PolymarketliveKalshisoonManifoldsoon

Not all whales are smart money.

Convexly ranks every Polymarket wallet by forward-looking skill, not backward-looking PnL. Pre-registered methodology on 8,656 wallets, fully reproducible.

Are you skilled or lucky?

Paste your Polymarket wallet. Free, 30 seconds, no signup.

Free. No signup. No private key.
Reads public on-chain data only.No wallet signature.No private key.
Sample skilled-money readout . Theo4 (Edge Score rank 1)
updated just now
0x56687bf4.....0f55839
Theo4
12 resolved positionsPolymarket leaderboard rank 1
Edge Score
81.2
81.2th pctile
Calibrationbaseline-adjusted Brier
Brier 0.185 . better than median
92.7th pctile
Convictionconcentration
57% of PnL from one event
43.7th pctile
Disciplineposition count
12 positions . low volume
87.5th pctile
Total PnL
$4.07M
Concentration
57%
Over-sizing risk
0.1x Kelly
Read
High Edge Score driven by conviction, not calibration. Track record dominated by one event (57%). Not structurally copy-tradeable.
Dataset
8,656 Polymarket wallets . 15,106 Manifold users
Validation
OOF Spearman +0.514 . p < 0.0001
What you get

A leaderboard that ranks skill, not luck.

Every Polymarket wallet with twenty or more resolved positions, scored on three pillars that actually predict forward profit. Refreshed daily. Free to browse.

Calibration

Does the wallet make money despite imprecise calibration?

Baseline-adjusted Brier score. The top-profit wallets tend to be mis-calibrated in a specific, learnable direction. Internally referred to as Posture in the methodology paper.

Conviction

Is the PnL from one event, or a repeatable edge?

Share of realized PnL from the wallet's single largest event. High conviction equals barbell concentration.

Discipline

Fewer, larger bets, or spray-and-pray?

Position count, inverted. On Polymarket, more positions predicts lower signed PnL. Focus wins.

Why the naive leaderboard is wrong

PnL is mostly luck. Skill is measured differently.

Across 8,656 Polymarket wallets, calibration (Brier) explains only +0.148 Spearman correlation with realized PnL. The top 1% by profit captured 36.2% of all signed gains, and most of them did not do it by being better-calibrated. They concentrated on a few high-conviction bets that paid.

Convexly's composite ranker pulls this apart. Three pillars, fit against the cohort, out-of-fold Spearman +0.514. The wallets you want to follow are not the ones with the biggest PnL screenshots.

V1-M paper (cross-venue invariance)
Share of total signed profit
n = 8,656
Top 1% of wallets capture 36.2% of all signed profit. n = 8,656.