Grid and martingale: the curves that lie
These two produce the prettiest equity curves on the internet, and both have genuinely negative expectancy. Not because the arithmetic is wrong, but because the curve measures something other than what it appears to. Understanding why is the cheapest vaccine there is against losing a whole account at once.
By the TradingCalculator.Pro team · Updated on · About us
Start your 7-day free trial →What you will learn
What a grid is
A grid of orders at fixed intervals: price drops a step, you buy more; it rises, you close for a small profit. No prediction and no stop. As long as price oscillates inside a range, the machine collects pennies and almost never misses.
Why the curve only goes up
Because it measures only what is closed, and the system never closes what is losing. Red positions stay floating, open, outside the statistics; the win rate climbs above 90 % counting only the ones that did close. The curve is real, and it is measuring half of reality.
Doubling after a loss
Martingale doubles the size after every loss, hoping one win recovers everything. Exposure grows 1+2+4+8+16… With a €100 base on a €10,000 account, seven losses in a row demand €12,700 — more than exists. And seven in a row is not bad luck, it is statistics: at a 50 % win rate it happens at some point in 200 trades more likely than not.
Same maths, both endings
In a range, averaging down is exactly right: price comes back and every added step pays. In a sustained trend it is exactly wrong, and no parameter tells the two apart in advance. The same machine that produces the unbeatable curve sideways produces near-certain ruin in a trend; these are not two strategies, they are one with two regimes.
How to spot them in an advert
Four signs, and one is enough to be suspicious: an equity curve that rises with no visible drawdown; an advertised win rate above 90 %; a short track record — weeks or a few months; and position size that GROWS while the account falls. The fourth is decisive: no sound system raises risk while it is losing.
Anti-martingale is defensible
Raising size after a WIN, not after a loss, is the exact opposite and it is the basis of pyramiding in trend following. The difference is not stylistic: it risks gains already made instead of base capital, so a bad run shrinks exposure rather than exploding it. Same gesture reversed, and the sign changes everything.
What to do with this
Rule out martingale and grid as income strategies: their real expectancy, once the tail is counted, is negative. If a grid is traded anyway, treat it as what it is — a bet that no trend will happen — with a hard aggregate loss limit, not a per-position stop the grid is designed never to use.