Adaptive Exits for Arbitrage Trades: Diff Excess Factor, Slippage Factor and Critical Slippage Friday October 23rd, 2026 – Posted in: Arbitrage Software

The short answer

Short answer: Diff excess factor % gives trades opened on a strong signal more room to run. Slippage factor % secures profit sooner when the fill was worse than the requested price. Critical slippage % puts a much tighter stop on a trade whose fill was so bad that the arbitrage is practically gone. All three are off by default (0), so nothing changes until you set them. The way the strategy trails and locks profit stays exactly as it is; only the numbers it uses for each trade change.

Why one exit for every trade is a compromise

In latency arbitrage a trade opens because the fast feed shows a price the slow broker has not reached yet. How much there is to earn depends on two things that are different for every trade: how large the price difference was when the signal fired, and how much of that difference was lost to slippage before the order was filled.

Fixed exit settings ignore both. A strong signal is trailed as tightly as a marginal one and is often closed before the price has finished catching up. A trade with a poor fill keeps waiting for a profit target it can no longer reach. And a trade filled after the broker had already caught up is protected by the same wide stop loss as a trade with a real edge, although what remains is a random price move. The three new parameters let each trade carry its own exit levels for exactly these three cases.

If slippage itself is new to you, our guide to measuring and reducing forex slippage explains where it comes from and how to read it in your own fills.

Where to find the parameters

The settings are per instrument, in four groups of strategies. Each one can be included in optimization like any other instrument parameter.

For the lock arbitrage strategies, Phantom Drift and Umbra Lock the parameters act only on the position that carries arbitrage risk. Locked legs, martingale orders and other positions keep your normal settings.

What the parameters measure against

All three parameters compare the trade with the signal threshold the strategy already uses:

• Diff to open in OneLeg;
• Diff to open 1 or Diff to open 2 for the broker that traded, in the Lock strategies;
• Diff to close in Umbra Lock;
• Min move when the strategy trades on the impulse alone (Diff to open 0 and Min move set). Min move is explained in our article on impulse entry and reverse exit.

Two values come from every fill: the difference (how large the signal was when the trade opened) and the slippage (how many points worse than requested the order was filled). The excess is the difference minus the threshold.

Diff excess factor %: more room for strong signals

Diff excess factor % raises the trade’s Min profit and Trailing step by this percentage of how far the signal went beyond its threshold.

Increase = (Difference − Threshold) × Diff excess factor %

Example: a trade opened on a 25-point difference with Diff to open 10 has an excess of 15 points. At 50%, both Min profit and Trailing step grow by 8 points (7.5, rounded).

The reasoning is simple. A larger difference means the slow broker has further to catch up, so the expected move is larger. With fixed exits a strong signal is trailed as tightly as a marginal one and is often closed before the price has finished catching up.

• Benefit: strong signals are given room to run, which raises the average profit of the best trades.
• Unchanged: a signal right at the threshold has no excess and keeps your normal Min profit and Trailing step.
• Switched-off exits stay off: a Min profit or Trailing step set to 0 stays 0; the parameter never switches on an exit you have switched off.

Slippage factor %: lock profit sooner after a bad fill

Slippage factor % lowers the trade’s Min profit and Trailing step by this percentage of the slippage the fill suffered.

Decrease = Slippage × Slippage factor % (neither value goes below 1 point)

Example: a fill 8 points worse than requested, at 100%, lowers both values by 8 points.

Slippage is part of the expected gain that is already lost: the price has less distance left to move in your favour. Lower exits secure what remains sooner, instead of waiting for a target the trade can no longer reach.

• Only adverse slippage counts. A fill better than requested changes nothing.
• Lock strategies, PhantomDrift and Umbra Lock: the slippage used is that of the close which created the unhedged position.
• Benefit: fewer trades that reach a small profit, fail to lock it and turn into a loss.
• Combined with Diff excess factor % the two adjustments add up: a strong signal with a bad fill can end up with values close to the normal ones.

