Avoid This Huge Mistake When Backtesting
By Dave Mabe
In today's newsletter, I want to address the biggest mistake I see traders make when they start backtesting.
Backtesting is a superpower - the sooner you start backtesting the sooner you'll start realizing the benefits.
Backtesting is a lot of work, but when done properly the payoff is immense.
The problem is most traders start backtesting in a way that still requires all that effort, but minimizes the actual benefits.
If you're going to put in the effort, it's important to do it in a way that will maximize the benefits, otherwise all your work will be wasted.
Most traders begin with an idea for a strategy they apply to a single symbol, usually a well-known market index like SPY or QQQ.
Creating a strategy that trades a single symbol like this can work, but it's vastly inferior to a strategy designed to trade any stock.
When you backtest using a single symbol, you're competing against every trader with the ability to simply look at a chart. That's almost every trader on the planet.
But when you start with a strategy designed to trade any stock in the market, the pool of traders you're competing against drops dramatically.
It's a hard technical problem to create software that backtests all of the thousands of symbols that actively trade each day.
Creating software to backtest a single symbol is trivial in comparison.
That's why you see so much software in the industry that claims to "backtest", but only a small handful can tackle the more lucrative problem of backtesting the entire market.
Why is it so much better to create a strategy for the entire market? Lots of reasons:
There are far more opportunities - it might take a decade to accumulate enough profitable trades in a single symbol to be worth trading
More opportunities give you more flexibility to improve your strategy
The types of strategies that work are easier to come by when applied to the entire market
Because of the technical issues of scanning the entire market, your pool of competing traders is much smaller
Easier to create a statistically valid strategy you can have confidence in
Your strategies will have more "staying power" - that is, they'll deteriorate less quickly