Skip to main content

Live Bittensor · Subnet 21 · updated daily

SN21 leaderboard

Results for 19 November 2026 · 130 miners scored of 256 registered

Full record
Verify this day

Every score for 19 November 2026 is in a signed daily receipt anchored on chain. Re-check it yourself with the public verifier from the SN21 repo:

python scripts/verify_day.py --url https://hope-bittensor-api.onrender.com --day 2026-11-19

It checks the signature, the hash chain and the on-chain anchor, and recomputes every score. Add --miner <your hotkey> to check only yours.

Miners earning

20 / 142

paid today, out of everyone scored

Needed to earn

+0.024

edge of place 20, the last paid miner

Top miner

UID 193

+0.032 edge · 0.588 accuracy

Scores settled

9,555,311

predictions checked against real outcomes

69,224 episodes served in 108 daily baskets · latest basket 503 (18 Nov) · basket size over time →

Leaderboard

142 miners · ranked by edge over the field

Place UID Edge accuracy below Result
1 193 +0.032 0.588 Earning · Elite
2 173 +0.031 0.596 Earning · Elite
3 44 +0.031 0.587 Earning · Elite
4 154 +0.031 0.596 Earning · Elite
5 237 +0.030 0.595 Earning · Elite
6 28 +0.030 0.595 Earning · Elite
7 30 +0.030 0.593 Earning · Elite
8 104 +0.030 0.583 Earning · Elite
9 26 +0.029 0.593 Earning · Elite
10 42 +0.029 0.588 Earning · Elite
11 194 +0.028 0.592 Earning · Elite
12 118 +0.028 0.593 Earning · Elite
13 255 +0.028 0.592 Earning · Elite
14 74 +0.028 0.584 Earning · Elite
— 160 +0.028 0.591 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
15 130 +0.027 0.592 Earning · Elite
16 34 +0.027 0.582 Earning · Elite
17 37 +0.026 0.592 Earning · Elite
18 103 +0.026 0.587 Earning · Competitive
— 61 +0.025 0.589 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
19 243 +0.024 0.582 Earning · Competitive
— 112 +0.024 0.579 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
20 143 +0.024 0.586 Earning · Competitive
Earning cut · place 20 · edge +0.024
— 253 +0.023 0.587 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 145 +0.023 0.587 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 158 +0.023 0.576 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 126 +0.023 0.585 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 94 +0.023 0.586 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 39 +0.022 0.584 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 251 +0.022 0.574 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 51 +0.021 0.586 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 137 +0.021 0.583 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 60 +0.021 0.581 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 52 +0.021 0.581 Below the alpha hold
Why?
  • Below the alpha hold Held 0.3 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 75 +0.021 0.581 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 72 +0.020 0.580 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 244 +0.019 0.580 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 12 +0.019 0.574 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 133 +0.019 0.580 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 218 +0.018 0.579 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 95 +0.018 0.578 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 17 +0.018 0.575 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 9 +0.018 0.571 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 18 +0.017 0.580 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 221 +0.017 0.580 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 231 +0.017 0.571 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 92 +0.017 0.578 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 177 +0.016 0.578 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 235 +0.016 0.579 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 171 +0.016 0.577 Below the alpha hold
Why?
  • Below the alpha hold Held 2.7 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 45 +0.016 0.577 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 91 +0.016 0.578 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 124 +0.016 0.577 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 6 +0.016 0.568 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 122 +0.016 0.582 Tenure
Why?
  • Tenure Scored on 6 of the 7 days needed. Scores still count and the model keeps running — tenure accrues by showing up.
— 119 +0.016 0.567 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 11 +0.015 0.566 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 213 +0.015 0.582 Tenure
Why?
  • Tenure Scored on 5 of the 7 days needed. Scores still count and the model keeps running — tenure accrues by showing up.
— 162 +0.015 0.576 Below the alpha hold
Why?
  • Below the alpha hold Held 1.3 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 66 +0.015 0.576 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 13 +0.015 0.576 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 141 +0.015 0.576 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 29 +0.015 0.576 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 178 +0.015 0.576 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 242 +0.014 0.577 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 56 +0.013 0.574 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 40 +0.012 0.572 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 102 +0.012 0.562 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 176 +0.012 0.575 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 196 +0.012 0.573 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 188 +0.012 0.573 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 199 +0.011 0.571 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 222 +0.010 0.571 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 148 +0.010 0.570 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 224 +0.010 0.571 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 165 +0.010 0.570 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 163 +0.009 0.570 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 131 +0.008 0.566 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 170 +0.008 0.570 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 226 +0.007 0.569 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 2 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 145 · 5GHpSi...DBfs
— 0 +0.007 0.566 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 238 +0.006 0.562 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 211 +0.006 0.566 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 63 +0.006 0.565 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 249 +0.004 0.564 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 246 +0.002 0.561 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 50 +0.002 0.558 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 113 +0.002 0.560 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 46 +0.000 0.555 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 76 −0.000 0.559 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 2 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 249 · 5DVFGX...ffqU
— 179 −0.001 0.559 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 120 −0.002 0.558 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 234 −0.005 0.552 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 25 −0.006 0.549 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 99 −0.009 0.535 Tenure
Why?
  • Tenure Scored on 5 of the 7 days needed. Scores still count and the model keeps running — tenure accrues by showing up.
— 254 −0.011 0.544 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 38 −0.011 0.550 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 125 −0.012 0.545 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 233 −0.012 0.549 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 240 −0.023 0.531 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 140 −0.027 0.530 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 90 −0.027 0.530 Duplicate model
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 3 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 44 · 5Evf2A...voUX
  • Duplicate model This model is already earning under an earlier submission. Identical models pay once. The seat is held by the hotkey shown. earns as UID 140 · 5C5kbk...BZGi
  • Same model lineage Same behaviour lineage as an earlier submission; the earliest in the group earns. earns as UID 140 · 5C5kbk...BZGi
— 32 −0.027 0.530 Duplicate model
Why?
  • Duplicate model This model is already earning under an earlier submission. Identical models pay once. The seat is held by the hotkey shown. earns as UID 140 · 5C5kbk...BZGi
  • Same model lineage Same behaviour lineage as an earlier submission; the earliest in the group earns. earns as UID 140 · 5C5kbk...BZGi
— 230 −0.028 0.529 Duplicate model
Why?
  • Duplicate model This model is already earning under an earlier submission. Identical models pay once. The seat is held by the hotkey shown. earns as UID 140 · 5C5kbk...BZGi
  • Same model lineage Same behaviour lineage as an earlier submission; the earliest in the group earns. earns as UID 140 · 5C5kbk...BZGi
— 5 −0.028 0.529 Duplicate model
Why?
  • Duplicate model This model is already earning under an earlier submission. Identical models pay once. The seat is held by the hotkey shown. earns as UID 140 · 5C5kbk...BZGi
  • Same model lineage Same behaviour lineage as an earlier submission; the earliest in the group earns. earns as UID 140 · 5C5kbk...BZGi
— 115 −0.028 0.533 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 209 −0.028 0.526 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 239 −0.033 0.524 Duplicate model
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 3 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 44 · 5Evf2A...voUX
  • Duplicate model This model is already earning under an earlier submission. Identical models pay once. The seat is held by the hotkey shown. earns as UID 140 · 5C5kbk...BZGi
  • Same model lineage Same behaviour lineage as an earlier submission; the earliest in the group earns. earns as UID 140 · 5C5kbk...BZGi
— 189 −0.033 0.524 Duplicate model
Why?
  • Duplicate model This model is already earning under an earlier submission. Identical models pay once. The seat is held by the hotkey shown. earns as UID 140 · 5C5kbk...BZGi
  • Same model lineage Same behaviour lineage as an earlier submission; the earliest in the group earns. earns as UID 140 · 5C5kbk...BZGi
— 245 −0.043 0.517 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 58 −0.048 0.512 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 19 −0.053 0.507 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 23 −0.057 0.490 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 107 −0.063 0.499 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 252 −0.084 0.465 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 31 −0.136 0.417 Below the earning cut
Why?
  • Below the earning cut Ranked below the earning cut. The score stands and the model keeps running — the paid set is a fixed size, so placing outside it is not a penalty and carries nothing forward.
— 8 −0.215 0.328 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 202 −0.382 0.136 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 250 −0.406 0.151 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 5 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 19 · 5G3pK6...shLV
— 241 −0.408 0.149 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 5 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 19 · 5G3pK6...shLV
— 136 −0.410 0.147 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 5 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 19 · 5G3pK6...shLV
— 138 −0.442 0.115 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 5 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 19 · 5G3pK6...shLV
— 70 −0.445 0.112 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 8 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 118 · 5CtypK...3TaM
— 20 −0.445 0.112 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 8 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 118 · 5CtypK...3TaM
— 108 −0.445 0.112 Duplicate model
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 8 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 118 · 5CtypK...3TaM
  • Duplicate model This model is already earning under an earlier submission. Identical models pay once.
  • Same model lineage Same behaviour lineage as an earlier submission; the earliest in the group earns. earns as UID 70 · 5Ek7Wq...KXaZ
— 152 −0.445 0.112 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 8 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 118 · 5CtypK...3TaM
— 167 −0.453 0.103 Duplicate model
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 8 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 118 · 5CtypK...3TaM
  • Duplicate model This model is already earning under an earlier submission. Identical models pay once.
  • Same model lineage Same behaviour lineage as an earlier submission; the earliest in the group earns. earns as UID 70 · 5Ek7Wq...KXaZ
— 169 −0.461 0.095 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 8 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 118 · 5CtypK...3TaM
— 216 −0.467 0.089 One seat per coldkey
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 8 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 118 · 5CtypK...3TaM
— 97 −0.532 0.025 Below the alpha hold
Why?
  • Below the alpha hold Held 0.0 alpha against the 1,000 required today. Zero weight for the day; the standing and every score stand, and the seat pays again from the next run on which the hold is met.
— 14 −0.554 0.000 Below baseline
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 5 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 92 · 5GZ1yJ...1PrF
— 54 −0.554 0.000 Below baseline
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 5 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 92 · 5GZ1yJ...1PrF
— 204 −0.554 0.000 Below baseline
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 5 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 92 · 5GZ1yJ...1PrF
— 55 −0.554 0.000 Below baseline
Why?
  • One seat per coldkey One coldkey holds one earning seat. This owner runs 5 hotkeys on the subnet; the one shown holds the seat today. seat held by UID 92 · 5GZ1yJ...1PrF
— 22 0.595 Building evidence
Why?
  • Placement floor Evidence 23.7 of the 250 needed for placement. Every settled entry adds to it; the row gets a standing and a rank once the floor is reached, and earns nothing before that.
— 121 0.584 Building evidence
Why?
  • Placement floor Evidence 23.7 of the 250 needed for placement. Every settled entry adds to it; the row gets a standing and a rank once the floor is reached, and earns nothing before that.
— 217 0.577 Building evidence
Why?
  • Placement floor Evidence 58.4 of the 250 needed for placement. Every settled entry adds to it; the row gets a standing and a rank once the floor is reached, and earns nothing before that.
— 172 0.552 Building evidence
Why?
  • Placement floor Evidence 58.4 of the 250 needed for placement. Every settled entry adds to it; the row gets a standing and a rank once the floor is reached, and earns nothing before that.
— 4 0.550 Building evidence
Why?
  • Placement floor Evidence 233.3 of the 250 needed for placement. Every settled entry adds to it; the row gets a standing and a rank once the floor is reached, and earns nothing before that.
— 144 0.547 Building evidence
Why?
  • Placement floor Evidence 58.4 of the 250 needed for placement. Every settled entry adds to it; the row gets a standing and a rank once the floor is reached, and earns nothing before that.
— 232 0.510 Building evidence
Why?
  • Placement floor Evidence 58.4 of the 250 needed for placement. Every settled entry adds to it; the row gets a standing and a rank once the floor is reached, and earns nothing before that.
— 49 0.495 Building evidence
Why?
  • Placement floor Evidence 157.7 of the 250 needed for placement. Every settled entry adds to it; the row gets a standing and a rank once the floor is reached, and earns nothing before that.

