Although I have a chart of the SPY open in my trading platform during the day, I am only mildly curious about what it’s doing. Even that might be overstating it. When the market gaps up strongly I don’t have a directional bias in my daily gap trades and neither when the market gaps down.
I realize that it sounds crazy to some traders to completely ignore the overall market, especially when you consider the fact that the movement of most stocks are correlated with it. Given the fact that so many traders watch the overall market and use it to determine how they trade individual stocks, why in the world would I choose to ignore it?
Benefits to Ignoring Overall Market
I imagine most systems traders consider how to include the overall market as an input to their trading systems. I thought a lot about it as I designed the systems that I trade.
Here’s a list of reasons why I concluded that I should have no overall market inputs into my systems:
Is the Overall Market Really Highly Correlated with your System?
Although most stocks are correlated with the overall market on a given day, is your trading system highly correlated with it? It may seem appealing if it’s true, but in my opinion this should be a strike against the quality of your trading system. Why? What does it say about your trading system if it is so dependent on behavior in the overall market? To me it means your system really isn’t all that unique or statistically significant. A robust trading system SHOULDN’T be strongly correlated with the overall market.
A Market Input Reduces the Number of Trades in your System
If you find an overall market input that is correlated enough to make a difference in your trading system’s performance, then including it will necessarily reduce the number of trades in your system. In fact, there might be entire days or even weeks where your trading system won’t take trades specifically because of your market filter. I’m trying to take as many profitable trades as I can so anything that reduces the number of trades I need to look at with a very critical eye to decide if it’s really worth the trade off.
The Overall Market is Finicky
I’ve taken stabs over the years at developing systems to trade the overall market using the e-minis. What I quickly found is that when even with 20 years of 1 minute bars, the relatively profitable situations that repeat over time is fairly small. When you do find reliably repeatable opportunities they are relatively infrequent. If you loosen your standards a bit to find more frequent opportunities, the uniqueness of the situation is reduced. Another way to think about this is that each market day is unique — including such a finicky filter in a separate trading system seems risky and somewhat arbitrary to me.
It Feels Like a Shortcut
I’ve looked very closely at how the overall market affects some of my strategies. When determining filters to use, the ones that relate to the overall market usually correlate somewhat strongly. It’s always felt like a shortcut to use them though. I’ve always concluded that my systems should be able to stand on their own. If I can’t find other ways to improve a system I’m designing then I take that as a strong sign I should go back to the drawing board and think harder about what I’m trying to accomplish with the trading system.
It’s a LOT Less Stressful
Rather than including a lot of indicators and filters into your system, it’s always preferable to exclude as many as you can. There’s ALWAYS more data points you could be looking at — the trick is having the confidence to ignore the data that’s unimportant. That takes time and work to research how indicators may or may not affect your system. If you don’t do that research, you’ll always have some doubt in the back of your mind that you cluelessly ignoring something important. This is especially true for the overall market which everyone on Twitter has an opinion about. It feels great to be able to confidently ignore that chatter. This reminds me of one of my favorite quotes from Albert Einstein:
“Everything should be made as simple as possible, but not simpler.” — Albert Einstein
Even with all the reasons I’ve listed above, I haven’t given up on incorporating overall market filters. This is another benefit of having my systems ignore them — I can always introduce them in the future in a more intelligent way.
For example, to counteract the effect of reducing the number of trades, you could increase your position size since in aggregate each trade should be more profitable and you should be more comfortable scaling up size, all things being equal. I’m also working on a method that instead of a binary yes/no for if the overall market filter is satisfied or not I could use a confidence score from 0 to 100 based on the value of the current market filters. More on this general idea in a future post.