ForecastEx

How it trades

Trade sizes, where on the probability axis the volume sits, and the biggest individual prints ForecastEx has published.

Trade size mix

Share of contracts, not of trade count — a book can be almost all small trades by count and still be majority block-traded by volume.

Volume by probability

Where traders choose to put money on the probability axis. This is behaviour, not accuracy — see Outcomes for whether those prices came true.

Largest individual trades

Switch to one-party stake to surface trades at extreme prices, where one side risks nearly the full dollar and the other almost nothing.

Largest trades in small markets

Trades that were large for the market they happened in, not just large in isolation.

How this is measured

Market names are whatever the venue publishes. ProphetX and Novig name their fixtures; DKeX, Polymarket US and Underdog publish an opaque contract id. Nothing is renamed here.

Taker stake appears only where there is a taker. Novig is the only competitor publishing an aggressor flag, so on every other venue that ranking is absent rather than blank.

“% of market” is share of a window, not of a lifetime. These tapes run only as far back as collection does, so a market that was already busy before collection started reads as more concentrated than it truly was. A market must also have traded at least 20 separate times to appear, which stops the table degenerating on venues whose market id is close to one-per-trade.

See also: ForecastEx · Activity, and Trading behavior across venues for the comparison.