Click a UID to open that miner's page. Edge is the average lead over the field on the same changes — it ranks and pays; accuracy is shown beneath it.

About this data

Daily stream — live. Each day's settled standings are published here as they mature, with outcomes measured at 7, 14 and 28 days. Earlier weekly epochs remain in the Epochs tab as the historical record. About the daily stream →

Prediction performance by change type. For every kind of account change: the best model's score, the current winner's beside it, and the score broken into direction, size, range and goal. Performance tab → Full page →

More on SN21: daily numbers, score distribution, history, method, API and performance

Pulse

Pulse · Operational snapshot

Headline numbers from the latest published day. SN21 publishes a new day of settled results every day; these tiles describe the most recent one. The leaders are in the leaderboard at the top of the page.

Pool size

130

miners

Trailing 4 epochs awaiting more closes

130/256 passed the participation gate

Baseline beat rate

100.0%

of pool

Trailing 4 epochs awaiting more closes

Share above the conditional-prior baseline

Elite tier

18

60.0% of emissions

Trailing 4 epochs awaiting more closes

EMA ≥ baseline + 1σ this epoch

Cohort context · this epoch vs all published

  • Pool size130→ 34th of 143 published (range 15–164)
  • Baseline beat rate100.0%→ highest beat rate yet (range 76.0%–100.0%)
  • Top score0.596→ 125th of 143 epochs with histograms

Tier band · Pool vs emissions

Where the qualifying pool sits — and where the rewards flow.

Live

Share of pool

measured

Share of emissions

policy constant

Elite

EMA ≥ baseline + 1σ

pool 13.8% · emit 60.0%

Competitive

beats baseline

pool 27.7% · emit 30.0%

Participating

passed gate

pool 26.9% · emit 10.0%

Tier rail · Emission shares + epoch context

4 metrics

Competitive

36

0.30 emissions share

Participating

35

0.10 emissions share

Multiplier

1.00×

Search

Formula

1.2.1

@ a1b20d8

Find yourself on chain

Opens taostats

Enter your numeric UID (the slot you registered into on netuid 21). This opens SN21's miner table on taostats with your UID copied: paste it into the table's search box to see your hotkey, stake, incentive and emission.

Don't know your UID? Paste your hotkey (SS58 address starting with 5). It is copied for you: paste it into the search box on SN21's miner table to find your UID.

Per IA D-05, no per-miner identifier ever leaves this server. The browser navigates straight to taostats.

Prediction performance

For every kind of account change since the rich-data baseline (2026-08-20): the best model, the current winner beside it, and each score broken into its parts — 9,555,311 settled scores from 238 models.

Full page →

Where the number comes from

91 days, 2026-08-20 to 2026-11-18
  1. 1. Captured380,502raw changes per day, before any rule
  2. 2. Measurable episodes111,013since 20 Aug, after the published rules and 72-hour clustering
  3. 3. Served to models47,936of 55,569 eligible; 10,778 held out for validation, 244 on our own accounts
  4. 4. Settled & scored33,107this page — served changes old enough to have settled
  5. 5. Predictions scored9,555,311every model predicts every change

Each stage is deliberate. The page counts the fourth: a change reaches it about 15 days after it happened, once its first outcome window has closed and settled.

