TWAP Orders Are Live: Spread Entries and Exits Across Time, Not Price

You decide to take a $200,000 EUR/CAD position because the fundamental setup is clean. You hit market. The venue's forex book is deep enough to fill the first $40,000 at the touch, the next $60,000 fifteen pips worse, and the remainder another ten pips through that. By the time the order's done, you paid up twenty-plus pips in slippage on a trade that was supposed to earn you forty. That's not a bad trade; it's a bad execution of a good trade.

TWAP — time-weighted average price — is the standard solution. Instead of one $200,000 market order, you tell the engine to send $20,000 every two minutes for twenty minutes. Each slice hits the top of a replenished book; the average fill price is close to the midpoint over the window; the market impact per slice is small enough that passive liquidity has time to refill between them. You get the position; you pay less for it.

TWAP orders are now live on JTX Markets across every tradable instrument.

How the UI works

On any symbol's trade page there's now a TWAP tab next to Limit and Market. The dialog asks for three things:

• Total size — the parent quantity you want filled, in contracts or base units.
• Duration — how long the execution window should last. Preset buttons for 5m / 15m / 30m / 1h, with a custom field underneath.
• Slice count — how many child orders to break the parent into. Defaults to one per minute of duration, rounded, so a 15-minute window sends 15 slices by default. You can override.

An optional Price cap field lets you bound the execution — the TWAP won't fill a slice above (for buys) or below (for sells) your cap. Use this when you have a view on fair value and don't want to chase a runaway move.

Submit, and the parent appears in the Orders tab with a status chip showing "TWAP · 3/15 filled · 12m remaining". Each child fill rolls up into the parent so the P&L and position accounting are clean. Cancel the parent at any time; any slices already filled stay filled, any remaining slices are cancelled.

When TWAP actually helps

Not every trade needs it. Three concrete cases where it does:

• Illiquid symbols. Anything beyond the top 10 crypto majors or the forex G7 pairs tends to have a top-of-book clip size smaller than most meaningful positions. On equity perpetuals and commodity perps, TWAP will cut your effective spread by half on positions in the $50K+ range.
• Low-volatility exit timing. You want out of a position but there's no urgency. A single market order concentrates all your exit slippage in one moment of the book's state; a 30-minute TWAP samples the book thirty times. If the market doesn't move much, the result is a cleaner average than you'd get hitting one random tick.
• Weekend overnight. On a weekend index position you want to roll out of before Monday's cash-market reopen, a Sunday-afternoon TWAP across 15 or 30 minutes avoids concentrating your exit at the thinnest liquidity moment of the week.

Cases where TWAP is the wrong tool:

• You're reacting to news. If the headline just dropped, you want fill *now*, not spread over 20 minutes. Market is the right call.
• The symbol is extremely liquid. On BTC-PERP or EURUSD at normal size, top-of-book depth will absorb most reasonable trades without meaningful slippage. TWAP adds latency with no benefit.
• You're hitting a stop or TP. Reduce-only exits should use the native TP/SL mechanism (STOPMARKET or reduce-only LIMIT), not a TWAP, so the engine's risk layer sees them as risk-reducing.

The mental model

TWAP is one tool in the execution-algorithm family. The others — VWAP (volume-weighted), POV (percent of volume), Implementation Shortfall — are all variations on the same core idea: break a large order into smaller pieces, time them against something, and reduce the market impact of the parent. For retail-sized trading on 24/7 markets with always-on liquidity, TWAP is usually the right starting point because the implementation is simple and the result is predictable.

Rule of thumb for duration: if a single slice would move the market by X basis points, you want slices small enough that each one moves the market by half a basis point or less. On a typical forex perp with top-of-book depth around $50K, that puts the slice ceiling around $25K. A $500,000 parent would then want 20+ slices, and 20 slices across a 20-minute window is one per minute — a common and workable TWAP profile.

Access and limits

• Minimum TWAP size: $20,000 USDT notional. Below that, a single market order is almost always cheaper after accounting for fees per slice.
• Maximum TWAP duration: 4 hours. Beyond that the trade starts looking more like a strategy than an execution.
• Maximum slice count: 120 slices per parent. 1-second TWAPs are supported but not recommended; the realistic floor is around 15 seconds per slice.
• Fees: standard per-slice. A 20-slice TWAP pays twenty takers' worth of fees, not one — which is still cheaper than the slippage saved on anything above the minimum threshold.

TWAP works on crypto perps, forex perps, indices, commodities and equity perpetuals. It respects reduce-only mode, so you can set up a TWAP exit from a position without worrying about the parent accidentally flipping you long when it meant to flatten you.

If you have an account, the TWAP tab is live on the trade page starting today. If you don't yet, open one at app.jtxmarkets.com or start on a challenge tier if you want to learn execution habits on paper before trading your own capital.