Numbers Don’t Lie – Or Do They?
April 7, 2013; Portland, OR, USA; Portland Trail Blazers power forward LaMarcus Aldridge (12) grabs a rebound against the Dallas Mavericks in the first half at the Rose Garden. Mandatory Credit: Jaime Valdez-USA TODAY Sports
I love numbers. I stand by the fact that numbers don’t and can’t lie – whoever uses and discusses numbers may present them in a false or questionable manner, but the numbers themselves can never lie. Or can they?
Twenty-seven points is 27 points, sure. But are all rebounds actually rebounds? What about blocks and steals? None of these actions are as clearly defined as scoring a point. Where does the human factor come into the equation? Here enters the murky world of NBA stat-keeping. According to an old DeadSpin article that has started to circulate again, this world may be shadier than most fans would probably want to believe (the article is a great read).
The former statistician that is interviewed in the article talks about just how much leeway the stat keepers were given in determining what occurred in a particular game. He said that particular actions such as rebounds and assists were particularly susceptible to being manipulated, since often times neither of them can be a clearly defined action.
He also posited that this action was generally condoned, if not actually encouraged while he was working. Fans like bigger numbers. The NBA likes happy fans. Ten rebounds sound much better than nine. So when there is a scrum for the rebound and it’s not clear who was really responsible for corralling it, what’s the harm in crediting it to the team’s star rebounding machine to push him to a double double?
The issue becomes most apparent, though, when considering the differences in home and away stats. As you may have guessed even without reading the article, home statisticians tend to be friendlier to their own players. This could be merely chalked up to human nature, but the levels that the interviewed statistician describes become more concerning.
There is a certain beauty to the box score. It’s pure. There is no room for outside thoughts, subjective analysis, far-fetched narratives or deviation. What you see is what you get. This is part of what has always appealed to me about the numbers aspect of basketball – if I am arguing with a friend about two players, I can bring up that Player X shot 43% and Player Y shot 45%. That’s a cold hard fact. That single stat has no room to give. It is what it is, no more no less.
While arguing, we as humans can try to massage it, or manipulate it suit our needs. What if Player X faced double teams and had to chuck up bad shots at the end of the shot clock more often? What if Player Y played in a better offensive system? These are certainly all factors worth discussing, but that original basis, the number itself, can never be changed.
That’s why I find these allegations so disturbing. The cold, polished, objective box score may not be what it seems. If the former stat keeper’s allegations are all true, how can I (we) still trust the box score? There isn’t really an alternative, but that still seems like a poor reason to keep going with a flawed system. The problem may have gotten better recently, though, as the statistician had not worked in the NBA for a while. Additionally, the problem may be mitigated if, in fact, all teams are guilty of this manipulation to some degree. Again, that is no real reason to accept the issue, but it could still hold true.
But, to get off my soapbox, I decided to look at some of the Blazers from last season to see if there were any shocking discrepancies among their home and away splits.
Damian Lillard seemed unaffected. In fact, he recorded more rebounds and blocks per game during away games, with the same number of assists. His number of turnovers in away games was significantly higher though. As an aside, I love to see this – away games are always harder for a team, so to see a rookie perform so consistently well on the road bodes well for the future. Especially when a team reaches the playoffs, road performances can become crucial.
LaMarcus Aldridge recorded more assists, blocks and rebounds per game during away games, similar to Lillard. He did have noticeably more blocks at home, but other than that, his numbers were remarkably consistent.
Nicolas Batum was nearly identical to Aldridge. He also had more rebounds, assists and steals at home, but his blocks were nearly identical. Also like Aldridge, none of the differences were astronomical by any stretch. The only potential takeaway could be that his consistency, especially on the road could stand to be improved, but as many Blazer fans know, a lack of consistency is just what you get with Batum.
When it’s all said and done, there was nothing remotely suspicious about the Blazers’ home and away splits. Perhaps my faith in the box score was not so misplaced after all. At the least, the topic teaches you to always question where information comes from. As much as I tell myself that numbers are black and white, sometimes they just can’t be.