14 groups covering 104 change types. Open a row to see the types inside it.

Budget change — when a daily budget was raised or lowered 7,127 Reliable UID 37 6 types

Across 7,127 such changes (816,885 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

down 1,557 Reliable UID 122
down large 1,707 Reliable UID 213
flat 414 Roughly right UID 169
unknown 19 Reliable UID 221
up 1,432 Reliable UID 122
up large 1,998 Reliable UID 22
Bid strategy switch — when the bid strategy was switched from one type to another 214 Reliable — 8 types

Across 214 such changes (24,851 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

maximize conversion value → maximize conversions 2 Too early —
maximize conversions → maximize conversion value 5 Too early UID 48
maximize conversions → target spend 16 Reliable —
target spend → maximize conversion value 1 Too early —
target spend → maximize conversions 9 Roughly right UID 5
switched to Maximize Conversion Value 54 Roughly right —
switched to Maximize Conversions 67 Reliable UID 204
unknown → target spend 60 Reliable UID 152
Bid target change — when a bid-strategy target value changed (e.g. Target CPA $60 → $50) 975 Reliable UID 169 13 types

Across 975 such changes (117,060 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

ad-group bid change 14 Reliable UID 198
maximize conversion value · down 94 Roughly right UID 167
maximize conversion value · down large 8 Reliable UID 115
maximize conversion value · flat 118 Reliable UID 242
switched to Maximize Conversion Value 2 Too early —
maximize conversion value · up 170 Reliable UID 216
maximize conversion value · up large 29 Reliable UID 226
maximize conversions · down 138 Roughly right —
maximize conversions · down large 41 Roughly right —
maximize conversions · flat 30 Roughly right UID 240
switched to Maximize Conversions 13 Roughly right UID 196
maximize conversions · up 240 Reliable —
maximize conversions · up large 78 Roughly right UID 122
Negative keyword changes — when negative keywords were added 6,682 Reliable UID 22 24 types

Across 6,682 such changes (828,223 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

negative keyword add 327 Reliable —
broad · 1 100 Reliable UID 131
broad · 11+ 22 Roughly right UID 204
broad · 2-10 92 Reliable UID 31
exact · 1 867 Reliable UID 108
exact · 11+ 579 Reliable UID 169
exact · 2-10 1,366 Reliable UID 26
mixed · 11+ 257 Reliable —
mixed · 2-10 427 Reliable UID 217
phrase · 1 335 Reliable —
phrase · 11+ 159 Reliable UID 55
phrase · 2-10 439 Reliable —
unknown · 1 9 Reliable —
unknown · 11+ 10 Reliable UID 115
unknown · 2-10 10 Roughly right —
negative keyword remove 150 Reliable —
exact · 1 2 Too early —
exact · 11+ 7 Roughly right UID 120
exact · 2-10 132 Roughly right UID 31
mixed · 11+ 993 Roughly right UID 122
mixed · 2-10 286 Roughly right —
phrase · 1 2 Too early —
phrase · 2-10 20 Reliable UID 216
shared negative keyword remove 91 Roughly right UID 213
Keyword changes — when this changed 1,364 Reliable — 2 types

Across 1,364 such changes (155,491 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

keyword add 795 Reliable UID 172
keyword remove 569 Reliable —
Other criterion changes (kind not recorded) — when this changed 2,968 Reliable — 6 types

Across 2,968 such changes (333,278 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

bid on a keyword / audience / placement 686 Reliable —
targeting criteria change 1,172 Reliable —
criterion enable 118 Reliable UID 55
criterion pause 942 Reliable UID 216
criterion remove 42 Reliable UID 234
criterion url change 8 Reliable —
Location, audience and demographic targeting — when this changed 219 Reliable — 3 types

Across 219 such changes (24,269 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

audience change 126 Reliable UID 14
demographic change 38 Reliable —
geo change 55 Reliable —
Device, schedule and placement targeting — when this changed 544 Reliable — 2 types

Across 544 such changes (66,719 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Roughly right Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

device targeting change 2 Too early —
schedule change 542 Reliable —
Campaign pause / enable — when this changed 1,634 Reliable — 1 type

Across 1,634 such changes (202,244 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Roughly right Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Ad group pause / enable — when this changed 223 Reliable UID 169 2 types

Across 223 such changes (25,946 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

adgroup enable 31 Roughly right UID 14
adgroup pause 192 Reliable UID 170
Other ad group changes — when this changed 301 Reliable — 3 types

Across 301 such changes (34,195 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

adgroup bid change 167 Reliable UID 31
adgroup change 132 Reliable —
adgroup remove 2 Too early —
Ad changes — when this changed 558 Reliable UID 172 5 types

Across 558 such changes (69,002 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

ad create 144 Roughly right —
ad edit 274 Reliable UID 240
ad enable 15 Reliable UID 30
ad pause 117 Reliable UID 55
ad remove 8 Reliable UID 189
Asset changes — when this changed 1,550 Reliable — 1 type

Across 1,550 such changes (179,233 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Combined changes — when several kinds of change happened together 7,829 Reliable — 28 types

Across 7,829 such changes (906,934 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

adgroup bid change 50 Roughly right —
adgroup change 15 Reliable UID 12
adgroup enable 19 Reliable —
adgroup pause 521 Reliable UID 108
ad-group bid change 1 Too early —
ad create 930 Reliable UID 141
ad edit 93 Reliable UID 31
ad enable 91 Reliable UID 17
ad pause 16 Reliable UID 115
ad remove 2 Too early —
asset change 82 Reliable UID 169
audience change 29 Reliable UID 34
bid strategy change 58 Reliable UID 169
budget change 2,063 Reliable UID 213
campaign pause 253 Reliable —
bid on a keyword / audience / placement 54 Reliable —
targeting criteria change 191 Reliable UID 4
criterion enable 60 Reliable UID 55
criterion pause 20 Reliable UID 193
criterion remove 12 Reliable —
demographic change 1 Too early —
geo change 2 Too early —
keyword add 1,619 Reliable —
keyword remove 203 Reliable —
negative keyword add 829 Reliable —
negative keyword remove 2 Too early —
shared negative keyword add 308 Roughly right UID 122
target value change 305 Reliable UID 144

14 groups covering 102 change types. Open a row to see the types inside it.

Budget change — when a daily budget was raised or lowered 6,692 Reliable UID 37 6 types

Across 6,692 such changes (772,545 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

down 1,459 Reliable UID 37
down large 1,618 Reliable UID 37
flat 377 Roughly right UID 251
unknown 19 Reliable —
up 1,344 Roughly right UID 202
up large 1,875 Reliable UID 37
Bid strategy switch — when the bid strategy was switched from one type to another 189 Reliable — 8 types

Across 189 such changes (22,068 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

maximize conversion value → maximize conversions 2 Too early —
maximize conversions → maximize conversion value 5 Not reliable UID 188
maximize conversions → target spend 15 Reliable UID 75
target spend → maximize conversion value 1 Too early —
target spend → maximize conversions 7 Roughly right UID 5
switched to Maximize Conversion Value 43 Roughly right —
switched to Maximize Conversions 61 Reliable UID 55
unknown → target spend 55 Reliable UID 42
Bid target change — when a bid-strategy target value changed (e.g. Target CPA $60 → $50) 958 Reliable UID 234 13 types

