Friday, October 19, 2007

Contradictions in an Article

I don't normally take on articles directly but there is an article on HoopsWorld that needs to be taken to task.

Making a list of the NBA's Biggest Contradictions is something that is a fair conversation and it could even be argued that last season Pau's performance didn't correlate to wins as well. However, without taking a look at anything but one season's performance and using such metrics as +/- rating as the sole determinant is a weak argument at best.

Why is +/- a weak determinant you ask? A fair question. Take this into consideration. While Gasol had a negative +/- ratio for the entire season he also was a member of the best 5 man units for the season. A curious dichotomy don't you think? How could Gasol be on the court when the team performed its best and still have a negative +/-? Could it be that Gasol was also on the court when their poorer performing units were on the court? Gasol also played the entire 2nd half of the season when many of the starters took the end of the season off. Could it also be that when he was playing himself into shape his game wasn't as good as when he got healthy?

Consider also that while Gasol had a negative +/- ratio his minutes occurred when the game was in contention against the opposing teams starters. Gasol didn't play against the backups very often. Look at the team's numbers. Do people really believe that Alexander Johnson (who has a positive 5.5 ratio) made the team better when he was on the court but Gasol made his teammates worse? I hope that isn't a conclusion many people would conclude but that is what you are being asked to believe solely using the +/- ratio.

When we use the Wages of Wins ratio instead we find that Gasol produced more wins than any other player on the team last season. Gasol's performance produced 11.5 wins or more than half of the wins the team actually produced last season. What's more the Win Score produced by Wages of Wins actually correlates to the teams actual wins in the season. No such attempt is made with the +/- ratio.

But that doesn't alter the fact that the team was outscored by a greater amount when Gasol was on the court than when he was off the court. Of course he was on the court for a greater percentage of the game than he was off the court as well. Gasol was on the court for 53% of the entire season for the Grizzlies and the team was outscored by 5.3 points during that time. When Gasol wasn't on the court the team was only outscored by 4.8 pts. Of course those 22 games were coached by Mike Fratello who coached a slower game and therefore fewer points were scored by the team. Couldn't that have a greater impact on the +/- rating than saying Gasol doesn't make his teammates better when he is playing?

In fact if you look at the actual numbers when Gasol was on the court vs off the court you will find that the team scored 4.8 more points when he was on the court while only surrendering 4.3 points. That is a positive .5 scoring production when Gasol was on the court vs when he was off it in games he played in.

But that isn't the big problem. The big problem I have with this is that the author put Gasol as the #1 biggest contradiction in the league despite his having a higher PER of any other player on the list and a lower +/- deficiency than anyone else on the list. Why would Pau be singled out as the biggest contradiction when players on winning teams such Allen Iverson (-0.9), Mehmet Okur (-6.6) and Deron Williams (-1.3)who had far worse +/- ratios aren't? Surely if a player had a negative rating on a winning team and was named to the All-Star game like Mehmet Okur would be a larger contradiction than Pau who's team performed poorly no matter who was on the court.

So with Pau we have a player who performed well against starters on a bad team and he is considered the biggest contradiction? Seems more likely the author had a personal bias in rating the players. I am not saying Gasol leads his team to victories but his performance last season surely isn't the largest contradiction in the league.

Especially when you consider that in the 2005-06 season when Gasol played the full year and was healthy his rating was a +4.87.


Anonymous said...

I had posted this on the Griz messageboard a few hours earlier last night, but it sure wasn't because I agreed with it. Thanks for the good comments about it.


ChipC3 said...

And I didn't reference the blog on the Grizzlies message board either but I do hope some people read a rational and statistical response to the article.

Spartacus said...

Chip, with the limited scope of reasoning that the original author used in determining who was a "contradiction", this was like shooting fish in a barrel. I'm amazed that you took it easy on him and left him with any dignity at all.

In fact, many of the players he listed are very productive in spite of their questionable +/- rating. This seems like an article by someone eager to get in on the statistical side of basketball, even if they don't fully understand them. Maybe it is one of John Hollinger's former interns then???

Ryan Schwan said...

Oh yeah? Well Pau is clearly a contradiction. He's Spanish, and Basketball was invented in America. Therefore, he's clearly not a good basketball player. Take that!

Oh, and Tim Duncan? Manu Ginobili? Virgin Islander. Argentine. No wonder they didn't win the Finals MVP or regular season MVP. Good Ol Tony Parker and Dirk Nowitzki, solid Americans.

Am I right?

ChipC3 said...

You're right about everything but basketball being an American game. It was invented by Dr. Naismtih who was Canadian.