Mispricd
Menu
DataEvery live figure on one page MethodEvery rule, source and correction

Markets · What the markets show now

BoardEvery open Kalshi temperature bracket, read hourly GamesKalshi game markets: the price and the numbers, hourly Whale finderPolymarket weather wallets, ranked from their fills, weekly

Record · What held up once it settled

ScorecardThe settled record, week by week CalibrationDoes a 20c bracket pay one time in five? CitiesEach city's record: ruled out, the leader, calibration, when the day is decided DaysThe daily box score: every settled day, city by city CallsDated calls with a probability, graded on their date ReceiptsThe calls, the settled weeks and the retired desk The deskEvery desk's lights, from the event log, live

Research · What we found when we looked hard

StudiesFive questions, called before they were measured PiecesLong reads, decoded from the fills and the tape SymbolsGematria, graded on 22,990 games

Terms Privacy X Owner sign-in

The whale finder · Polymarket temperature ladders · last two weeks

Who is making money on the weather ladders, from their own fills.

Every taker fill of every temperature market on Polymarket over the window, grouped by wallet, resolved against the venue's own result and scored the way the whale piece scores: first-in first-out, what is still held settles at a dollar or nothing. Each wallet is then profiled from its fills alone, when it buys against the market's day, how long it holds, which side and which brackets. And for the wallets that rank, the paper mirror: what a copier who saw each fill on the public feed and lifted the next ask would have made. No order is placed anywhere; this is a measurement.

loading the finder…

Hypothetical results · the copier tileThese results are based on simulated or hypothetical performance results that have certain inherent limitations. Unlike the results shown in an actual performance record, these results do not represent actual trading. Also, because these trades have not actually been executed, these results may have under- or over-compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated or hypothetical trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to these being shown.

The wallets that rank

Wallets with at least twenty buys and five hundred dollars paid over the window, by gross profit on the buys that have settled. The window is the last two weeks; the feed names takers, so a wallet that only rests orders does not appear. The address is the wallet's public identity on the chain; the name is the pseudonym Polymarket shows for it. Per share is gross, before the venue's fee, over the shares that settled.

#walletprofilebuysmarketssharespaidgrossper shareheld, wonmedian holdon the daylong NO

The profiles

A label from the fills alone. The thresholds are the whole rule; a wallet that fits none is "mixed", and one below the sample floor is "small sample" and does not rank.

observation bot80% or more of its dollars bought on the market's own day, and a median hold of an hour or less: it buys once the station has decided and sells at once.
tail seller70% or more of its dollars long NO, 60% or more of those on the ladder's tail brackets, and it holds six hours or more, or to the end.
longshot buyerA median entry of 20c or less, YES-equivalent, and 70% or more of its dollars long YES.
two-sidedBetween 35% and 65% of its dollars on each side over a hundred or more buys: a book, not a bet.

The paper mirror

How the mirror works

For every settled buy of a ranked wallet, the copier sees the fill on the public feed, waits the lag, and pays the price of the next ask lifted on that outcome; no print within six hours is no fill. "Hold" keeps the position to resolution. "Exit" copies the wallet's sales too, at the next bid hit after each. Slip is the median cents between the wallet's price and the copier's. The wallet's own figure is on the same buys, so the two columns are the cost of arriving second.

A single wallet contributes a handful of buys to this, far too few to decide anything, so the table that counts is the pooled one: every wallet of a kind, added together. "Ranked" is every wallet past the sample floor.

grouplagwalletsbuysfilledown, per sharecopier, holdcopier, exitslip

wallets = those with a settled buy in the mirror, of those in the group · the copier takes the wallet's full size at the next print, which flatters it: the print may have been smaller than the order it would have needed.

Hypothetical results · the paper mirror, above and belowThese results are based on simulated or hypothetical performance results that have certain inherent limitations. Unlike the results shown in an actual performance record, these results do not represent actual trading. Also, because these trades have not actually been executed, these results may have under- or over-compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated or hypothetical trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to these being shown.

The same, wallet by wallet

The published wallets' own rows. Read the sample size before the cents: most of these are a handful of buys.

walletlagbuysfilledown, per sharecopier, holdcopier, exitslip

The crowd, by the price it paid

Every taker buy in the window, not only the ranked wallets', by the price of what was bought: a No bought at 93c is a 93c favourite, not a 7c long shot. "Per dollar" is the gross result on the buys that have settled, sales at their price and the rest at $1 or $0, over what those buys cost; "held, won" is how much of what was held to the end paid. Before any fee. It is call C-011's measure: whether Polymarket's weather long shots lose money the way every venue's do.

price paidbuyspaidsettledgrossper dollarheld, won

Where the numbers come from

agents/whale_finder.py lists every temperature event on Polymarket whose day falls in the window (Gamma), pulls every taker fill of every market from the public data API, and scores each wallet with the same first-in first-out rule as the whale piece, one market at a time as the fills arrive. It runs weekly inside the logger, checkpointing its place every couple of thousand markets so a restart resumes rather than starting the three-hour pass over, writes one line per market and wallet to the private ledger bucket, and publishes the ranked rows here; --report rebuilds the leaderboard from that file offline. What it cannot see: the feed names the taker only, so a wallet that only rests orders is invisible to it, and so is any trading outside the temperature ladders.