# The best looking backtest I can find has no stop, and its profit factor is 1.20 — retained record

> Retained reddit record from reddit:algotrading.

Canonical: https://highsignal.app/data/records/04db75df7dd6db77

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:algotrading)  published 2026-11-19 04:13 ingested 2026-11-20 02:40

[original source ↗](https://www.reddit.com/r/algotrading/comments/1y004hd/the_best_looking_backtest_i_can_find_has_no_stop/)

## retained text



```
Two published runs, both NQ, both openable without an account. I put them side by side because one of them makes the other look broken, and it shouldnt.

First one is an opening range breakout on the 15 minute. Nov 2020 to Sep 2026, 1,493 trades, **+257.5% net**, 21.7% max drawdown, 53% win rate, **profit factor 1.18**, Sharpe 1.28.

Second one buys the last candle of the regular session and sells the first candle of the next one. No stop, no target, no filter, 100% of the account on every entry. May 2016 to Oct 2026, 2,604 trades, **+3,784.4% net**, 49.3% max drawdown, 56% win rate, **profit factor 1.20**.

So the second one posts a net return almost **15x larger**, and its gross profit per dollar of gross loss is two cents higher.

1.18 and 1.20 are the same rule quality for any practical purpose. Everything between +257% and +3,784% came from somewhere other than the rules: ten and a half years of window instead of five and a bit, a position held through every overnight session instead of being flat by the close, and the whole balance committed on every entry so the compounding runs on the full account.

Two people both earn 7% a year. One puts in 500 a month for six years, the other puts in his whole paycheque for eleven. The second finishes with a far bigger number and he is not the better investor, he just had more money exposed for longer.

The part i keep coming back to is that the no stop version is the one that looks best. A strategy with no stop cannot show you a stop out, so its worst trades are capped by wherever the next open happens to land, and the pain shows up in the equity curve rather than in the trade log. 49.3% on a hundred grand is a $49,300 hole sitting inside that 3,784%.

**\_\_\_\_\_**

The overnight gap thing is still a real result and id say so if it was my run getting picked apart. NQ has drifted up for a decade and a lot of that drift lands outside regular hours, 2,604 trades is not a small sample, and holding 1.20 across ten years is something i would want to look at properly. What it is not is a 3,784% edge, and sorting any list by net return puts it above the strategy with a quarter of the drawdown.

Both of these are on the [Agenticks](https://www.agenticks.ca/?utm_source=reddit&utm_medium=post&utm_campaign=autopilot&utm_content=algotrading) community feed with the full rule set and the trade log attached, so the arithmetic is checkable rather than something you take off me. The breakout run reports a Sharpe of 1.28 and the overnight run does not report one at all.
```



document fetched 2026-11-20 00:44 · canonical https://reddit.com/r/algotrading/comments/1y004hd/the_best_looking_backtest_i_can_find_has_no_stop
