Research and strategy testing

Trading strategies, tested before they're trusted.

Tireon Capital is a quantitative research lab. Submit a strategy and our AI agents replicate it exactly, backtest it with out-of-sample discipline and real costs, and tell you whether it holds up. We publish the negative results too.

Submit a strategySee how we test

What we do

{ }

Strategy evaluation

Pine, Python, or plain rules. Replicated as written, never tuned.

≈

Rigorous backtesting

Lookahead-safe, cost-aware, IS/OOS split, bootstrap significance, checked by exact timeframe.

✓

A clear verdict

Fund, forward log, or don't fund, with the numbers and the bugs we found.

Latest evaluation

Forward log Submitted strategy Replication

Opening gap fade with relative-volume filter

Fade gaps > 0.8× 20-day ATR when opening volume < 0.7× average · SPY, QQQ, IWM · daily · Sep 2023 – Sep 2026

3 Sep 2026
Return, 3y (SPY)
+61.2%
buy and hold +58.4%
Max drawdown
−10.4%
buy and hold −17.2%
Sharpe
1.28
buy and hold 1.19
Win rate
63%
94 trades
Profit factor
1.47
net of 1 bp round trip
Beats buy and hold?Marginally on return, clearly on drawdown. 3 of 3 instruments.
StabilityPositive all 3 years. Bootstrap P(net > 0) 96.4%. OOS Sharpe 1.19 vs IS 1.40.
CaveatRoughly 40% of profit came from April–May 2025. Sample is 94 trades in one dominant regime.
Full report: per-instrument tables, year-by-year, exit-rule sensitivity, cost ladder, 10-year extension Read full report
Do not fund28 Aug

Dual moving-average crossover with RSI sleeve

SPY, QQQ, ES, NQ · daily · 12y

Return vs buy and hold+94% / +186%
Max drawdown−27.5% / −31.2%
Sharpe · PF · Win0.58 · 1.14 · 68%

Trails buy and hold in every window tested. The regime filter reacts too slowly to cushion a fast drawdown, and the exit rule can strand a position when the regime flips.

Fund with limits21 Aug

Overnight hold with contango filter

ES, NQ · daily · 5y

Return vs buy and hold+71% / +64%
Max drawdown−9.6% / −25.2%
Sharpe · PF · Win1.08 · 1.38 · 58%

Holds across all 5 years. Filter removes 22% of days. Sized for a 2% daily loss cap.

Browse all evaluations

Illustrative only. The evaluations shown are examples of report format and are not results of actual submitted strategies. Not investment advice.

How we test

Replicate as written
Apply real costs
Split in-sample / out-of-sample
Bootstrap significance
Check by exact timeframe
Compare to buy and hold
Verdict

No parameter search. No window selection. Negative results published.

Submit a strategy

T
Paste your strategy or describe the rules. Pine, Python, or plain English all work. I'll ask what I need to replicate it exactly.
U
Here's a Pine script. Fades opening gaps on SPY when volume is light.
//@version=6 strategy("Gap fade", overlay=true) atrLen = input.int(20) gapMult = input.float(0.8) …
T
Got it. Three things before I run it:
1. Instruments: SPY only, or SPY, QQQ, IWM?
2. Bar timeframe: daily, or does the gap need intraday bars to fill?
3. Should I treat process_orders_on_close as fill-at-close of the signal bar? That's how Pine does it.
U
All three. Daily. Yes.
T
Running: SPY, QQQ, IWM · daily · 3- and 10-year windows · 1 bp round trip · versus buy and hold. No parameters changed. I'll flag anything in the code that doesn't do what it looks like it does.
ReplicatedBacktesting 2 of 3BootstrapReport
Paste code or describe your rules
AttachSend

Illustrative conversation. Live chat requires an account.

How it worksThe agent reads your code, asks only what it can't infer, confirms its interpretation, then runs the full pipeline. Report lands in the thread and by email.
What we don't doOptimize parameters, pick windows, or soften a bad result.
ConfidentialityYour code stays private. Publication only with consent, anonymized.
TurnaroundMost daily-bar strategies finish in the session. Intraday and options take longer.

Who it's for

Self-directed traders

Want an honest answer before risking capital.

Strategy developers

Want an independent replication of their work.

Quant-curious investors

Want to see how testing is actually done.

Hedge AI Labs

Our product for doing the work yourself. Build a strategy in plain language or code, test it on real market data with the same lookahead-safe, cost-aware pipeline we use for evaluations, then turn it into a live scanner for US stocks, ETFs and futures. Ask the AI analyst a market question. Paper-trade it in a simulated account with realistic costs.

{ }

Build and test

Strategies from rules or code, backtested on real data. Negative results are shown, not hidden.

≈

Scan live

Turn a tested strategy into a custom scanner across US stocks, ETFs and futures.

✓

Simulate

Simulated accounts with realistic costs, so the first time a rule meets the market it isn't your money.

Visit hedgeailabs.comAsk a question

Hedge AI Labs is a product of Tireon Capital LLC. Educational tools, not investment advice.

Request access

Invite-only. Members get the evaluation chat, full reports, and new evaluations as published.