Basket-Grid / Martingale Bots
Grid of averaging-down orders with a money-target basket close — the classic 'smooth equity curve' grid bot.
Blew the $5k account to the ruin floor in ~15 months; a zero-cost control pins the basket at PF 1.01 — a coin flip with no directional edge.
- Why it fails
- The spec fully defines the risk machinery but never an entry edge — risk management cannot manufacture alpha, and real costs bleed a zero-edge bet to ruin.
- When / how it stopped
- Run spec-exact on real 2023–2026 M15 data, the bot posts PF 0.514, a −90.26% drawdown, and was negative every month.
The basket-grid martingale bot is the classic “smooth equity curve” promise: stack a grid of averaging-down orders, close the whole basket at a money target, and let historical retracements bail you out. The equity curve looks beautiful — until it doesn’t.
Run exactly as specified over 3.4 years of real M15 data, it does not hit its 5–7%/mo target — it loses and blows the account:
- From a $5,000 start it bled to $490 (−90.20%) and hit the ruin floor on 2024-04-02, ~15 months in, then sat flat (dead).
- Across 1,276 baskets: win rate 37.5%, profit factor 0.514, monthly Sharpe −5.56, −90.26% drawdown, and 0 of 16 months profitable (average month −13.1%).
The kill-switches, recovery throttle, two-strike override and regime veto all fired as designed. They slowed the bleed; none made the bot profitable.
The decisive test is the zero-cost control. Strip spread, slippage and swap and the basket sits at profit factor 1.01 / ~54% win — a coin flip with no directional edge. Removing the martingale grid on top changes almost nothing. The entire live loss is transaction cost applied to a zero-edge bet, compounded over thousands of baskets at ~$3–5/basket. The spec fully defines the risk machinery but never an entry edge — and risk machinery cannot manufacture alpha.
Worse, every backtest bias here points toward optimism (survivorship, fixed-pip slippage, constant spread, a soft ruin floor), so a live account would do everything this run does, only worse.
The full report, cost decomposition and equity charts are in our validation write-up.
→ Read our full validation report: /strategy/basket-grid
Sources
Frequently asked
Do grid trading bots work?
Not on a real cost model. Run spec-exact on 2023–2026 M15 data, this basket-grid bot lost money every way it was measured — profit factor 0.514, −90% drawdown, account blown to the ruin floor in about 15 months, and 0 of 16 months profitable. With transaction costs removed it sits at profit factor 1.01 (a coin flip), so there is no directional edge for the grid machinery to compound.
Is martingale trading profitable in the long run?
Not here, and the reason is structural. The spec fully defines the risk machinery — kill switches, recovery throttle, two-strike override, regime veto — but never defines a directional entry rule. Risk management cannot manufacture alpha; once real spread and slippage hit a zero-edge bet across thousands of baskets, expectancy is reliably negative at roughly $3–5 lost per basket. The safety net slows the fall, it does not turn a losing strategy into a winning one.
Not investment advice — your mileage may vary, but the burden of proof is on the person claiming an edge. This entry describes general research and published evidence (or its absence), not a recommendation. See the full disclaimer.