How I Plan To Use Alternative Data
StatsEdgeTrading
If you’re testing stock strategies and not layering in external data, you’re missing edge. In this week’s Line Your Own Pockets, I get selfish — asking Dave about something I’ve wanted to test for a while: using implied volatility data from the options market to improve a simple mean reversion system.
Most traders backtest with just price and volume. That’s fine. But what if you could overlay the expected move from market makers (implied by options pricing) — and see what happens when price gaps outside that expected range?
The idea: when a stock reports earnings, the options market “prices in” a move. Most of the time, the actual move is smaller — that’s how option sellers make money. But when the price overshoots that expected move? That’s where I think mean reversion might kick in harder.
Here’s the twist: this whole concept hinges on pulling in external data and marrying it with your standard OHLC dataset. Sounds complex, but as Dave and I discussed, there’s a surprisingly simple way to do it — even without Python chops.
The process: run your backtest as normal, export results, and merge in the options-derived expected move data post hoc. You can do this with a lightweight script, and once set up, automate daily ingestion and archiving going forward.
Why bother? Because the harder the data is to use, the fewer people test it. Which means more edge for those who do.
Action Plan
Define your data column (e.g. implied move).
Build the baseline strategy first (e.g. gap fill rate).
Layer in the implied move and compare.
If it shows value — automate, archive, and scale.
Curious if it works? So am I. Stay tuned for results.
Want strategies with proven edge? Check out the free course, newsletter, and StatsEdge Pro access at www.statsedgetrading.com