The D'Alembert strategy

The sensible cousin of Martingale. Add a unit after a loss, remove one after a win — the same recovery idea, but arithmetic instead of exponential, which is why it's the popular choice for running longer sessions without white knuckles.

Load D'Alembert in one click →

How it works

Pick a starting stake and a unit size. Lose, and the next stake goes up by one unit. Win, and it comes down by one unit. The stake never falls below the initial stake. Because the steps are additive rather than multiplicative, a long losing run raises your exposure gradually rather than violently.

Key parameters

A worked example

1. Start with an initial stake of 1 USD and a unit of 2 USD.
2. The first trade wins. The stake does not reduce below the floor, so it stays at the initial stake of 1 USD.
3. The second trade loses. The stake for the next trade rises by 2 USD, and keeps rising by 2 USD per loss.
4. The following trades win, so the stake steps back down by 2 USD each time — from 3 USD back to 1 USD.

How long your balance lasts

D'Alembert is less risky than Martingale, but it still has a floor you can hit. The rounds you can sustain are:

R = ( √( s² + 2·f·B + (f − s)²/4 ) − s + f/2 ) / f, approximately

• R — consecutive losing rounds you can sustain
• B — your loss threshold
• s — the initial stake
• f — the unit increment

With a loss threshold of 100 USD, an initial stake of 1 USD and a unit of 2 USD, R works out to 10 rounds — ten consecutive losses cost 100 USD and stop the bot. Compare that to Martingale, where the same 100 USD buys you roughly six losses in a row.

The trade-off you're making

Gentler escalation is not the same as safety. D'Alembert loses money at the same expected rate as every other staking plan on a negative-edge contract; it simply loses it as a steady bleed rather than one dramatic wipeout. Traders often find this harder to notice, and therefore run it longer.

The other trap is that recovery is slow. After a long losing run the stake is elevated, and it takes an equally long winning run to bring it back down — during which you are risking more per trade than you intended.

Profit and loss thresholds

Both thresholds stop the bot automatically. Set a profit threshold of 100 USD and the bot stops once total profit across all trades passes it; the loss threshold does the same on the way down. Use both.

Run it in Binary Pro

1. Open the Bot Builder and choose Quick strategy > D'Alembert.
2. Select the market and trade type, then set your initial stake, unit and thresholds.
3. Hit Run to load the blocks, and adjust anything you like before starting.
4. Or download the ready-made version — "Apex Switcher" on the free bots page.

Related: Reverse D'Alembert flips the rule to raise the stake after wins. Oscar's Grind is another slow-progression approach with a defined profit target per session. Model your settings first with the backtester.

Illustrative figures are rounded. A 1 USD stake does not return exactly 1 USD on a successful trade — payouts vary by contract and market.