Problems with analytics

After the recent defeat of Houston Rockets, there have been a lot (more) debates on whether analytics works – which as most of us basketball fans know, Houston Rockets is absolutely religious when it comes to analytics.

This is what I think – both sides are not totally right.

First, there is nothing wrong with analytics. Analytics is about looking at things that happened in the past, “after the fact”. So by definition, it can not be wrong because these things have taken place! Sure, analytics is more about the patterns identified amongst these historical facts. But again the patterns themselves cannot be wrong because they are there! If there is anything wrong, it is more about the interpretations and application of these patterns.

And here are the 3 problems with (the interpretations and application of) analytics:

1. Just because historical data says it is stupid to shoot from 30 feet out (and it’s true), it does not prevent the birth of transcendent players. If everyone listens to analytics, there will not be Steph Curry (really shouldn’t be shooting way out there) or Kevin Durant (tall players are not good at playing outside). So analytics is good if you want to repeat what has happened in the past, but not very good if you want to aspire to be something out of the ordinary.

2. Players’ brain doesn’t work like supercomputers with state of the art algorithms. There are simply too many scenarios in every second of the game (if we take into account fatigue, etc, then every situation is unique) and players can’t react on a real-time basis. Surely analytics supports 3 pointers (36%) rather than 2 points (46%) because simple it is simple maths. But this is really only a statistical representation that reflects differently in different situations. For example, if you are 3 on 1, in the NBA finals game seven with 2 seconds to go and you are down 1, it is stupid to take the 3.

3. Finally, any analytics, or statistics, is only true about a very large population. But we play the game, or life for that matter, only once. When we take a shot at that moment, it is not about 36.25%, but it is about whether it goes in or not. It’s binary, it’s 1 or 0. That’s why winners win because they make those shots. And winners, again by definition, are not the general population.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s