Across 958 such changes (117,753 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

ad-group bid change 14 Roughly right UID 138
maximize conversion value · down 86 Roughly right UID 160
maximize conversion value · down large 5 Not reliable UID 138
maximize conversion value · flat 138 Reliable UID 221
switched to Maximize Conversion Value 2 Too early —
maximize conversion value · up 173 Reliable UID 242
maximize conversion value · up large 23 Reliable UID 165
maximize conversions · down 134 Roughly right —
maximize conversions · down large 41 Roughly right UID 28
maximize conversions · flat 29 Reliable —
switched to Maximize Conversions 13 Roughly right UID 196
maximize conversions · up 231 Reliable —
maximize conversions · up large 69 Roughly right —
Negative keyword changes — when negative keywords were added 5,398 Reliable — 23 types

Across 5,398 such changes (661,217 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

negative keyword add 330 Roughly right UID 245
broad · 1 93 Roughly right —
broad · 11+ 20 Roughly right UID 55
broad · 2-10 86 Reliable UID 31
exact · 1 772 Reliable UID 37
exact · 11+ 513 Reliable UID 169
exact · 2-10 1,207 Reliable UID 26
mixed · 11+ 232 Roughly right —
mixed · 2-10 382 Roughly right —
phrase · 1 310 Reliable —
phrase · 11+ 143 Roughly right —
phrase · 2-10 392 Reliable UID 237
unknown · 1 8 Reliable UID 66
unknown · 11+ 10 Reliable UID 115
unknown · 2-10 10 Roughly right UID 91
negative keyword remove 139 Roughly right —
exact · 1 1 Too early —
exact · 2-10 157 Roughly right UID 61
mixed · 11+ 408 Roughly right UID 104
mixed · 2-10 137 Roughly right UID 12
phrase · 1 2 Too early —
phrase · 2-10 20 Reliable UID 216
shared negative keyword remove 26 Reliable UID 119
Keyword changes — when this changed 1,337 Reliable — 2 types

Across 1,337 such changes (155,646 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

keyword add 759 Reliable —
keyword remove 578 Reliable —
Other criterion changes (kind not recorded) — when this changed 2,846 Reliable UID 216 6 types

Across 2,846 such changes (329,460 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

bid on a keyword / audience / placement 657 Reliable UID 253
targeting criteria change 1,139 Reliable UID 152
criterion enable 111 Roughly right —
criterion pause 892 Reliable UID 55
criterion remove 40 Reliable UID 234
criterion url change 7 Reliable —
Location, audience and demographic targeting — when this changed 208 Reliable — 3 types

Across 208 such changes (23,191 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

audience change 124 Reliable UID 74
demographic change 35 Reliable —
geo change 49 Reliable UID 31
Device, schedule and placement targeting — when this changed 509 Reliable — 2 types

Across 509 such changes (63,459 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Roughly right Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

device targeting change 2 Too early —
schedule change 507 Reliable —
Campaign pause / enable — when this changed 1,640 Reliable — 1 type

Across 1,640 such changes (201,609 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Ad group pause / enable — when this changed 213 Reliable UID 234 2 types

Across 213 such changes (25,410 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

adgroup enable 30 Roughly right UID 42
adgroup pause 183 Reliable UID 23
Other ad group changes — when this changed 279 Reliable — 3 types

Across 279 such changes (32,810 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

adgroup bid change 157 Reliable UID 23
adgroup change 120 Reliable —
adgroup remove 2 Too early —
Ad changes — when this changed 431 Roughly right UID 55 5 types

Across 431 such changes (51,794 model predictions), a prediction here is roughly right. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

ad create 138 Roughly right UID 158
ad edit 161 Reliable —
ad enable 12 Roughly right —
ad pause 110 Reliable UID 55
ad remove 10 Roughly right UID 75
Asset changes — when this changed 1,406 Reliable — 1 type

Across 1,406 such changes (171,796 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Combined changes — when several kinds of change happened together 7,429 Reliable UID 37 27 types

Across 7,429 such changes (871,276 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

adgroup bid change 46 Reliable UID 238
adgroup change 14 Reliable UID 253
adgroup enable 19 Roughly right —
adgroup pause 505 Reliable —
ad-group bid change 1 Too early —
ad create 846 Roughly right UID 37
ad edit 89 Roughly right UID 167
ad enable 90 Roughly right UID 104
ad pause 14 Roughly right —
ad remove 1 Too early —
asset change 75 Reliable UID 104
audience change 28 Reliable UID 91
bid strategy change 49 Reliable —
budget change 1,934 Reliable UID 37
campaign pause 256 Reliable UID 37
bid on a keyword / audience / placement 54 Reliable UID 94
targeting criteria change 181 Reliable UID 102
criterion enable 55 Reliable UID 55
criterion pause 19 Roughly right UID 95
criterion remove 12 Reliable UID 107
demographic change 1 Too early —
keyword add 1,536 Reliable —
keyword remove 209 Reliable —
negative keyword add 752 Reliable —
negative keyword remove 2 Too early —
shared negative keyword add 366 Roughly right UID 44
target value change 275 Reliable —

14 groups covering 93 change types. Open a row to see the types inside it.

Budget change — when a daily budget was raised or lowered 5,182 Roughly right UID 237 6 types

Across 5,182 such changes (548,194 model predictions), a prediction here is roughly right. What that's made of:

  • Roughly right Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

down 1,163 Roughly right UID 130
down large 1,229 Roughly right UID 130
flat 327 Roughly right —
unknown 11 Reliable UID 176
up 1,009 Roughly right UID 237
up large 1,443 Roughly right UID 237
Bid strategy switch — when the bid strategy was switched from one type to another 144 Reliable UID 170 8 types

Across 144 such changes (15,323 model predictions), a prediction here is reliable. What that's made of:

  • Roughly right Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

maximize conversion value → maximize conversions 2 Too early —
maximize conversions → maximize conversion value 4 Too early —
maximize conversions → target spend 15 Reliable UID 92
target spend → maximize conversion value 1 Too early —
target spend → maximize conversions 6 Roughly right —
switched to Maximize Conversion Value 38 Roughly right —
switched to Maximize Conversions 42 Reliable UID 55
unknown → target spend 36 Reliable UID 145
Bid target change — when a bid-strategy target value changed (e.g. Target CPA $60 → $50) 582 Roughly right UID 118 13 types

Across 582 such changes (61,478 model predictions), a prediction here is roughly right. What that's made of:

  • Roughly right Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

ad-group bid change 14 Roughly right UID 254
maximize conversion value · down 63 Roughly right —
maximize conversion value · down large 4 Too early —
maximize conversion value · flat 15 Roughly right UID 165
switched to Maximize Conversion Value 2 Too early —
maximize conversion value · up 70 Roughly right UID 91
maximize conversion value · up large 23 Roughly right UID 204
maximize conversions · down 104 Roughly right UID 51
maximize conversions · down large 39 Roughly right UID 222
maximize conversions · flat 21 Reliable UID 165
switched to Maximize Conversions 13 Roughly right UID 196
maximize conversions · up 166 Reliable —
maximize conversions · up large 48 Roughly right —
Negative keyword changes — when negative keywords were added 3,440 Roughly right UID 253 16 types

Across 3,440 such changes (368,822 model predictions), a prediction here is roughly right. What that's made of:

  • Roughly right Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

negative keyword add 332 Roughly right UID 218
broad · 1 71 Roughly right —
broad · 11+ 19 Roughly right —
broad · 2-10 76 Roughly right —
exact · 1 544 Roughly right —
exact · 11+ 392 Roughly right UID 107
exact · 2-10 864 Roughly right UID 28
mixed · 11+ 186 Roughly right —
mixed · 2-10 256 Roughly right UID 173
phrase · 1 189 Roughly right UID 130
phrase · 11+ 97 Reliable UID 107
phrase · 2-10 286 Roughly right UID 170
unknown · 1 9 Reliable —
unknown · 11+ 9 Reliable UID 126
unknown · 2-10 8 Roughly right UID 58
negative keyword remove 102 Roughly right UID 51
Keyword changes — when this changed 1,044 Reliable — 2 types

