The glossary: 58 terms on the tape.

Event contract

A yes or no position on a real-world outcome that pays $1 if you are right and nothing if you are not.

The price, between 1 and 99 cents, is what you pay for that claim. There is no leverage and no margin call; the most you can lose is what you paid.

YES / NO

The two sides of every Kalshi market; buying NO at 30 cents is the same position as selling YES at 70 cents.

Every executed trade pairs a YES holder with a NO holder, and the two prices sum to $1. The tape records which side was the taker, which is how flow direction gets read.

Price

A contract's price in cents is the market's live estimate of the probability the event happens.

A 63 cent YES means the market puts the odds near 63 percent. Price moves are probability revisions, not share appreciation.

Market

A single yes-or-no question with its own order book.

"Will the Fed cut rates in September" is one market. Every print, sweep, and whale event on WhaleScanr attaches to a specific market.

Event

A group of related markets on the same question, one per outcome or strike.

A Bitcoin price event might carry a dozen markets, one per range. Markets nest inside events, and events inside series.

Series

The recurring family a market belongs to, like every weekly jobless claims market.

Series keep repeated questions comparable over time. The largest, KXBTC15M, was 28.9 percent of all trades, measured from one full day of Kalshi's public tape, July 2026.

Ticker

The short code identifying a series, event, or market, like KXBTC15M.

Tickers nest: the series ticker prefixes the event ticker, which prefixes the market ticker.

Strike

The threshold a ladder market is set at, like "above $70,000" in a Bitcoin range.

Adjacent strikes in one event let the market price a whole distribution rather than a single line.

Settlement

The moment a market resolves and every contract pays $1 or $0.

Settlement follows the market's written rules and named resolution source, not the headline. The payout is final; the position ceases to exist.

Resolution source

The named data source a market's rules use to decide the outcome.

If the rules name the Bureau of Labor Statistics, the BLS figure settles the market even when other sources disagree.

Expiration / close

When trading in a market stops.

Close can arrive on schedule or early, once the outcome is known. A market can stop trading well before it settles.

Order book

The live list of resting bids and offers waiting to trade.

Kalshi runs one central book per market. Every print on the tape is a resting order meeting an incoming one.

Bid / ask

The best price buyers will pay and the best price sellers will take.

A market quoted 47 bid, 49 ask trades the moment someone accepts one of those prices.

Spread

The gap between best bid and best ask; the cost of trading right now.

Cross it and you pay it. In thin markets the spread is often a larger cost than the fee.

Limit order

An order that names a price and waits.

If it fills, you were the maker. The risk is that it never fills and the market moves on without you.

Market order

An order that takes the best available price immediately.

Certainty of fill, uncertainty of price. Large market orders in thin books produce slippage, and sometimes sweeps.

Maker

The trader whose resting order gets filled; they set the price and waited.

Makers pay a quarter of the taker fee. Patience is priced into the schedule.

Taker

The trader who crosses the spread to fill against a resting order; they paid for immediacy.

The tape's side field records the taker, so tape reading is mostly taker reading.

Maker vs taker fees

Kalshi charges takers 0.07 times price times one minus price per contract, peaking at 1.75 cents on a 50 cent contract; makers pay a quarter of that.

The fee is largest exactly where trading is heaviest: 23.9 percent of taker dollars traded in the 40 to 60 cent band, measured from Kalshi's public tape by WhaleScanr, July 2026.

Liquidity

How much you can trade without moving the price.

Depth in the book absorbs size; volume on the tape only records it. Whale detection is calibrated to each market's own liquidity for exactly this reason.

Slippage

The gap between the price you saw and the price your full order actually got.

A 5,000-contract order fills through several price levels, each one worse. On the tape, that arithmetic looks like a sweep.

Volume

Contracts traded over a period.

Volume measures activity, not commitment. A contract bought and sold ten times is 10 volume and 0 open interest.

Open interest

Contracts currently held open; money committed, not trades made.

Volume counts turnover, open interest counts positions still alive. Rising open interest alongside rising volume means new money rather than recycling.

Position

The contracts you currently hold in a market.

Positions are private. The public tape never shows them, which is why WhaleScanr works in flow, not portfolios.

Market maker

A firm or trader quoting both sides continuously to earn the spread.

Their business is the spread plus the maker discount, not an opinion on the outcome. Tight books are largely their work.

Tape

The public record of every trade as it prints: price, size, side, and time, with no names attached.

Kalshi's tape is fully anonymous. Everything WhaleScanr shows is built from it, roughly 7.33 million trades on a single measured day in July 2026.

Print

A single executed trade on the tape.

Each print carries price, contract count, taker side, and timestamp. Everything else a reader concludes is inference.

Whale

Outsized flow judged against the market it lands in, not a flat dollar bar.

$20,000 is a whale in a thin Senate race and wallpaper in a 15-minute Bitcoin market. The relative framing is the whole point.

Whale event

A print or sweep big enough to clear WhaleScanr's market-relative size thresholds.

Events are what the alerts page lists for the tracked wallets. A market with too little history produces no events: unknown liquidity means no judgment.

Sweep

A rapid burst of same-side prints within seconds, usually one trader clearing the book.

WhaleScanr flags a sweep at 5 or more same-side prints with gaps of 10 seconds or less, treated as one event. On Kalshi there is one order book, so a sweep is a burst in time, not a route across exchanges.

Block trade

A large privately negotiated trade reported to the exchange; effectively absent from Kalshi's public tape.

