You always hear TV announcers say it: “if you turn the ball over, you lose”. On an intuitive level, it makes sense — you don’t want to be giving the ball over to your opponent unnecessarily, duh. But it turns out…
- Alright, so what’s going on here?
Basically, I took last season’s turnovers per 100 possessions (for each team) and plotted it against the team’s winning percentage.
- So…what does this show?
It basically shows that there is zero correlation between turnovers (per 100 possessions) and winning percentage.
- Uh…so what does “zero correlation” mean exactly?
Basically, correlation is how related things are. For example, kids that roll around in the mud usually take a shower afterwards(hopefully). Thus, correlation between rolling around in the mud and taking showers is pretty high. It’s important to remember that correlation does not imply causation.
- So what’s surprising about the graph?
Like I said earlier, the graph shows zero correlation between turnovers and winning percentage. That means that there is no relationship between turnovers and winning percentage.
For example, the Oklahoma City Thunder had the second-best winning percentage in the NBA last year — and they averaged just as many turnovers as the New Orleans Pelicans. The Cleveland Cavaliers and the Charlotte Bobcats averaged fewer turnovers than the Miami Heat…and the Cavaliers and the Bobcats obviously had a much worse winning percentage than the Heat.
From what you hear from TV broadcasters (turnovers, turnovers, turnovers), you’d think that more turnovers should lead to less winning…but it doesn’t.
I don’t know. This graph can’t tell you that. My hypothesis is that in order to get good shots (for example, a post-up in the paint) you have to take some risks (like an entry pass through traffic), and these risks lead to turnovers. However, because teams get good looks, they score at a greater rate than a team that doesn’t take any risks and takes bad shots — and thus, they win.
You could have a team that takes contested jump shots all day, and not turn the ball over at all…or you could have a team that drives to the rim for a layup, but turns the ball over occasionally. Team B might have more turnovers, but because its offense is superior to Team A’s, it’ll still win more games.