Across 1,044 such changes (111,818 model predictions), a prediction here is reliable. What that's made of:

  • Roughly right Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

keyword add 586 Roughly right —
keyword remove 458 Reliable —
Other criterion changes (kind not recorded) — when this changed 2,264 Reliable UID 216 6 types

Across 2,264 such changes (242,396 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

bid on a keyword / audience / placement 546 Reliable UID 216
targeting criteria change 877 Reliable UID 216
criterion enable 102 Roughly right —
criterion pause 706 Roughly right UID 239
criterion remove 28 Reliable UID 141
criterion url change 5 Too early —
Location, audience and demographic targeting — when this changed 191 Reliable UID 170 3 types

Across 191 such changes (20,247 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

audience change 116 Reliable UID 221
demographic change 32 Reliable —
geo change 43 Roughly right UID 173
Device, schedule and placement targeting — when this changed 351 Reliable UID 216 2 types

Across 351 such changes (38,882 model predictions), a prediction here is reliable. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Roughly right Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

device targeting change 2 Too early —
schedule change 349 Reliable UID 216
Campaign pause / enable — when this changed 1,071 Roughly right UID 130 1 type

Across 1,071 such changes (113,840 model predictions), a prediction here is roughly right. What that's made of:

  • Roughly right Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Ad group pause / enable — when this changed 158 Roughly right UID 170 2 types

Across 158 such changes (17,085 model predictions), a prediction here is roughly right. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

adgroup enable 25 Roughly right —
adgroup pause 133 Reliable UID 170
Other ad group changes — when this changed 206 Reliable — 3 types

Across 206 such changes (21,869 model predictions), a prediction here is reliable. What that's made of:

  • Roughly right Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

adgroup bid change 113 Reliable UID 130
adgroup change 91 Reliable —
adgroup remove 2 Too early —
Ad changes — when this changed 300 Roughly right UID 55 5 types

Across 300 such changes (32,409 model predictions), a prediction here is roughly right. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

ad create 64 Roughly right UID 61
ad edit 127 Reliable UID 154
ad enable 9 Roughly right —
ad pause 91 Roughly right UID 55
ad remove 9 Roughly right —
Asset changes — when this changed 870 Roughly right — 1 type

Across 870 such changes (92,339 model predictions), a prediction here is roughly right. What that's made of:

  • Reliable Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Combined changes — when several kinds of change happened together 5,524 Roughly right UID 130 25 types

Across 5,524 such changes (586,245 model predictions), a prediction here is roughly right. What that's made of:

  • Roughly right Direction — called the change up or down correctly on the goal metric
  • Reliable Size — how close the predicted size of the effect was
  • Not reliable Range — how often the real result fell inside the predicted range
  • Not reliable Goal — how close on the goal number itself (CPA or ROAS)

Today's leaderboard #1 (UID 193) on this change type: Reliable.

Broken down by kind:

adgroup bid change 34 Roughly right UID 29
adgroup change 10 Reliable UID 103
adgroup enable 17 Roughly right —
adgroup pause 412 Roughly right UID 61
ad-group bid change 1 Too early —
ad create 648 Roughly right UID 51
ad edit 62 Roughly right UID 224
ad enable 25 Roughly right UID 126
ad pause 11 Roughly right UID 91
asset change 52 Roughly right UID 18
audience change 20 Roughly right UID 95
bid strategy change 42 Reliable —
budget change 1,510 Roughly right UID 216
campaign pause 209 Roughly right —
bid on a keyword / audience / placement 47 Reliable UID 169
targeting criteria change 141 Reliable UID 216
criterion enable 49 Reliable UID 194
criterion pause 17 Roughly right UID 72
criterion remove 9 Reliable UID 63
demographic change 1 Too early —
keyword add 1,248 Reliable —
keyword remove 162 Reliable —
negative keyword add 573 Roughly right —
negative keyword remove 1 Too early —
target value change 223 Roughly right UID 74

A row reads Too early until at least 500 of its changes have settled. Open any row for the score breakdown and the best model.

Today's slice — accuracy by change type on BD-2026-11-19

Accuracy by change type

Change type Predictions Field average Winning model
Budget change 11706 0.561 0.573
Combined changes 10719 0.537 0.572
Negative keyword add 7881 0.547 0.576
Asset change 2032 0.572 0.633
Criterion pause 773 0.579 0.567
Targeting criteria change 1386 0.593 0.668
Bid target change 3720 0.495 0.469
Keyword removal 465 0.494 0.514
Campaign pause 3415 0.460 0.569
Schedule change 154 0.720 0.648
Keyword add 616 0.568 0.587
Criterion bid change 625 0.528 0.553
Negative keyword remove 616 0.507 0.508
Ad edit 157 0.419 0.388
Adgroup bid change 157 0.703 0.784
Bid strategy switch 154 0.573 0.515
Geo change 924 0.619 0.655
Adgroup pause 154 0.463 0.364
Ad enable 616 0.651 0.662
Criterion enable 154 0.603 0.715
Demographic change 154 0.554 0.512
Change type Predictions Field average Winning model
Budget change 43355 0.584 0.628
Combined changes 29000 0.572 0.605
Negative keyword add 16965 0.523 0.550
Asset change 20155 0.547 0.575
Criterion pause 11020 0.523 0.550
Targeting criteria change 7540 0.530 0.574
Bid target change 1160 0.573 0.635
Keyword removal 1595 0.529 0.583
Campaign pause 725 0.548 0.498
Keyword add 3480 0.527 0.541
Criterion bid change 1740 0.576 0.579
Ad edit 1015 0.503 0.548
Adgroup bid change 435 0.663 0.670
Bid strategy switch 870 0.538 0.604
Shared negative keyword add 1015 0.484 0.415
Ad pause 145 0.553 0.530
New ad 145 0.618 0.648
Adgroup enable 145 0.615 0.792
Audience change 145 0.583 0.566
Criterion remove 145 0.461 0.427
Change type Predictions Field average Winning model
Budget change 18000 0.561 0.600
Combined changes 21750 0.561 0.587
Negative keyword add 18500 0.551 0.573
Asset change 5875 0.541 0.543
Criterion pause 1500 0.520 0.521
Targeting criteria change 3375 0.570 0.594
Bid target change 2625 0.582 0.593
Keyword removal 4000 0.626 0.649
Campaign pause 1625 0.545 0.532
Schedule change 5250 0.635 0.648
Keyword add 1125 0.514 0.565
Criterion bid change 2375 0.612 0.630
Negative keyword remove 1750 0.490 0.543
Ad edit 1000 0.581 0.611
Adgroup bid change 1125 0.581 0.607
Bid strategy switch 375 0.504 0.453
Adgroup pause 625 0.635 0.653
Ad pause 500 0.502 0.542
Criterion enable 375 0.528 0.558
New ad 250 0.548 0.618

Scores are on the same 0-to-1 scale as the leaderboard standing. "Winning model" is the current leaderboard #1's average on that change type; "—" means it has no settled predictions there yet.

Score distribution

Distribution · Score histogram + tier breakdown

Per IA D-05 the public surface publishes the shape of the score distribution — not per-miner identifiers. Bins are k-anonymity-collapsed to a 5-miner floor; the chain remains the source of truth for raw per-UID data.

Results · BD-2026-11-19

Search ×1.00

Plain-language summary

20 of 142 miners are earning today.