The block flag was true on 0 of 7.33 million trades measured from one full day of Kalshi's public tape, July 2026, and on 0 of roughly 14 million rows examined overall. Size on Kalshi arrives as prints and sweeps.

Notional

The dollar value of a trade: contracts times price.

1,000 contracts at 62 cents is $620 notional. WhaleScanr ranks flow by notional because raw contract counts flatter cheap contracts.

N× typical

How a trade compares with this market's typical large trade, its 95th percentile print; 8× means eight times that size.

The yardstick is big trades, not the average, because the average trade in a thin market is a $40 ticket. Past 25× the badge reads 25×+, since everything beyond that is simply off the charts.

Entry price

The price a whale actually paid for its side, which is also the probability it bought.

A whale entering YES at 34 cents bought a 34 percent claim. Entry price is what makes whale flow readable as opinion.

Lean

Which side the recent whale dollars in a market have favored, YES or NO.

Lean is an aggregate of recent whale notional, not a forecast. A market can lean YES and settle NO without anything being wrong.

Flow cluster

A heuristic grouping of anonymized trades that look related; not a verified person or account.

The tape has no identities, so clusters are behavioral shapes: same series, same side, similar size, close in time. Every cluster carries a confidence score for exactly that reason.

Confidence

How likely a cluster is one trader rather than several; an estimate, never a certainty.

More events, steadier sizing, and regular timing raise it. It is capped below certainty by design, because anonymous tape data can never confirm a single actor.

Whale Score

A 0 to 100 read on how big, sustained, and broad a cluster's flow is; it describes activity, not advice.

The score is gated by confidence, so a loud but loosely grouped cluster ranks below a quieter, tighter one. It measures presence on the tape, not correctness.

Flag

A trade WhaleScanr marked as whale-sized for its own market, at the moment it printed.

A flag is a record that something unusual happened, not a prediction that it was right. Every flag's outcome, including the ones that settled against the flagged side, is on the record page.

Typical

A market's own normal trade size, which is what a print is measured against before it can be called large.

Ten thousand dollars is enormous in a quiet market and unremarkable in a busy one, so a flat cutoff would flag the wrong things. N times typical means N times that market's own bar, never a sitewide number.

Realized

Profit that has actually settled, as opposed to the paper value of a position still open.

The word is used for two different sets on this site, and they are labelled so they cannot be confused. REALIZED ON THESE N is the profit from the settled positions WhaleScanr has sampled. LEADERBOARD is Polymarket's own lifetime figure for that wallet, which is larger because it counts everything we have not sampled.

ROI

Realized profit divided by the dollars staked to earn it, on the same set of positions.

ROI is weighted by stake while an average entry price is not, which is why a wallet can win every position at an average entry of 51 cents and still show a smaller ROI than that suggests. A large stake entered high and small stakes entered low produce exactly that shape.

Convergence

Two or more tracked wallets taking the same side of the same market within a short window.

It is counted per cluster, not per wallet, because wallets that trade together would otherwise be counted several times over and make one decision look like several. Convergence describes agreement among traders, and agreement is not evidence that they are right.

Edge

The gap between the probability you believe and the price you pay.

Buying at 40 cents with a genuine 45 percent view is 5 points of edge, before spread and fee take their share.

Expected value

What a position is worth on average across outcomes, after its cost.

A 40 cent contract with a true 45 percent chance is worth 5 cents on average, minus fees. One trade proves nothing; expected value is a long-run quantity.

Calibration

Whether prices match reality over many outcomes; well-calibrated 70 cent contracts win about 70 percent of the time.

Calibration is measured across hundreds of markets, never one. Departures from it, like longshot bias, are among the best documented facts about prediction markets.

Longshot bias

The tendency of cheap contracts to win less often than their price implies.

Published research on Kalshi finds longshot buyers overpay on average, with the difference collected by the resting side of the book. The 5 cent lottery ticket is usually priced above its true odds.

Arbitrage

Buying and selling the same outcome at different prices to lock in the difference.

On a single exchange this can mean YES and NO in the same market summing below $1. Fees usually close the gap before retail can reach it.

Hedge

A position taken to offset risk you hold elsewhere.

A trader long rate-sensitive stocks might buy NO on a rate cut. On the tape a hedge looks identical to an opinion, one reason flow is read with caution.

Degen

Self-mocking slang for a trader taking reckless positions.

Worn as a badge, almost never as an apology. Natural habitat: the 15-minute Bitcoin market at 2am.

Sharp

A trader with a real record of beating the market.

Borrowed from sports betting, where books limit sharps. On an anonymous tape, sharpness is claimed often and verified never.

Printing

Making money fast; a position paying off.

From money printers, not the tape's prints, though a winning position tends to involve both.

Full port

Betting the whole portfolio on one position, almost always told as a cautionary tale.

Stories that begin with full port rarely end with printing.

Fade

To bet against another trader's position or the public narrative.

Whale watchers split into followers and faders. The tape does not say which camp is right.

Tailing

Copying someone else's bet after they place it.

The opposite of fading. By the time a whale print is visible, the price it got is usually gone.

Mogged

Outtraded, badly.

What happened to the resting orders on the wrong side of a sweep.

Fudded

Sold out on fear rather than information.

From FUD: fear, uncertainty, doubt. Every panicked sale has a buyer, and sometimes the buyer is a whale.

Rulescucked

Losing on a technicality in the settlement rules rather than the outcome everyone thought they were trading.

The market asked one question and the rules answered another. Reading the resolution source in advance is the only known vaccine.