FSP Lab
Log inStart free
Backtest results

The curve is the easy part.

This is one real strategy run end-to-end through the full honesty suite, the same report you get for your own. Anyone can show you a rising equity line. The verdict beneath it is what tells you whether it is a real edge or just a story the data told itself.

Run your own backtest

The interactive builder + runner is in early access.

Early access

No code yet? You can still explore the strategy builder and join the waitlist there.

Example run

opening_range_breakout

ES · NQ · Stocks · 5m

Hypothetical result
Verdict

Did not pass review

Low confidence, read with care

Didn't hold up under review. Not running it.

Sharpe ratio
1.73

Return earned per unit of risk taken. Higher means smoother gains.

Max drawdown
6.0%

The deepest peak-to-valley drop. Smaller is easier to live through.

Win rate
62%

How often a trade finishes in profit. Not the whole story on its own.

High, but what matters is size: winners average 0.96R, losers -1.03R.

Avg R / winner
0.96R

Average reward on a winning trade, in multiples of the risk staked.

Trades / week
1.4

How often the strategy actually trades. Too few and results are noisy.

Walk-forward stability
83%

Share of rolling time windows where the edge still showed up.

Deflated Sharpe
0.55

Confidence the edge survives after counting every variation tried.

Equity vs. 400 resamples

Re-drawing your trades with replacement, to show how much the result depends on the specific trades you got.

Re-drawing the trades 400 times, 100% of runs finished in profit. The middle outcome was +33.5%, the unluckiest 5% landed at +9.1%, and the risk of a 20%+ drawdown was 0.0%. Trades are resampled with replacement, so this varies the mix you might get, not just the order.

ES1!
Long entryShort entryExit

Candlestick price chart for ES1! on the 5-minute timeframe, showing 1489 bars with 148 trade entries marked by up arrows below the bar (green for long, red for short) and exits marked by down arrows above the bar. A visual aid; the full trade list and performance figures are in the breakdown and quant-detail sections.

Tested on data it never saw

We trained on 70% of the history and kept the last 30% hidden, then measured the edge on that untouched slice. A strategy that only works on the data it was built from is overfit. This is the test that catches it.

In-sample (the data it learned from)70% of history
Out-of-sample (held-out, the honest test)30% it never saw

The edge held up on data it never saw, the strongest sign it's real, not fitted.

Walk-forward stability

83%

Re-fitting the strategy as it rolled forward through time, it stayed profitable in 5 of 6 independent windows. It was inconsistent across periods; the edge comes and goes.

Where the edge lives

Expectancy is the average result per trade, measured in R (multiples of the amount risked). Bars right of the center line made money in that bucket; bars to the left lost. The same strategy can be sharp in one regime and flat in another. This is where you see it.

By market regime

  • Uptrends
    51 trades
    +0.96R
    avg / winner

    Strong in Uptrends (+0.5R avg over 51 trades, 78% won).

  • Volatile markets
    57 trades
    +0.95R
    avg / winner

    Mildly positive in Volatile markets (+0.2R avg over 57 trades).

  • Ranges
    20 trades
    +0.97R
    avg / winner

    Soft in Ranges (-0.1R avg over 20 trades).

  • Downtrends
    20 trades
    +0.97R
    avg / winner

    Soft in Downtrends (-0.3R avg over 20 trades).

By session

  • NY afternoon
    17 trades
    +0.96R
    avg / winner

    Mildly positive in NY afternoon (+0.5R avg over 17 trades).

  • NY morning
    131 trades
    +0.96R
    avg / winner

    Mildly positive in NY morning (+0.2R avg over 131 trades).

By symbol

  • ES1!
    148 trades
    +0.96R
    avg / winner

    Mildly positive in ES1! (+0.2R avg over 148 trades).

Risk-on vs risk-off

  • Risk-on
    99 trades
    +0.96R
    avg / winner

    Mildly positive in Risk-on (+0.3R avg over 99 trades).

  • Risk-off
    33 trades
    +0.97R
    avg / winner

    Mildly positive in Risk-off (+0.1R avg over 33 trades).

  • Neutral risk
    16 trades
    +0.82R
    avg / winner

    Soft in Neutral risk (-0.3R avg over 16 trades).

Positive expectancy (made money)Negative expectancy (lost money)Bars scaled within each section.

Would this have passed a funded-account challenge?

Scored against typical prop-firm rules: a profit target to clear without breaching a max daily loss or a max total drawdown. Evaluated on this in-sample backtest (the same equity curve above), so treat it as an honest indication, not a live guarantee.

Funded-account check
PASS
  • Max drawdown 6.0%, within 10%
  • Worst intraday drawdown −1.5%, within 5%
  • Hit +10% target on day 31
Hypothetical performance disclosure (CFTC 4.41)

HYPOTHETICAL PERFORMANCE RESULTS HAVE MANY INHERENT LIMITATIONS, SOME OF WHICH ARE DESCRIBED BELOW. No representation is being made that any account will or is likely to achieve profits or losses similar to those shown. In fact, there are frequently sharp differences between hypothetical performance results and the actual results subsequently achieved by any particular trading program. … Simulated trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to those shown.

These figures are simulated, not the result of live trading. No representation is being made that any account will or is likely to achieve profits or losses similar to those shown.