Wallet Ratings
LivePositive-PnL Polymarket wallets ranked by Edge Score V3b, a descriptive behavioral composite of calibration, conviction, and discipline. It describes the resolved record, not a forecast of future returns. Paste your wallet at the analyzer to compare against the cohort.
Wallets ranked
37
clear the positive-PnL gate
Top Edge Score
98.4
Median Edge Score
54.7
Reference cohort
50
frozen fit
Showing 11 wallets with 50+ resolved positions. Calibration stabilizes around 30 resolved positions per the V1 methodology paper, so this view drops the noisier tail. 26 smaller-sample wallets are hidden by this filter.
Leaderboard caveat
Leaderboard rank explains why a wallet scored highly in the historical cohort. It is not a recommendation to mirror trades or assume future wallet performance.
How to read this board. Edge Score is a descriptive behavioral composite of how a wallet positioned on its resolved public history (calibration, conviction, and discipline). It describes the past record; out-of-sample forward-skill validation is still pending, so a rank is not a forecast of future returns.
Each row also carries a skill read (skilled / not separable from luck / thin sample / too concentrated) with the wallet's realized-edge sign and 95% interval, so a high Edge Score never stands alone. A "too concentrated" marker is a flag on the read, not a verdict. The board Edge Score is a snapshot from 2026-07-01; a live analyzer read may differ as new positions resolve, so a board score and a live analyzer score read as snapshot-vs-live, not a contradiction.
This board is shown for wallets with positive realized PnL. 13 cohort wallets are held back from the public view because realized PnL is not positive or is unavailable.
0x••••…•••• · 67 positions · $57.7K PnL
Realized edge -0.2pp (95% CI [-11.2, +11.0])
72
Edge · 2026-07-01
0x••••…•••• · 61 positions · $476.9K PnL
Realized edge +0.0pp (95% CI [-11.2, +11.2])
58
Edge · 2026-07-01
0x••••…•••• · 70 positions · $145.5K PnL
Realized edge +10.2pp (95% CI [-1.5, +21.2])
55
Edge · 2026-07-01
0x••••…•••• · 218 positions · $1.6M PnL
Realized edge -1.2pp (95% CI [-5.9, +3.7])
31
Edge · 2026-07-01
0x••••…•••• · 139 positions · $955.6K PnL
Realized edge +0.2pp (95% CI [-5.3, +5.4])
30
Edge · 2026-07-01
0x••••…•••• · 79 positions · $487.2K PnL
Realized edge +0.5pp (95% CI [-6.8, +8.3])
27
Edge · 2026-07-01
0x••••…•••• · 82 positions · $86.8K PnL
Realized edge +0.6pp (95% CI [-8.9, +9.9])
22
Edge · 2026-07-01
0x••••…•••• · 72 positions · $71.6K PnL
Realized edge +2.8pp (95% CI [-4.2, +10.2])
16
Edge · 2026-07-01
0x••••…•••• · 124 positions · $132.1K PnL
Realized edge -3.1pp (95% CI [-8.6, +3.1])
15
Edge · 2026-07-01
0x••••…•••• · 111 positions · $172.9K PnL
Realized edge -0.5pp (95% CI [-4.3, +3.1])
6
Edge · 2026-07-01
0x••••…•••• · 193 positions · $348.7K PnL
Realized edge +2.5pp (95% CI [-2.5, +7.7])
5
Edge · 2026-07-01
Edge Score vs same-window PnL (in-sample diagnostic)
Monthly rolling rank correlation between Edge Score V3b and signed-log PnL computed on the SAME in-window positions. This is an in-sample contemporaneous diagnostic, not a forward test: the score and the realized PnL share the same positions and the dominant input (conviction) is PnL-derived, so it overstates predictive skill. The honest forward result is the ex-ante per-wallet temporal holdout (out-of-sample Spearman +0.11, 95% CI [0.05, 0.18], which did not clear the +0.30 threshold it was filed against). Brier-only baseline shown for contrast.
median ρ = 0.375
26 monthly windows
Edge Score distribution
How the published cohort distributes across the 0-100 scale. Median + p25/p75 marked inline.
median = 54.6
n = 50
Watch up to 10 rated wallets.
Researcher gives you a Monday digest of score movement, concentration drift, and new-position context for wallets you choose to monitor.