TWAP stands for time-weighted average price. The term is used in two ways: as a benchmark (the average of a price sampled over a period, each sample weighted by the time it covers) and as an execution method (a large order cut into smaller slices sent on a schedule across that period). Neither predicts where the price will go.

TWAP as a price average

Take a price sampled once a minute at 100, 102 and 101. Each sample covers the same time, so the TWAP is (100 + 102 + 101) ÷ 3 = 101. If samples cover unequal intervals, each price is weighted by its duration. Any TWAP figure depends on four choices: which price source (last trade, bid/ask midpoint or an oracle), the start and end time, the sampling interval, and how gaps in the data are handled. Two charts labelled "TWAP" can show different numbers because they made different choices.

TWAP as an order method

A TWAP order spreads one big order over time. For example, buying 10 units over 5 hours might mean sending 2 units every hour. Traders use it to:

  • reduce the price impact a single large order would have on a thin market;
  • avoid revealing the full order size at once;
  • get an average fill close to the period's average price.

The final average still depends on liquidity, how far the price moves during the window, the order settings (limit prices, slice size) and whether every slice actually fills.

How to use TWAP sensibly

  1. Pick the window from the order size and the market's volume. A slice should be small next to the normal volume in each interval.
  2. Decide what happens if the price runs away. A limit price on each slice protects you but may leave part of the order unfilled.
  3. Compare the result with the right benchmark. Judge the fill against the TWAP of the same window and price source, not against a different chart.
  4. Paper-trade first if the tool supports it, to see how the schedule behaves, remembering that simulated fills are not proof that real ones would happen at the same prices.

Common mistakes

  • Treating TWAP as a strategy that makes money. It only controls how an order is executed.
  • Slicing too thin or too fast, so each slice still moves the price or fees pile up.
  • Comparing TWAP values from different sources (exchange trades vs an oracle) as if they were the same number.
  • Using a predictable schedule in a market others watch; some tools add randomness to slice timing for this reason.

Alternatives

VWAP (volume-weighted average price) weights each slice by expected trading volume, so more is traded when the market is busier; it usually suits markets with a clear daily volume pattern. A single limit order works when the order is small next to market volume. Many exchanges and brokers offer TWAP or other algorithmic orders built in; check your venue's own documentation for how its version handles slices, limits and cancellations.

What's changed

9 November 2026: this answer used to end with a section on TwapTrade, a Quantumines product, as a trading workspace with bot controls, paper trading and TWAP guides. TwapTrade has since replaced that workspace with a prediction market, and its TWAP guides are no longer online, so the section and its links were removed. The explanation of TWAP above is unchanged.

This answer explains a method; it is not investment advice.