What is TWAP? Definition, Formula, and Example
TWAP (time-weighted average price) is an execution benchmark and algorithm that slices a parent order into equal-sized child orders released at uniform time intervals across a fixed window, regardless of trading volume.
Plain-English Definition
TWAP is time-weighted average price — both a benchmark and a class of execution algorithms that split a parent order into equal-sized child slices released at uniform time intervals across a fixed window, weighting every minute equally regardless of trading volume. It originated on institutional desks in the early 1990s as the simpler alternative to volume-weighted execution and is now standard on every EMS (Bloomberg EMSX, Portware, Instinet) and increasingly exposed to retail through brokers like Interactive Brokers and tastytrade.
How TWAP is Calculated
The benchmark formula is the arithmetic mean of midpoint prices sampled at fixed intervals:
TWAP = (Σ P_i) / N
where N is the number of equal-duration intervals in the window and P_i is the price (usually midquote or last) at the end of interval i.
For execution scheduling, a parent order of size Q is sliced into N child orders of size Q/N, released every T/N minutes. Production implementations randomize slice timing ±10–20% and slice size ±15% to defeat HFT pattern-detection that would otherwise front-run the next predictable child.
TWAP vs. VWAP
| TWAP | VWAP | |
|---|---|---|
| Weighting | Equal per minute | Proportional to volume |
| Volume forecast | Not required | Required |
| Slice size | Constant | Variable, U-shaped |
| Best for | Illiquid names, low urgency | Liquid names, full-day fills |
Worked Example
A fund needs to buy 100,000 shares of AAPL between 10:00 and 14:00 ET — a 240-minute window. A pure TWAP algo releases ~417-share child orders every minute. The benchmark price is the simple mean of the 240 one-minute midpoints; if that average is $187.42, the fill is graded against $187.42 (for a buy, lower is better).
In a comparable VWAP schedule on the same name, the algo would front-load ~12% of volume into the first 30 minutes and ~18% into the closing 30 — the U-shaped intraday curve. The TWAP, by contrast, would have traded exactly 12.5% in each of those two windows. On a fund-wide 5-million-share program, that difference shifts roughly 280,000 shares away from the close.
When Traders Use It
- Low-urgency rebalancing where minimizing market impact matters more than minimizing tracking error against the day's VWAP.
- Illiquid stocks where VWAP would over-trade during a volume spike that the algo itself triggered.
- Stealth accumulation — equal slices conceal parent size better than VWAP's volume-following profile, which broadcasts size whenever printed volume jumps.
- Multi-day benchmarking — TWAP is the standard yardstick for grading execution quality on programs that span more than one session.
Limitations and Misconceptions
TWAP systematically underperforms VWAP in liquid large-caps where 30–40% of daily volume prints in the closing auction. Predictable slice cadence makes un-randomized TWAPs vulnerable to HFT detection — Pragma Securities measured a 4–7 bp impact penalty on fixed 5-minute slicing in mid-cap names. TWAP also ignores adverse news: a 3% gap mid-window does not stop the schedule, which keeps lifting offers at every interval. Finally, TWAP is not "passive" execution — most implementations cross the spread on every slice to guarantee completion, which makes it a takes-liquidity benchmark, not a saves-liquidity one.