Research / Martingale grid
DE EN
Sign in
Research · Risk

Is a high win rate bot actually safe?

We ran a classic grid bot with Martingale lot doubling on 7 years of gold prices. It won 98.9 % of its 5,757 cycles. Here is what happened to the account.

QuantLab ResearchPublished 2026-11-12Data June 2019 – June 20268 min read

01How a grid bot works

A grid bot buys more every time price falls, and closes everything on a small bounce. It almost never closes at a loss.

A common way to build a bot with a very high win rate is a grid. The bot opens a trade. If price moves against it, it does not take the loss. It opens another trade further down, and another, and waits.

With Martingale, every new trade is twice as big as the one before. That pulls the average entry price close to the current price, so even a small bounce is enough to close the whole basket in profit.

Gold 5-minute chart with five grid buys and the basket target
Figure 1. A real cycle from the test: 5 buys, then a bounce of a few dollars closes all of them in profit.

That is why the win rate is so high. The problem is the other side of the same mechanic:

Lot size of each of the ten buys in a Martingale grid
Figure 2. With doubling, the last buys are huge. A cycle that goes all the way down holds 102.3 lots instead of 0.1.

02What we tested

One fixed grid bot on gold, set up in advance and never tuned — plus three variations for comparison.

MarketGold (XAUUSD), 5-minute candles, June 2019 – June 2026
StartA new cycle opens 0.1 lot at the open of the candle after the last cycle ended
Grid spacing3 × ATR (14 five-minute candles), fixed when the cycle starts
AddingOne more buy each time price falls one spacing further, up to 10 buys. Martingale: each new lot ×2
TargetClose the whole basket when price is one spacing above the average entry price
Stop-lossNo stop at all. Only the broker's stop-out at 50 % margin level (leverage 1:100)
Account$10,000. No new buy if margin level would fall below 100 %

When the account fell so low that it could not even open the first 0.1 lot safely, we counted it as blown and started again with a fresh $10,000.

We ran the same rules two more ways: Martingale short instead of long, and long with the same lot size on every buy (no doubling). As a contrast we added a simple system with a stop-loss (section 9).

03How a cycle is counted

One cycle is one trade: from the first buy until the whole basket is closed.

  • Win: the basket closed in profit after costs.
  • Loss: the broker closed it at the stop-out level, or it was still open and under water when the data ended.
  • Costs: every single order pays the real spread plus $0.50 per ounce slippage.
  • Worst case: if one candle touches both a new buy and the target, the buy (and a possible stop-out) comes first.

All parameters were fixed before the test. We still split the results into 2019–2023 and 2024–2026 to see whether the picture changes over time.

04Wins against losses

The bot won 98.9 % of the time and still lost money.

Total profit of winning cycles against total loss of losing cycles
Figure 3. Every winning cycle added together against every losing cycle added together, in US dollars.

5,691 winning cycles made $5,116,697. The 66 losing cycles lost $5,245,973. Net: −$129,276.

Typical and average win and loss sizes on a log scale
Figure 4. Note the log scale: each grid line is ten times the one below.
The win rate you would need

With an average win of $899 and an average loss of $79,484, a bot has to win 98.88 % of the time just to break even. This one won 98.85 %. A win rate of 98.9 % can be a losing system.

05The worst cycle

Gold fell 5.5 % in 11 days. That was enough to lose $821,523.

On 22 September 2023 the bot started a normal cycle with 0.1 lot. Gold kept sliding. The bot kept buying, doubling each time, until it held 102.3 lots — about 10,230 ounces of gold.

Worst grid cycle on a 1-hour gold chart
Figure 5. 10 buys, no stop. On 3 October 2023 the broker closed the whole position.

The account held $907,683 before this cycle and $86,161 after it: 90.5 % gone in one trade. That single loss equals 914 average winning cycles.

06The account

The account grew for months, then collapsed — again and again.

Account balance of the Martingale grid on a log scale with blown accounts marked
Figure 6. Account balance after every cycle. Red dots: account blown, restarted with $10,000.

This is what makes grid bots so convincing on a short track record. For long stretches the curve only goes up. The first account was blown on 11 November 2019, 5 months in.

Over the whole test, 14 accounts were blown. Counting every top-up back to $10,000, a trader would have paid in $141,943 and been left with $12,667.

