Where it stands
- IdeaRule written down
- BacktestedRun after costs
- Last-year pickSettings picked in-sample
- Forward testingLogged on new data, rules frozen
The rule
This is the published rule from Zarattini, Aziz and Barbon (2024), tested exactly as written with nothing tuned:
- For each minute of the New York session, measure the average absolute move from the open over the previous 14 sessions. That builds a 'noise area' band around the open.
- Every half hour from 10:00 to 15:30 New York time, be long if price is above the band and above session VWAP, short if below both, and flat otherwise.
- Flat by 22:00 GMT+3 at the latest. One micro contract, no leverage scaling.
The frozen test
Rules were written and hashed before the run. It was run once on July 2023 to October 2026, a window never used for this rule.
| Measure | MNQ |
|---|---|
| Trades | 590 |
| Net after costs | +$9,884 |
| Net per trade | +$16.75 |
| Win rate | 42% |
| Max drawdown (closed trades) | $2,615 |
| Per trade at 2x slippage | +$15.75 |
Why it needs forward testing
- Statistics. It sits just short of the desk's strict multiple-testing bar (p = 0.0135 against 0.0125).
- Concentration. A small number of days carried most of the profit.
- Drawdown. $2,615 for one micro, above the desk's $2,000 limit.
- Seasonality. Results differ between summer and winter, because the fixed GMT+3 exit moves against New York time.
Status: in forward testing. This is research, not a recommendation and not a signal. No result here is a promise.
What happens next
The frozen rule is logged on new data from 22 October 2026, with no changes. A handful of weeks proves nothing either way: at roughly 180 trades a year, one year of new data alone cannot settle it. The plan is to pool new data with the frozen window and re-judge it under the same pre-registered rules.
All dollar figures are hypothetical backtest results for one contract, after exchange and broker fees and one tick of slippage on each side of every market or stop order. They describe the past under stated assumptions and do not predict future results.
Research notes for education only. Not financial advice, not a recommendation and not a signal. No strategy on this site has passed final validation. Futures trading involves substantial risk of loss.