What is realized entry edge?

The answer first

Every prediction-market entry is an implied probability: pay 40 cents on a binary market and you are acting as if the event is at least 40 percent likely. Realized entry edge asks, across a wallet's resolved record, how much more often its entries resolved true than the prices it paid implied. The unit is probability points: +5.0pp means entries resolved true five points more often than the paid prices implied. Convexly reads the statistic with a bias-corrected and accelerated (BCa) bootstrap 95 percent interval, never a point estimate on its own, because at realistic sample sizes the point estimate is mostly noise.

Worked example: the bigger number is the weaker read

Two real rows from the frozen 2026-06-09 scan of our own published top-50 cohort (full table at /research/top50-skill-scan):

• Wallet 0xaaaf7f: realized entry edge +12.2pp across 32 resolved positions, 95 percent interval [-0.7, +23.7]. The interval includes zero, so the record is not separable from chance despite the large point estimate.
• Wallet 0xc2fb28: realized entry edge +5.6pp across 137 resolved positions, 95 percent interval [+1.2, +9.5]. Less than half the point estimate, but the interval clears zero: the stronger read, though as one uncorrected test of 35 it is consistent with chance at the cohort level, it is not in the FDR-corrected cleared set, and its net PnL on the board is negative.

That inversion is the whole lesson. A +12.2pp point estimate on 32 positions tells you less than a +5.6pp estimate on 137, because the interval width scales with sample size. Anyone quoting a wallet's edge without its interval and its denominator in the same breath is quoting noise.

The four-state verdict it feeds

On the analyzer, the skill badge, and the cohort scans, realized entry edge feeds a deterministic four-state read, evaluated top-down with demotions first:

• Flagged: a single event drives at least 60 percent of the net result, so the figure cannot be separated from one outcome. This overrides any point estimate.
• Insufficient: fewer than 30 resolved positions, or no usable interval. Too thin to tell an edge from chance either way.
• Not separable from chance: the BCa interval includes zero. Not a claim of no skill; a claim that the record cannot establish one at this sample size.
• Skilled: renders only when all gates pass and the BCa lower bound is strictly above zero. Retrospective and in-sample; a past read is not a forecast.

The frozen definitions live in the lexicon, and the interval construction is documented on the methodology page.

What realized entry edge does NOT do

It does not measure dollar PnL, which on a fat-tailed market is dominated by a few large positions. It does not forecast future trades. And a single positive interval does not survive being one test among many: that is the job of the false-discovery-rate correction, which is applied whenever a whole cohort is screened at once.

Read a wallet's edge

Paste any Polymarket wallet address at the analyzer to get its realized entry edge, the BCa 95 percent interval, the resolved-position count, and the four-state read. Free, no signup, public on-chain data only.