Martingale longCyclesWin rateStop-outsNet USD
2019–20233,26199.6 %13 +$171,533
2024–20262,49697.9 %52 −$300,809

To be fair: measured in dollars, 2019–2023 even ended in profit, although both of the two largest losses happened then. A larger account lets the bot add more and bigger buys before the margin runs out, so wins and losses both grow with it. From 2024 on, stop-outs came far more often: 52 against 13.

07Without doubling, or short

Every version won at least 94.8 % of its cycles. Only the one without doubling survived.

Blown accounts for three grid variants
Figure 7. Same grid, three versions. The win rate barely differs; the number of blown accounts does.
Grid on goldCyclesWin rateBlownNet USD
Martingale ×2, long5,75798.9 %14 −$129,276
Martingale ×2, short4,11294.8 %55 −$495,771
Same lot size, long1,25099.9 %0 +$70,679

The grid without doubling ended with a profit of +$70,679 and was never stopped out. That result deserves a close look rather than a headline:

Account of the grid without doubling next to the gold price
Figure 8. The flat grid only ever bought. Gold went from about $1,328 to $4,014 in this period.
  • It only buys, in a market that went up about 3-fold. The short version of the same idea blew 55 accounts.
  • One cycle stayed open for 19 months (7 August 2020 to 8 March 2022) and still counted as a win.
  • The last cycle, opened on 29 January 2026, was still open at the end: −$140,058 on paper, 63.5 % of the account. At its worst point the account was 78.4 % below its high.

08The random test

Compared with random direction, the long Martingale grid did worse than most random versions.

We also ran each grid with the same rules but a random direction for every cycle and a randomly shifted start — 1,000 times for the main bot, 200 times for the others. The measure is the average result per cycle.

Grid on goldRandom runsBetter thanp-value
Martingale ×2, long1,0009.4 % of random runs0.9061
Martingale ×2, short20095.0 % of random runs0.0547
Same lot size, long200100.0 % of random runs0.005
How to read this

A p-value near 0 means random versions rarely did as well. The flat long grid "beats random" clearly — but its random versions buy and sell at random, while the real one only buys in a rising market. That is mostly the gold rally, not the grid.

09A system with a stop

A stop-loss caps the worst trade. It does not create an edge.

For contrast we tested a simple session system on the same gold data: trade in the direction of the first New York 5-minute candle (its close against a 12-period EMA), stop at 8 × ATR, trail the stop along a 120-period EMA once the trade is 0.5R in profit. One trade per day, same cost model.

Cumulative result of the first New York candle system on gold
Figure 9. Results in R (one R = the distance to the initial stop). Shaded: 2024 onwards.

It won 57.8 % of 1,397 trades and ended at −35.0R after costs (+8.9R before). Against 200 coin-flip versions it had a p-value of 0.4279: no edge. At 1 % risk per trade its worst drawdown was 44.1 %.

So it is not a good system either. But look at the shape of its risk:

Worst single loss measured in average wins for both systems
Figure 10. Worst single loss divided by the average winning trade.

Its worst trade was −1.82R, about 2.9 average wins. The grid's worst cycle wiped out 914. With a stop, a bad system loses slowly and visibly. Without one, it looks perfect until it isn't.

10Verdict

Not safe

A 98.9 % win rate came with 65 stop-outs, 14 blown accounts and a net loss of $129,276. The win rate hid the risk instead of measuring it.

A grid or Martingale bot does not remove losses. It postpones them and makes them bigger, until price moves far enough in one direction. On gold, that happened 65 times in 7 years.

Before you trust a high win rate

  1. Ask for the largest loss, not the win rate — and compare it with the average win.
  2. Ask whether there is a stop-loss. "No stop" usually means the broker's stop-out is the stop.
  3. Ask how long the track record is. A few good months prove little: here the first account was blown after 5 months.

Limitations

  • One market, one price path (June 2019 – June 2026), dominated by a strong gold rally.
  • One grid setup (3 × ATR spacing, 10 buys, ×2). Other settings were not tested.
  • Overnight swap was not included. Grid positions are often held for days or months, so real results would be worse.
  • Candle-based test: the order of prices inside one candle is unknown, so we assumed the worse order.
  • Margin and stop-out rules (1:100, 50 %) differ between brokers.
Is a high win rate bot actually safe? 60-second version · 231,491 views

The 60-second version

This study started as a Reel. Share it with someone who is looking at a bot with a near-perfect win rate.

