MVPs

I'm a basketball nut. I never see enough basketball for my liking. I'm also not one of those college only basketball fans. I love watching the NBA. After games I like to look at the box scores to see what kind of picture the game stats paint. With that in mind, I have to say that I absolutely hate the way the MVP is determined in the NBA All Star Game.

I know the All Star Game is completely meaningless in the greater NBA picture (which I will concede is meaningless), but the NBA people could at least put in a little effort in determining the MVP. It may as well be a mathematical equation. At the very least it could be put into flowchart form, an extremely simple flow chart at that. It would go something like this:

1. Find winning team
2. Find player with most points
3. Give MVP trophy

That's about all there is to it. Sure points provide the final decision in who wins and loses, but it overestimates the value of a single aspect of how those points are scored. About the only exception to the above is when there is some story involved, but even in some of those cases the players will do their part to "force" that player into the most points.

It would be interesting to see how a composite points + 2x assists would change who received the MVP each year. It would be interesting to see rebounds, steals, shooting percentage, and possibly a few other stats taken into account. This year at least the MVP, Dwyane Wade, looks like he was pretty much the runaway winner even looking past just the points. He also led his team in assists, rebounds and steals.

The place where it falls apart, and seemingly flies in the face of logic, is in a fascinating stat called "+/-". Essentially this stat is a representation of the score while a particular player is on the court. If the good guys score, it is positive points. If the bad guys score, it is negative points. The result is that the good guys score that many more or less points than the opponent while that particular player is on the court. It isn't a "natural" stat in that it isn't directly measuring any specific facet of the players performance, but it gives a feeling of how well the team performed because of the player being in the game. It isn't perfect, but it is an interesting thing to look at.

If "+/-" was a factor in determining the MVP, it might have been LeBron James hoisting the trophy at the end of the game. Dwyane Wade was the only starter for the Eastern squad with a negative "+/-". Dwyane Wade was -2 on 31 minutes of play. LeBron James led the team with a +14 on 32 minutes of play. In a 48 minute game it is reasonable to assume that two players who were in the game for about 2/3 of the game would have about the same numbers in these stats. Instead there is a 16 point difference. LeBron James had fairly comparable stats in points, rebounds, and steals. His shooting numbers were not as hot, along with the assists, but for the most part the numbers were comparable. When it gets to the discrepancy in that one stat it really gets weird. Why do two guys with almost identical numbers have such wildly different "+/-" numbers?

Some of it can be explained by which people were on the floor with the player in question. Match ups away from the player in question could have played a big role in team offense and defense. It could be a whole lot of other things, but I still think it is an interesting line of reasoning.