Critical slippage %: a tight stop when the gap is gone

When slippage takes more than this percentage of the signal’s difference, the trade’s stop loss is replaced by a much tighter one based on the broker’s spread.

If Slippage > Difference × Critical slippage %: Stop = Spread × max(2, Difference ÷ Slippage)

Example: at 50%, a trade opened on a 53-point difference and filled 46 points worse (87% of the difference) gets the tight stop. With a broker spread of 10 points at the moment of the fill, that is a 20-point stop.

Such a trade has practically no arbitrage left: the slow broker had already caught up while the order was executing. What remains is a random price move, so the best course is to keep its possible loss small.

• The tight stop is the broker’s spread at the moment of the fill times 2, or times difference ÷ slippage when that is larger. A spread of 10 points gives a 20-point stop.
• It can only reduce risk: the tight stop is never wider than your Stop loss. It also applies when Stop loss is 0.
• Umbra Lock: the tight stop may go below the usual 50-point minimum of Stop loss, for the same reason.
• Hidden stop only: it changes the strategy’s hidden stop; a stop loss already sent to the broker with the order is not modified.
• Benefit: the trades that cause the biggest avoidable losses, the ones filled after the opportunity was gone, are cut early.

Worked example: three XAUUSD trades

Three XAUUSD trades (1 point = 0.01) show how the same settings produce different exits. Instrument settings: Diff to open 40, Min profit 20, Trailing step 20, Stop loss 50, broker spread 10. New parameters: Diff excess factor 50%, Slippage factor 100%, Critical slippage 50%. Without them, all three trades would get Min profit 20, Trailing step 20 and Stop loss 50.

• A has a 20-point excess and little slippage: 20 + 10 − 2 = 28. Both exits move out by 8 points and the trade has room to run.
• B has a 5-point excess (+3, rounded) but lost a third of its difference to slippage (−15): it locks profit at 8 points instead of waiting for 20.
• C is a real trade from live trading. Slippage took 87% of the difference, so Min profit and Trailing step fall to the 1-point floor and the trade gets the critical stop of 20 points. The price then moved 51 points against it: the loss would have been about 20 points instead of 51.

Recommended starting values

Start with the values below and confirm them on your own instruments and brokers with a backtest. They are starting points, not guarantees: the best values depend on how often and how far your brokers slip.

How to tune them, step by step

Tune these parameters after the core ones. In our step-by-step video on building your own template (below), the order is: Diff to open, Min profit, Pips for min profit and Take profit first, then Stop loss and Take profit, then Reverse factor and the newer controls, including these three.

Good to know

• Calculated once per trade, at the moment of the fill, in whole points, and kept for the whole life of the trade.
• Only arbitrage trades are affected. Lock legs, martingale orders and other positions keep your normal settings.
• Trailing step in percent: when Trailing units is “Percent”, the trailing step is a share of the profit and is left unchanged; Min profit is still adjusted.
• After a restart, trades that were already open use your normal settings.
• OneLeg needs the Advanced optimization licence for these parameters. Without it they are hidden, not used and not offered in optimization. Instruments with Diff units “Percent” are supported.
• OneLeg with “Keep limit orders”: these fills carry no slippage information, so only Diff excess factor % acts on them.
• Check it in the log. For every affected trade the strategy log has a line starting with “Exit values for”. It shows the difference, the excess, the slippage, the old and new Min profit and Trailing step, and the tight stop when Critical slippage % applied.

Video: build your own template

This 10-minute video shows the full process on LockCL2: collecting ticks, measuring your real execution time, optimizing in stages and stress-testing the result with slower execution.

FAQ

Trading forex, CFDs and crypto carries a high risk of loss. The examples in this article explain how the parameters calculate exit levels; they are not a forecast of results. BJF Trading Group is a software developer, not a broker or investment advisor.