The Wages of Wins Network current consists of the following five blogs:
Arturo Silly Little Stats
Courtside Analyst
Nerd Numbers The Blog
Roblog
The NBA Anti-Expert
In the past few days a number of excellent posts have been offered in these forums.
Although each of these posts is excellent, I wanted to highlight a recent column offered by Robbie O’Malley. The following post reviews John Hollinger’s comparison of Carlos Boozer and Al Jefferson. Hollinger employs his Player Efficiency Rating to evaluate the two players. Robbie discusses the problems with this approach. If you have not seen this discussion, here it is in its entirety:
In his analysis of the Utah Jazz trade for PF Al Jefferson ESPN’s John Hollinger begins by posing to his readers this question – Would you trade Carlos Boozer for Al Jefferson?
Hollinger comes to the conclusion that he would make such a trade.He says, “You’d at least have to think about it. Jefferson is three years younger and had better numbers over the past three seasons. He can play center — Boozer cannot — which allows the Jazz to paper over Mehmet Okur‘s expected absence and move Andrei Kirilenko up to power forward in stretches, where he’s more effective anyway.”
I would probably agree with his assessment on opportunities this trade provides the Jazz in their front court. Kirilenko is feasibly more effective at PF and Jefferson is more suited to play center than Boozer. The thing that caught my eye, however, is that he believes Jefferson had better “numbers” over the past three years than Carlos Boozer. I wonder what numbers he is referring to? He is known in the media as “Professor Hollinger” and is ESPN’s only resident stat geek. If anyone should know numbers it is him. I suspect the main number he is referring to is his Player Efficiency Rating (PER) metric. On average, over the past three years Carlos Boozer has a PER of 20.18. Al Jefferson’s PER over this time period is 1.5 points better at 21.68. So, according to Hollinger’s own PER, Jefferson has been the better player.
Is PER accurate? Let’s take a closer look at each players numbers:
Table One: Boozer vs. Jefferson
Table Two: Boozer vs. Jefferson relative to average PF
Carlos Boozer is 6’9″ 260 pounds while Al Jefferson is 6’10″ 260 pounds.They’re roughly the same size, although Jefferson has a longer standing reach than Boozer.They have played roughly the same amount of time at 65-70 games a year at ~34 minutes a game.They both have played similar minutes at PF and C. Bottom line is that this is as about as apples to apples a comparison you can make between two players in the NBA.
What do we find when we look closer at the numbers? In terms of net possessions (REB + STL – TO) they are virtually equal. They are both very good at generating extra possessions for their teams. Carlos Boozer is the better passer but Al Jefferson is the better shot blocker – essentially a wash there. Boozer commits about an additional foul each game. So far this is the biggest separation we have. I wouldn’t think that should account for a 1.5 points better PER, however.
So what else is different? When we look at the scoring categories we see that Jefferson takes nearly 3 more shots per game that Boozer. This allows him to score less than half a point more per game. That doesn’t seem like it’s in Jefferson’s favor though. A half point after three shots is not very efficient. When we look at the chart we see that Boozer is a good amount more efficient than Jefferson. This means that Boozer gets more out of the shots he takes than Jefferson does. Making the shots you take is important for winning games.
But essentially – according to John Hollinger – if Boozer took an additional 2.71 shots, making only .44 at a 16% rate, he would be as efficient as Al Jefferson. That just doesn’t work! Boozer is the more efficient player. There is no value in taking more shots if you don’t make those shots. It’s so strange that a metric that is named Player Efficiency Rating, does not value efficiency! If you shoot more shots your PER will be higher – even if you are not an efficient scorer.
I don’t get it. John Hollinger has the same numbers available to him that everyone else does. He also has more monetary incentive and available time than most others to use to get it right. He has underrated Boozer before while overrating the ability to take shots – HERE. You can see an explanation of why a players PER will increase with additional shots and nothing else – HERE.
Hollinger goes on to make a few other comments. He notes that Jefferson is three years younger and cheaper. He fits the center position better than Boozer. These points I agree with. He also said that Boozer is probably the better defender – backed up HERE and HERE. So Boozer has better numbers and is the better defender. Jefferson is younger, cheaper, and more versatile. A Boozer for Jefferson swap hurts the Jazz.
But it wasn’t exactly a trade. Carlos Boozer left via free agency. The Jazz did not control whether he returned or not. They basically added young PF Al Jefferson for nothing. This is a good move by them. Despite what I said, Jefferson can be a very good player and is better than the nothing Boozer left behind. He’s just not as good as Boozer, at least not yet.
