Fast Thinking, Losing Streaks and Basketball

It’s that time of year: Playoff sports. NBA edition. Toronto is firmly past the first round, so things are starting to get as exciting here as they’ve ever been. It’s fun (for once) to have an excited city with a reason to stand behind a team. But that’s not what this is about.

Instead…

One of the cool things about basketball is that statistics, chance and the law of large numbers lend themselves so well to this sport. This is on my mind because I’m reading the wonderful book, Thinking Fast and Slow, recommended to me by a YouTube viewer.

Now, as an active investor and trader, I’m always doing my best to think in probabilities. But man, basketball really drives the point home.

Let me give you an example:

In the last Toronto vs. Miami showdown, the two star Raptors players scored 16 points combined in the entire game and were 6/23 from the field. Brutal, right? But get this…

Part way through the first quarter of the next game, these two stars combined to score 12 points and were 6/9 from the field (or something like that). They had dramatically outperformed and were suddenly hot again.

But were they?

The book Thinking Fast and Slow (review coming soon) shows different ways our brain processes information and creates illusions. One of these illusions is the “hot hand” in basketball. Science has categorically proven it does not exist.

And yet…

Announcers and fans alike rant and rave about basketball stars being hot and cold. But in a sport with so many games and so many shots per game (particularly by the star players/Kobe) the shots taken in a single quarter vs. a multi-season career are completely insignificant. The sample size is too small.

Let me show you where I’m going with this.

Below are two graphs of a random-number generator that feeds into a rule-based trading system.

Each set of graphs is a unique set of random numbers.

The graph on the left shows an equity curve over 3,500 trades, and the graph on the right shows a random 12 trade sub-set of the same equity curve.

Behold:

Screen Shot 2016-05-11 at 10.19.57 PM

This first curve is probably what you’d expect. But it’s not exactly typical.

The returns can still be smooth in the long-run and bumpy in the short-run…

Screen Shot 2016-05-11 at 10.18.55 PM

Below is another volatile sub-set. Remember, these are all just subsets of the bigger curve (which consistently looks great, don’t you think?)

But on a short term basis…

Screen Shot 2016-05-11 at 10.19.16 PM

It’s not always pretty. And sometimes…

It’s downright ugly:

Screen Shot 2016-05-11 at 10.18.36 PM
Losing streaks happen, in basketball and in trading and investing. The plain truth is, our brains aren’t well-wired to deal with stats like this. We quickly jump to conclusions based on key inputs.

But like it or not, stats like this are how the world works…

While the narrative appeal of a super-star on a hot-streak sounds nice, it’s probably more likely that they’re temporarily oscillating from one extreme to another and you’re exaggerating the contrast. You might also want to keep this in mind before you give up on your investing methodology in search of the next hot trading system.