Houston Rockets: Four reasons why Moreyball analytics work for this team
By Neal Sharma
Reason #2: The misconception about the numbers
Many people believe that a number can tell the story better than anything. There are also, unfortunately, a number of people who believe that a number tells them all they need to know.
I’m calling B.S. already!
If there’s one thing I learned the year I took A.P. Statistics in high school, it’s this:
A statistic is never wrong.
A statistic is never right.
A statistic is not weak.
A statistic is not strong.
What the hell?
By statistic in the above statement, I refer to values of measurement such as points, rebounds, height, etc. Not the general field of statistics.
Here’s the best way to explain all of this mess so let’s try an understandable example as it relates to the Houston Rockets.
Somebody walks up to me and asks, “Who is the best shooter in the NBA?”
The first thing that pops up to my mind is field goal percentage. According to ESPN Stats, Clint Capela has the highest field goal percentage in the NBA.
Field Goal Percentage is used for the following reason…
Therefore my response would be, “Statistically speaking, I can conclude Clint Capela is the best shooter in the NBA because he has the highest field goal percentage.”
I mean, I wouldn’t be wrong. I have evidence to prove my response, as Clint Capela does have the highest field goal percentage. But if I hosted a shoot out with Clint Capela and Stephen Curry from the three-point line, I don’t think it would be hard to predict the winner. I wouldn’t be right either.
The percentage, or statistic, isn’t weak. Capela has taken 676 shots in the regular season, so the percentage is pretty darn consistent. But then again, 675 of those shots were inside the paint, implying that the shots were lay-ups or dunks. Maybe a rare mid-range shot. So the statistic isn’t strong as the shot variety isn’t large.
Hopefully the recent statement I said makes a little bit more sense.
If you are a NBA follower, you would immediately say something like, “Capela is a good player. But there is no damn way he is the best shooter in the NBA. Forget all this math crap you’re telling me. IF I put Capela in a shooting contest versus Stephen Curry, James Harden, Klay Thompson, and Kevin Durant, he would get whooped. $100 million my you know what.”
So the dude who asked the question in the first place responds with, “No..no. I meant like, if it were a shootout with all NBA players. If they just straight up shot the ball all around the court without any injury, mental doubts, etc, who would be the best one out there?”
That dude just changed the question. That is one of the most fundamental things to understand in the world of analytics. The best response possible only comes from a question that can provide the best sample data with it.
In simple terms, a crappy question will get a crappy response. Now the response, conclusion rather, may be formal and have plenty of evidence. But if the evidence is only based on the question that is possibly misinterpreted, then the response will be considered crappy by the guy who asked the question in the first place.
Welcome to the world of analytics, where a simple communication issue can screw up everything.
Seems as if the question is more important than the number itself…