Guides
How 2manyoptions measures a backtest
Backtests are history, not prophecy.
This page explains what the numbers are built from and where they stop. Terms are defined in the glossary.
Last reviewed: 2026-10-15
SPX, 0DTE and option legs
SPX is the underlying index used in these studies. 0DTE means the options expire on the session being studied. A leg is one option position; a spread combines a short and a long leg.
Strike is the option exercise level. A short leg is sold and a long leg is bought. An Iron Condor entry combines a put spread and a call spread; the width is the strike distance within each spread.
Bid, ask, mid and slippage
Bid and ask are the quoted buying and selling sides of the market. Mid is their midpoint. A quote is an observation, not proof that an order filled there.
Slippage represents a modeled execution cost or difference from a reference price. The execution profile determines how quotes, triggers and fills are used; broker commission is a separate cost.
VIX, GEX, SMA and FOMC
VIX describes an option-implied volatility measure. GEX refers to gamma exposure context. SMA is a simple moving average. FOMC identifies scheduled Federal Reserve policy decisions.
These names do not by themselves specify a usable signal. Read the observation time, lookback and source in the relevant view; an SMA of entry P&L is not an SMA of SPX.
From selection to outcome
A modeled result follows a sequence: select strikes, attempt entry, evaluate exit conditions, close or settle at expiration, then apply the selected costs.
A quote is not a broker execution. Entry failure, an exit that cannot be confirmed and settlement are distinct outcomes. Check the execution profile and coverage rather than assuming every scheduled entry became a complete trade.
Compare like with like
Keep configuration, execution profile, costs, date range and filters aligned when you want to isolate one difference.
Periods in Multi Timeframe overlap: agreement between them is useful context, not independent replication. Differences in sample size and excluded outcomes can also change a metric.
Selection bias and small samples
Exploring many combinations makes it easy to find a strong historical result by chance. A small or repeatedly filtered sample can make that result look more stable than it is.
Treat the best-ranked entry as a candidate for further examination. Review neighboring parameters, other periods and sample coverage; none of these checks guarantees future performance.
Iron Condor (IC)
An iron condor entered at several fixed times of day (e.g. 9:33, 9:45, 10:00) on SPX 0DTE options with a target premium and width. Each entry collects premium and profits if SPX stays within the defined range through expiration or until the profit target is hit.
Iron Condor is the public strategy in 2manyoptions. Backtested history starts 2023-01-02. Each entry time is studied on its own, so the results cover both a single condor and several condors opened at fixed times through the session — each managed by its own stop and target. The multiple-entry variant is known in the community as MEIC (Multiple Entry Iron Condor).
Data Resolution (100 ms)
The engine evaluates fills, stops and targets against the historical SPX option quote record at 100-millisecond resolution — not minute bars, not end-of-day snapshots, not interpolation.
Historical option quotes are sampled at 100 ms. Resolution does not guarantee complete coverage or executable fills. Trades without a confirmed outcome are excluded individually where supported, with coverage shown alongside the study.
Pure Measurement
No correction or adjustment factors are applied to P&L — the numbers are pure measurement of the modeled fills against the historical option chain; invariants and internal tests guard that measurement.
Earlier releases documented per-stop adjustment factors from a minute-data era; those were retired when the engine moved to 100 ms data and the correction layer was removed. Results are validated internally — determinism tests, invariant checks, data-quality audits — rather than against any external benchmark.