Earning today

20/142

14% of the field

Not earning

122

most: holding less alpha than the published hold

Top score

0.596

leader of the pool

Score range

0.052

Mixed range

Why the other 122 are not earning

  • 48 holding less alpha than the published hold
  • 40 ranked below the earning cut
  • 15 a second hotkey on one owner
  • 8 running a model that already earns
  • 3 still building a track record
  • 8 below the paying positions

Each row below names the rule that applied to it, and who holds the seat where that matters. Scores are unaffected — they are facts either way.

Verifiable against chain

Signed and chain-anchored

2026-11-19

Every published day is attested and hash-chained to the day before, and the feed's rolling root is committed on chain — so one anchor covers the whole history, not a single block span.

Reproduce locally

python scripts/verify_day.py --url https://hope-bittensor-api.onrender.com --day 2026-11-19

Recomputes every score from the day's signed receipt. Prints "ok": true when they reproduce, and names the entry that disagrees when they do not.

Scoring · v1.2.1 · commit a1b20d870b

The per-miner leaderboard is at the top of this page.

Score range · 130 miners scored

Aggregate summary

Scores landed between 0.544 and 0.596 — top score 0.596

Range collapses to a 8.5-point window

mean 0.519
p25 0.544
median 0.572
p75 0.581
top 0.596

0.515

0.600

Axis is auto-zoomed to the data range so percentile positions are legible even when the underlying histogram bins are coarse. Score cuts shown above are exact; positioning is proportional.

Histogram suppressed · bins too coarse to add signal

The upstream aggregator emitted bin edges that span much wider than the actual data range (e.g. [0, 1] with all scores clustered around 0.572). A bar chart would compress into one column; the score range strip above shows the same shape more legibly. Once the aggregator emits ≥10 bins or quantile-based edges, the full histogram returns automatically.

Position bands · Find where you sit

Score cuts at every quartile — match your score to a band.

5 bands

Band Score cut Tier hint What it means
Top score
0.596 Elite · 60% Highest score achieved by any miner in this epoch’s qualifying pool.
Top 25%
0.581 Elite · 60% Scoring at or above this puts you in the top quartile of the pool.
Mean score
0.519 — Average across the qualifying pool — useful proxy for the conditional-prior baseline.
Median
0.572 Competitive · 30% Half the pool scored at or above this, half below.
Bottom 25%
0.544 Participating · 10% Scoring at or below this puts you in the bottom quartile of the pool.

Tier hints are heuristic from this epoch's percentile cuts. Actual tier membership depends on EMA across multiple epochs vs the conditional-prior baseline + 1σ. See Methodology.

Score → band calculator

Heuristic

Enter a value between 0.000 and 1.000. Defaults shown match this epoch's distribution.

p25 0.544 · p50 0.572 · p75 0.581 · max 0.596

Band placement

Enter a score above

—

—

Headline numbers

Pool size
130
Baseline beat rate
100.0%
Epoch type
Search multiplier 1×
Registered UIDs
256

130/256 passed the gate.

Tier distribution

Tier Count Share of pool Share of emissions
Elite 18 13.8% 60.0%
Competitive 36 27.7% 30.0%
Participating 35 26.9% 10.0%

Elite-tier floor: met this epoch.

Accuracy by change type moved to the Performance tab — the cumulative picture with best model, current winner and score parts.

What does the score mean?

The score is an accuracy score from 0 to 1, where 1 is a perfect prediction. A score of 0.6 means the model is about 60% of the way to perfect. It is not the percentage of predictions that were correct.

It blends four things:

  • Closeness (half the score). The model gives a low, middle and high estimate for each change. This measures how close that range sat to what actually happened.
  • Honest confidence (a fifth). When the model claims to be sure, the real result should land inside its range about as often as it claims. Sounding confident and being wrong costs score.
  • Direction (15%). Did it call the move up or down correctly.
  • The account's own goal (15%). A lead business is judged on leads, a shop on sales value — the model has to get the metric that account actually cares about.

The leaderboard number is not a single day: it is the average of a model's recent predictions, where recent work counts more than old (a score twelve days old counts half). Predictions are locked before any outcome exists, and a day's first score settles about 15 days after its basket — 7 days of outcomes plus a settling week so late conversions count.

Mining? Look up your own scores by hotkey — per-day entries, win/lose by change type, and the signed receipt that lets you reproduce every number.

On-chain record

Block range
0–0
Scoring formula version
1.2.1 @ a1b20d8
Validator output snapshot
2026-11-19T09:42:30.113Z
Chain fetch
2026-11-19T09:42:30.065Z
Horizon set
7d, 14d, 28d

Published epochs

Epochs · Frozen records, newest first

Every closed epoch publishes as a permanent record at /predictions/epochs/[release_key]. Published rows are immutable per IA D-13; corrections appear as separate rows, not edits.

