Free NT8 strategies for MES, MNQ and MYM (developer here)

General Discussions
1

Hi all, I’m the developer of Theory of Edge. We’ve made three of our NinjaTrader 8 strategies free:

  • TheoryOfEdge_MES: short-RSI mean reversion with a VWAP side and trend filter, break-even stop, flat by 11:05 CT.
  • TheoryOfEdge_MNQ: late-session continuation, ATR trailing stop, flat by 14:30 CT.
  • TheoryOfEdge_MYM: long-only morning noise-area breakout, ATR trail and RSI exit, flat by 13:20 CT.

Install with Tools > Import > NinjaScript Add-On (keep the zip as is). 1-minute chart or Strategy Analyzer, ETH trading hours template, 60 days of history minimum.

They’re on the free algos page of the Theory of Edge website (new member here, so I can’t post links yet; the download link is sent by e-mail).

Results are hypothetical; please run them on sim first. Questions and bug reports welcome here.

2

I remember when Neural Networks were all the rage a few years ago.

There were more vendors than I can remember, “Sigma Research”, “SirTrade’s Safir-X”, “Bowfort Technologies”, “Cornice Research”, “LouisMendelsohn’s ProfitTaker and VantagePoint”, “NeuroDimension”, “Ward System’s NeuroShell Trader”, “Nirvana System’s OmniTrader”… and So. Many. More!

I was really excited about a company called “BioComp Systems” and particularly the module, “BioComp Profit”. I used it and trained it and trained it, in sample and out of sample. After months and months, I found out I was a better trader with the neural network that resided inside my own head. Swapping between Tradestation/MultiCharts and Biocomp was mind numbing.

I had bought every trading book I could get my hands on that dealt with Neural Nets.

I remember talking to Murray Ruggiero, a trading math genius, for an hour at some conference over 20 years ago about data sets and error reduction cross market matrices.

Ah, The good old days. I don’t even know if most of these vendors are still in business.

Maybe this go around A.I. is different? Maybe some independent trader will discover a profitable A.I. that actually works long term. Of course you’ll never hear about it. Who would advertise they had the golden goose?

~RobotMan

1 Like
3

Thanks for sharing that, RobotMan, great bit of history.

I share the skepticism, and it’s a big reason this isn’t a neural net. The engine doesn’t invent anything: it starts from strategies that are documented in research (opening range breakout, late-day continuation, short-term mean reversion), only tunes a few settings, and keeps what still holds on data the search never saw. Fewer moving parts, less room to fool yourself.

And you’re right that nobody advertises a golden goose. That’s why these three are free: no performance claims, just something people can run in the Strategy Analyzer on their own data and judge for themselves. If you try one, I’d genuinely like to hear where it breaks.