Next in the research library

All studies →

Reproducibility. Script idee110_hohe_trefferquote_hohes_risiko_run95.py, commit 03a3d8f, data 2019-06-11 to 2026-06-29, split at 2024-01-01. The two example cycles were recomputed independently from the price data and match the test exactly. Past results do not guarantee future results. Trading leveraged products involves risk of loss.

What survived the test

Most ideas fail. These systems didn't.

Same process as every study: exact rules, data they had never seen, real costs, compared against random trades. These are the systems that passed — and they trade live today, with every trade visible.

Session Open Momentum NAS100 · M5 Free plan
live

One window a day — Nasdaq at the New York open

2020202120222023202420252026
+505.8 %Profit since start
56.4 %Win rate
1.24Profit factor

Max. drawdown −26.60 % Risk per trade 1.00 % Trades 1,792 · 4.4/week

Monthly returns outlined: forward test (live)
JFMAMJJASONDYear 19+7-70-6-4+5+8+4+2-4-6+6+3% 20+8+6-2+1+7+5+6+4+6+5-4-3+46% 21+50-1+16+12+6-8+8+6+1-5+2+50% 22+4+4+1+4+8-3-5+3-2-1+2+1+16% 23+2+6+4+30-3-7+3+4-3-8+3+3% 24-5-10+6+2+2-5+90-4-4+5+7+2% 25+12+40+80-6-4+14+7+15+9+6+82% 26-7+100+1+8-20+6+6+22%
Prop: suitable Free to use in the free plan

Free backtest file after signing in

UK100 3x Drop Monday Pullback UK100 · H1 Premium
live

Three strategies on the FTSE 100: post-drop rebound, Monday dip and pullbacks in the trend – with loss streak defense

20192020202120222023202420252026
+312.9 %Profit since start
50.8 %Win rate
1.27Profit factor

Max. drawdown −12.03 % Risk per trade 0.75 % Trades 1,908 · 3.8/week

Monthly returns outlined: forward test (live)
JFMAMJJASONDYear 17+1+2+4% 18-2+2-4+10+6+7-40+3+1+1+10% 19+4+4-2+1+1+6+2-3+30-4-3+12% 20+8-7+7-1-4+1+2+1-3+1+50+9% 21-1-2+600+2-10+1+50-1+8% 22+50+4-1-4-4+5+4+3+2+6-2+18% 23+7+80+8+2+1+2-4+50+7+6+52% 24-1-20+1+2+6-2-3+2+3+3+2+10% 25-1-1-2+4+12+3+11-4+40-7+4+23% 26-2+8-4+3-3+1+5+5+30+17%
Prop: suitable Premium subscription only

Free backtest file after signing in

XAUUSD 3x Speed Q4 Turn XAUUSD · H1 Free plan
live

Three gold strategies in one EA: momentum, a fourth-quarter breakout and the turn of the month

2020202120222023202420252026
+210.1 %Profit since start
44.8 %Win rate
1.26Profit factor

Max. drawdown −14.09 % Risk per trade 0.75 % Trades 1,309 · 3.2/week

Monthly returns outlined: forward test (live)
JFMAMJJASONDYear 190-400-3+7+1+5+3-1-3-2+3% 20+6+2-3+50-2+4-2-20+5+4+17% 21+2+5-4+2+4-1-1-5+3-5+6+1+7% 22+1+1+20+1-2+1+4-1-1+8+2+17% 23+5+1+1+2+4+3-2-20+400+16% 24-6-4+13+9+1-4+4-5+30-2-70% 25+6+5+1-2+1+4-20+8+13-3+6+42% 26+7-5+20+3+3+1+8+6-1+26%
Prop: suitable Free to use in the free plan

Free backtest file after signing in

NQBlade NAS100 · M3 Premium
live

EMA cross on NAS100, in four parts — with 12.5 months of forward testing.

20252026
+391.9 %Profit since start
40.1 %Win rate
1.35Profit factor

Max. drawdown −28.98 % Risk per trade 1.50 % Trades 1,528 · 2.8/week

Monthly returns outlined: forward test (live)
JFMAMJJASONDYear 24+6+4+8-9+11-4+12+16-7+11-8+3+48% 25+19+14+4+10-1+20-2+12+9+10+2+3+170% 26+70+1+20-8-9-4+17+23%
Prop: suitable Premium subscription only

Or join our free algo trading community on Telegram — talk to other traders and ask your questions.