106 epoch slots · newest first · 24 have corrections

  1. BD-2026-11-19 Search · campaign-level pool 130 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ a1b20d8
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  2. BD-2026-11-18 Search · campaign-level pool 133 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 62815f8
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  3. BD-2026-11-17 Search · campaign-level pool 131 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 65ba8a8
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  4. BD-2026-11-16 Search · campaign-level pool 133 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 65ba8a8
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  5. BD-2026-11-15 Search · campaign-level pool 159 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 65ba8a8
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  6. BD-2026-11-14 Search · campaign-level pool 164 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 65ba8a8
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  7. BD-2026-11-13 Search · campaign-level pool 164 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ ef6f0ab
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  8. BD-2026-11-12 Search · campaign-level pool 163 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 0c2d9a0
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  9. BD-2026-11-11 Search · campaign-level pool 162 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ cf16a53
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  10. BD-2026-11-10 Search · campaign-level pool 162 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 3f8ef7a
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  11. BD-2026-11-09 Search · campaign-level pool 159 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 3f8ef7a
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  12. BD-2026-11-08 Search · campaign-level pool 157 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ bb3b1ab
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  13. BD-2026-11-07 Search · campaign-level pool 153 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ bb3b1ab
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  14. BD-2026-11-06 Search · campaign-level pool 149 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ bb3b1ab
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  15. BD-2026-11-05-COR-1 Search · campaign-level Corrected · 1 prior pool 147 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 7e5e0aa
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  16. BD-2026-11-04-COR-1 Search · campaign-level Corrected · 1 prior pool 145 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 1d4f3fa
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  17. BD-2026-11-03 Search · campaign-level pool 145 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 34dbdfe
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  18. BD-2026-11-02 Search · campaign-level pool 143 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 34dbdfe
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  19. BD-2026-11-01 Search · campaign-level pool 140 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 34dbdfe
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  20. BD-2026-10-31 Search · campaign-level pool 140 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ af0fb44
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  21. BD-2026-10-30 Search · campaign-level pool 136 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b0cbd6c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  22. BD-2026-10-29 Search · campaign-level pool 132 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b0cbd6c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  23. BD-2026-10-28 Search · campaign-level pool 136 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b0cbd6c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  24. BD-2026-10-27 Search · campaign-level pool 140 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b0cbd6c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  25. BD-2026-10-26 Search · campaign-level pool 137 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b0cbd6c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  26. BD-2026-10-25 Search · campaign-level pool 126 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b0cbd6c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  27. BD-2026-10-24 Search · campaign-level pool 124 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ f106cbf
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  28. BD-2026-10-23 Search · campaign-level pool 120 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ e85eacb
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  29. BD-2026-10-22 Search · campaign-level pool 117 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ e2d4c12
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  30. BD-2026-10-21 Search · campaign-level pool 118 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ e2d4c12
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  31. BD-2026-10-20 Search · campaign-level pool 127 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ e2d4c12
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  32. BD-2026-10-19 Search · campaign-level pool 126 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ e2d4c12
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  33. BD-2026-10-18 Search · campaign-level pool 117 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 808bc22
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  34. BD-2026-10-17 Search · campaign-level pool 117 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 808bc22
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  35. BD-2026-10-16 Search · campaign-level pool 117 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 8aec3c8
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  36. BD-2026-10-15 Search · campaign-level pool 121 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 8aec3c8
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  37. BD-2026-10-14 Search · campaign-level pool 123 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 128f7d9
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  38. BD-2026-10-13 Search · campaign-level pool 124 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ d0fad2d
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  39. BD-2026-10-12 Search · campaign-level pool 124 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 662fad7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  40. BD-2026-10-11 Search · campaign-level pool 124 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 662fad7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  41. BD-2026-10-10 Search · campaign-level pool 122 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 662fad7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  42. BD-2026-10-09 Search · campaign-level pool 122 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 662fad7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  43. BD-2026-10-08 Search · campaign-level pool 118 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b0bc708
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  44. BD-2026-10-07 Search · campaign-level pool 118 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b0bc708
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  45. BD-2026-10-06 Search · campaign-level pool 118 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b0bc708
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  46. BD-2026-10-05 Search · campaign-level pool 115 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  47. BD-2026-10-04 Search · campaign-level pool 128 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  48. BD-2026-10-03 Search · campaign-level pool 128 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  49. BD-2026-10-02 Search · campaign-level pool 127 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  50. BD-2026-10-01 Search · campaign-level pool 123 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  51. BD-2026-09-30 Search · campaign-level pool 125 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  52. BD-2026-09-29 Search · campaign-level pool 125 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  53. BD-2026-09-28 Search · campaign-level pool 125 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  54. BD-2026-09-27 Search · campaign-level pool 127 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  55. BD-2026-09-26 Search · campaign-level pool 128 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  56. BD-2026-09-25 Search · campaign-level pool 130 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  57. BD-2026-09-24 Search · campaign-level pool 130 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  58. BD-2026-09-23 Search · campaign-level pool 130 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  59. BD-2026-09-22 Search · campaign-level pool 130 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6b24e0c
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  60. BD-2026-09-21-COR-1 Search · campaign-level Corrected · 1 prior pool 128 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 85ed3c8
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  61. BD-2026-09-20-COR-1 Search · campaign-level Corrected · 1 prior pool 129 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 46a35dd
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  62. BD-2026-09-19 Search · campaign-level pool 130 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 725fdf7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  63. BD-2026-09-18 Search · campaign-level pool 131 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 098210f
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  64. BD-2026-09-17 Search · campaign-level pool 132 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 098210f
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  65. BD-2026-09-16 Search · campaign-level pool 132 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 2375996
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  66. BD-2026-09-15 Search · campaign-level pool 132 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 8b68ec4
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  67. BD-2026-09-14-COR-1 Search · campaign-level Corrected · 1 prior pool 132 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 884ae5e
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  68. BD-2026-09-13-COR-2 Search · campaign-level Corrected · 2 prior pool 128 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 73b7de0
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  69. BD-2026-09-12-COR-2 Search · campaign-level Corrected · 2 prior pool 128 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 73b7de0
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  70. BD-2026-09-11 Search · campaign-level pool 130 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 57c792d
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  71. BD-2026-09-10 Search · campaign-level pool 129 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 63faef0
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  72. BD-2026-09-09 Search · campaign-level pool 127 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 2c4b5b7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  73. BD-2026-09-08-COR-1 Search · campaign-level Corrected · 1 prior pool 125 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 2c4b5b7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  74. BD-2026-09-07-COR-1 Search · campaign-level Corrected · 1 prior pool 122 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 0c0b23a
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  75. BD-2026-09-06-COR-1 Search · campaign-level Corrected · 1 prior pool 122 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 861fdfc
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  76. BD-2026-09-05-COR-2 Search · campaign-level Corrected · 2 prior pool 122 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b845527
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  77. BD-2026-09-04 Search · campaign-level pool 121 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 75618fb
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  78. BD-2026-09-03 Search · campaign-level pool 118 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 799f4b3
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  79. BD-2026-09-02-COR-2 Search · campaign-level Corrected · 2 prior pool 118 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 799f4b3
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  80. BD-2026-09-01 Search · campaign-level pool 118 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6450130
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  81. BD-2026-08-31-COR-2 Search · campaign-level Corrected · 2 prior pool 119 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 6450130
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  82. BD-2026-08-30-COR-1 Search · campaign-level Corrected · 1 prior pool 122 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ f678ad5
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  83. BD-2026-08-29-COR-5 Search · campaign-level Corrected · 5 prior pool 123 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 9e0da63
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  84. BD-2026-08-28-COR-1 Search · campaign-level Corrected · 1 prior pool 124 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 28d9e84
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  85. BD-2026-08-27 Search · campaign-level pool 124 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 121aba4
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  86. BD-2026-08-26 Search · campaign-level pool 114 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ a83e1d6
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  87. BD-2026-08-25 Search · campaign-level pool 118 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 4fb93af
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  88. BD-2026-08-24 Search · campaign-level pool 120 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 8463fde
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  89. BD-2026-08-23 Search · campaign-level pool 120 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 84e001d
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  90. BD-2026-08-22-COR-1 Search · campaign-level Corrected · 1 prior pool 120 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 01d18d9
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  91. BD-2026-08-21 Search · campaign-level pool 117 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ 5e39f21
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  92. BD-2026-08-20 Search · campaign-level pool 117 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b279dca
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  93. BD-2026-08-18-COR-1 Search · campaign-level Corrected · 1 prior pool 112 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ b279dca
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  94. WR-2026-W32-PUB-E1 Search · campaign-level pool 15 beat 93.3% —
    Block range
    8414168–8813452
    Formula
    1.2.1 @ 7cdf7d7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  95. WR-2026-W31-PUB-E1 Search · campaign-level pool 141 beat 100.0% —
    Block range
    8414168–8763768
    Formula
    1.2.1 @ 7cdf7d7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  96. WR-2026-W30-PUB-E1 Search · campaign-level pool 125 beat 100.0% —
    Block range
    8414168–8713416
    Formula
    1.2.1 @ 7cdf7d7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  97. WR-2026-W29-PUB-E1 Search · campaign-level pool 120 beat 100.0% —
    Block range
    8414168–8662242
    Formula
    1.2.1 @ 7cdf7d7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  98. WR-2026-W28-PUB-E1 Search · campaign-level pool 100 beat 100.0% —
    Block range
    8355702–8611981
    Formula
    1.2.1 @ 7cdf7d7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  99. WR-2026-W27-PUB-E1-COR-1 Search · campaign-level Corrected · 1 prior pool 92 beat 100.0% —
    Block range
    0–0
    Formula
    1.2.1 @ c79e255
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  100. WR-2026-W26-PUB-E1-COR-1 Search · campaign-level Corrected · 1 prior pool 98 beat 100.0% —
    Block range
    8355702–8512403
    Formula
    1.2.1 @ 7cdf7d7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  101. WR-2026-W25-PUB-E1 Search · campaign-level pool 88 beat 100.0% —
    Block range
    8355702–8461824
    Formula
    1.2.1 @ 7cdf7d7
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

