Option Omega

Ever trade $XSP?

What happened to 5/7 Double Calendars in the past few years? Did our video on them in 2024 kill the trade?

11/ We explored this in today’s Fix My Backtest with Mark Anderson. Watch: youtube.com/watch?v=j6FHeYPD… The episode’s A/B comparison uses invented teaching data to explain the method. Historical backtests don’t guarantee future returns.

5/ Here’s where the Option Omega MCP gets interesting. Connect your AI assistant to the backtester and ask it to coordinate the work: Optimize on past data → freeze the selected settings → backtest the next quarter → retrieve the trades → repeat.

3/ Walk-forward analysis puts that question to work: • Choose settings using 12 months of history. • Freeze them. • Test on the next 3 months. • Move forward 3 months and repeat. Each selection gets judged on data that wasn’t used to make it.

1/ Your AI can find the best backtest settings. But if it picks them using the same data you use to judge them, what have you learned? Give it a harder assignment: “Pick on the past. Test on months those settings never saw.” Here’s how. 🧵

Today at 1:30 on our YouTube Channel

We asked AI to backtest a 0DTE SPX iron condor since 2023, fees and slippage included. 9:32 entry: +$14,212, but it fell 27% first. Add a 200% stop: profit drops to $4,787. Test 7 more entry times: 10:32 made $39,216 with a 12.8% max drawdown. Then it warned us the winner could be luck.

BREAKING: We just gave Claude and ChatGPT a real options backtester, and it's not a demo. It's the Option Omega MCP Server. Describe a trade in plain English. Your AI builds it, runs the backtest and walks you through the results. 0DTE iron condors, credit spreads, calendars. Test entry times. Dig through your worst trades. No code.

10 trades-- supposedly different (all various double calendars)... but in a crash, the risk was magnified, not diversified. Romeo on the moment the correlation matrix stopped meaning anything. Full interview with @Sheridanoptions linked below. youtu.be/Wbj3fy4zCb4?utm_sou…