  102. WR-2026-W24-PUB-E1-COR-1 Search · campaign-level Corrected · 1 prior pool 22 beat 100.0% —
    Block range
    8413094–8413096
    Formula
    1.2.1 @ 8cd3fdc
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  103. WR-2026-W21-RERUN-E1-COR-1 Search · campaign-level Corrected · 1 prior pool 26 beat 100.0% —
    Block range
    8219538–8363696
    Formula
    1.2.1 @ b8b77a9
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  104. WR-2026-W22-PUB-E1-COR-3 Search · campaign-level Corrected · 3 prior pool 54 beat 100.0% —
    Block range
    8219538–8313781
    Formula
    1.2.1 @ a04e036
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  105. WR-2026-W21-COR-3 Search · campaign-level Corrected · 3 prior pool 16 beat 100.0% —
    Block range
    8208900–8258914
    Formula
    1.2.1 @ 91705b5
    Multiplier
    1.00×
    Registered UIDs
    256

    Open permanent record

    Prior versions of this epoch slot (archived)

  106. WR-2026-W99-SMOKE-E1 Search · campaign-level pool 50 beat 76.0% —
    Block range
    4823091–4830290
    Formula
    1.2.1 @ a1b2c3d
    Multiplier
    1.00×
    Registered UIDs
    64

    Open permanent record

Methodology

Methodology · Verification map

Every number on this dashboard is reconcilable to on-chain validator output. The links and procedure below are the entry points for verifying any published epoch independently — the chain is the source of truth; this site mirrors the spec and surfaces aggregates.

Verification path · From spec to chain

3 steps

Step 1 / 3

Read the spec

Mirrored canonical SN21 documents on this site — whitepaper, reward mechanism, epoch structure.

Open whitepaper

Step 2 / 3

Run the verifier

scripts/verify_day.py checks any published day: the signed receipt, the hash chain, the on-chain anchor, and every score recomputed.

View on GitHub

Step 3 / 3

Cross-check the chain

Block range + scoring formula commit on every record lets you query the chain directly.

Latest record

Canonical SN21 documents (mirrored on this site)

Independent verification

scripts/verify_day.py in the SN21 repo fetches a day's signed receipt, checks its signature, the hash chain and the on-chain anchor, re-runs the open-source scoring code, and names the exact entry if anything disagrees.

bash verify_day.py
python scripts/verify_day.py \
    --url https://hope-bittensor-api.onrender.com \
    --day YYYY-MM-DD \
    --miner <your hotkey>   # optional: only your entries

No keys and no archive node needed: the receipt, proof and allocation audit are public. Add --expect-anchor with the root you read from the chain to close the loop yourself. Weekly epochs before 3 August 2026 use scripts/verify_epoch.py.

How to verify

Why aggregate-only · IA decision D-05

Per-miner emissions on SN21 are derivable from per-miner scores (Reward Mechanism §"Component 2"). Publishing per-UID scores publishes per-UID emissions — doxxing-adjacent, and a strong anti-coordination signal. The chain already exposes per-UID data to anyone who queries it. The dashboard publishes what the chain does not: the shape of the distribution, the size of the qualifying pool, the baseline-beat rate, and the scoring formula version that produced them.

API & schema

API & schema · Public read surface for researchers and validators

Every published epoch is reachable as JSON without a key. Drop the cURL line into a terminal, the Python snippet into a Jupyter cell, and you have the same data the dashboard is rendering — plus the provenance fields needed to reconcile against the chain.

Endpoints · Read-only, no auth, JSON

Latest published epoch

GET 200

https://adtao.io/api/sn21/latest.json

Returns { epoch: EpochReport | null } — the freshest published row, or null until the first publish click. 60-second cache.

All published epochs

GET 200

https://adtao.io/api/sn21/index.json

Returns { count, epochs: EpochReport[] } newest first. Canonical snake_case pydantic shape per the SN21 source-of-truth model. 60-second cache.

Total fields

21

Nullable

3

Aggregate-only

D-05

Schema version

v1.0

Quickstart · Read in 30 seconds

bash terminal
curl -sSL "https://adtao.io/api/sn21/latest.json" \
  -H "Accept: application/json" | jq '.epoch'

Returns the full EpochReport JSON. Pair with jq for pretty-printing.

python stdlib only
import json, urllib.request

URL = "https://adtao.io/api/sn21/latest.json"

with urllib.request.urlopen(URL) as r:
    resp = json.loads(r.read())

epoch = resp["epoch"]
if epoch is None:
    raise SystemExit("no published epochs yet")

print(f"Epoch        {epoch['epoch_id']}")
print(f"Pool size    {epoch['pool_size']} of {epoch['total_registered_uids']}")
print(f"Beat rate    {epoch['baseline_beat_rate']:.1%}")
print(f"Formula      {epoch['scoring_formula_version']} @ {epoch['scoring_formula_commit'][:7]}")
print(f"Blocks       {epoch['block_range_start']}-{epoch['block_range_end']}")

Zero dependencies — uses urllib from the stdlib.

Schema · 21 public fields

Pydantic source-of-truth

Field Type Units Semantic
epoch_id string release-key Canonical SN21 release key — BD-2026-08-20 in the daily stream, WR-2026-W21-PUB-E1 in the weekly era. Stable URL path component.
epoch_type string enum Standard epoch classification (Daily, Search, Discovery, …). Drives epoch_type_multiplier.
epoch_subtype string | null — Optional sub-classification when epoch_type alone is insufficient.
block_range_start integer block height Inclusive lower bound of the chain blocks aggregated for this epoch.
block_range_end integer block height Inclusive upper bound. Reconcile via a Bittensor archive node.
scoring_formula_version string semver Released aggregator wire-shape version. Pinned per epoch.
scoring_formula_commit string git sha (40) Commit hash of SN21-adtao at epoch close. Links to the SHA-pinned source.
horizon_set string[] duration Time horizons the scoring formula evaluated — 7d, 14d, 28d in the daily stream; 7d, 14d in the weekly era.
epoch_type_multiplier float × Emission multiplier applied to this epoch_type. Constants per epoch type.
pool_size integer miners Count of qualifying miners (passed the participation gate).
total_registered_uids integer miners Count of UIDs registered on netuid 21 at epoch close.
pool_size_below_distribution_floor boolean — True when pool_size < 15; histogram is omitted in that case.
baseline_beat_rate float ratio (0–1) Share of pool with EMA above the conditional-prior baseline.
score_distribution object | null histogram bin_edges, bin_counts, summary {mean, median, p25, p75, max}. Null when pool below floor.
tier_distribution object — elite, competitive, participating { count, share_of_pool, share_of_emissions } + elite_floor_met.
tier_split_active boolean — False when the pool collapses to a single undivided group.
emergency_intervention object — { triggered, type, outcome, summary_text }. Routine epochs read { triggered: false }.
validator_output_snapshot_timestamp string ISO 8601 UTC When the validator output was captured for this aggregation.
chain_fetch_timestamp string ISO 8601 UTC When the chain reading completed.
commentary_markdown string | null markdown Optional human commentary. Null on routine epochs.
aggregator_version integer — Wire-shape version of the aggregator. Bumped when the same input produces different output.

Full reproducibility · Re-run an epoch end-to-end

The dashboard reads from the JSON endpoint. The JSON came from the aggregator. The aggregator reads chain + validator output. The chain is the source of truth. Re-run the aggregator with the same inputs and you get the same JSON — or a discrepancy you can act on.

bash SN21-adtao
git clone https://github.com/ippcteam/SN21-adtao.git
cd SN21-adtao && pip install -e .
python scripts/verify_epoch.py \
    --epoch-id WR-2026-W21-PUB-E1 \
    --validator-hotkey 5GxVLdpRGZN... \
    --tier-2-base https://archive.example.io

Needs Python 3.11+ and a Bittensor archive node URL. The full score recomputation also needs the operator's truth-file (reveal blob per §9.A.2). Chain-integrity mode does not.

verify_epoch.py

Reading paths · One surface, four audiences

Pick your entry