I was thinking of waiting until Monday to comment on LeBron in the playoffs. Tomorrow is the decisive game seven between the Cavaliers and Pistons, so as I write this I do not know if LeBron and the Cavaliers have dethroned the defending Eastern Conference champs, or started looking for a good place to play golf this summer.
Although I do not know how the series will end, I thought I would still spend a bit of my Saturday commenting on the media’s coverage of LeBron.
After reading and watching the media’s accounts of LeBron’s exploits in the playoffs I am left with this impression of the conventional wisdom: LeBron has raised his level of play in the NBA’s post-season and consequently his team is on the verge of eliminating the team that finished the regular season with the best record.
As we note in the book, though, the conventional wisdom is often incorrect. And with respect to LeBron in the playoffs, it appears to be incorrect again.
Now before I talk about LeBron in the playoffs, I need to review what he did in the regular season. LeBron this past week was named to the All-NBA first team. He was joined by Dirk Nowitzki, Shaquille O’Neal, Kobe Bryant, and Steve Nash. Of these players, LeBron’s 20.4 Wins Produced in the regular season, led the field. Nash, Nowitzki, Bryant, and O’Neal finished with 18.6, 18.0, 14.3, and 8.5 Wins Produced respectively. Shaq’s relative low total reflects the fact he only played in 59 games and averaged less than 31 minutes per game.
From this we see that LeBron was very, very good in the regular season. Is he playing better in the playoffs? As I detailed in “Simple Models of Player Performance”, one can easily see if a player has improved by looking at his Win Score. In the regular season LeBron posted a per-minute Win Score of 0.258. The average small forward has a per-minute Win Score of 0.152, so as we saw with Wins Produced, LeBron’s Win Score tells us that he was really, really, good in the regular season.
In the first round of the playoffs, though, LeBron’s per-minute Win Score dipped to 0.201. Again, he is well above average. But Regular-Season-LeBron was better than First-Round-LeBron.
What about the second round against the Pistons? Again, LeBron’s productivity has dipped. Against Detroit LeBron’s per-minute Win Score is 0.184. He is still above average. He is still playing well. But Regular-Season-LeBron was better than Second-Round-LeBron.
What has caused the decline? The key is turnovers. In the regular season LeBron turned the ball over about once every thirteen minutes he played. In the playoff, though, he is turning the ball over once every nine minutes played. As a result his turnovers-per-game has risen from 3.3 per contest in the regular season to 5.2 per game in the playoffs. So although LeBron is offering similar levels of shooting efficiency and rebounds, his increased turnovers has reduced his effectiveness.
So why are the Cavs winning? Two players have played better in the post-season: Drew Gooden and Donyell Marshall. Now one can argue that these players are more effective because the Wizards and Pistons focus their attention on LeBron. And I cannot dismiss that argument. But the media is not showing many highlights of Gooden and Marshall and noting that their performance is possible because of the greatness of King James. No, the media is mostly showing highlights of LeBron and noting his greatness. And he is still great. But the numbers tell us he was better in the regular season.
By the way, this is the same story I told earlier about Kobe Bryant. And it is the same story we told about Michael Jordan in the book. An excerpt detailing Jordan in the playoffs is available here. So LeBron’s slight dip in the post-season is not unique. Most stars play worse in the playoffs. This should not be surprising. LeBron is not playing the Atlanta Hawks this week. He is playing the Pistons. And the Pistons won 64 games this year because Detroit is pretty good. And I think after committing seven turnovers last night and only hitting 40% of his field goal attempts, LeBron just might agree.
— DJ
Wages of Wins Journal 
cw
May 22, 2006
The problem with statistical analysis of sports performance is that it doesn’t capture enough. For one, it is very difficult to measure defense, at least in basketball, the sport I know the most about. Neither have I seen any kind of good statistical quantification of timeliness. It’s often not some much how many shots you make, or rebounds, steals, blocks, or stops you get, when you get them. And I think that is what they talk about when they talk about raising your game in the playoffs. Certain players get the ball at the end of the game and the best of these do something good with it more often than those who don’t. This is something that we can recognize instictively, but not statistically.
The same is true of something you mention in your post, and that is that the effort needed to stop one player like Labron or MJ or Shaq creates oportunities for other players. In the Cleveland series this was very much the case with Labron.
So I think statistics can be very vaulable, but you shouldn’t totally discount your intuitive impression. Intuition can be fooled, but at the same time, it captures and processes information that pure mathematical statistics cannot.
zydrunas
May 23, 2006
huh? “raising your game” is relative, not absolute. if a .330 hitter in a baseball league that averages .260 bats .320 in the postseason against pitching staffs that held opponents to a .230 batting average, the hitter “raised his game” simply because what he did was harder.
similarly your assessment of lebron has to take into consideration detroit’s World’s Best Defense (TM). at what level should we expect him to play against detroit’s defense, and how does that compare to how he actually played?
and of course marshall & gooden’s game results from the extra attention paid to lebron. i agree that the observer doesn’t know much, but that one is hard to miss.
zydrunas
May 23, 2006
also: cw, statistics can capture when you do stuff. all you have to do is say: how many points, rebounds, assists, turnovers, blocks, etc., did player X contribute during the last six minutes of a game while the score differential was no greater than four. for instance. stats can capture whatever you want, as long as you define what you want carefully.
queenjames
May 23, 2006
Honestly, I think your assertion is a load of crap. Your statistics can’t be denied, and your argument is interesting and well calculated. But you do not account for lebron’s impact on the game. You mention “…Now one can argue that these players are more effective because the Wizards and Pistons focus their attention on LeBron. And I cannot dismiss that argument…” but you aren’t considering that this is the only argument that matters. Donyell Marshall made a big block on Tayshaun, lebron did not influence that. Drew Gooden stepped up a game or two (most notably his 20 point 10 rebound performance against the wizards- a game the cavs lost). But many of the rebounds and open 3s and putbacks by Gooden and Marshall are due to defenders not paying attention to them and leaving them open. In game 7 against the Pistons, according to your formula, lebron played almost the whole game and lost, so therefore his per-minute win score is low – but did you see the game? He single handedly kept his team in the game the first half. His teammates shot a miserable percentage from the field, and thats even with the Pistons triple-teaming him whenever he touched the ball. No, it’s not Lebron’s fault that the Cavs don’t win. It’s because nobody else on the Cavs can score.
Ryan Early
May 25, 2006
I concur with the comment from zydrunas. Instead of comparing Lebron’s regular season stats to his postseason stats, shouldn’t you have compared his regular season stats against the Pistons to how he performed in the postseason series with the Pistons?
However, you are essentially correct. The analysts’ “stepping up” story line is about marketing, not expert analysis. TV ratings are generated by stars. Stars are created as much by marketing as by talent. Why are they hyping LeBron so much? Because they are being paid to.
cw
May 27, 2006
I agree with queenjames (LaBron’s wife?). If he wasn’t on the team the cavs don’t even make the playoffs. His abliites and the attention they draw make that team viable. It is obvious if you watch the game but I don’t see how you can capture his effect through those basic statistics.
Maybe to really describe basketball you have to model it like they do the weather. Bball is a complex dynamic system. It’s ten people acting all at once. One players abilities are effected by his opponent, but also by all the other players on his team.
There’s no way you can measure the subtle effects a player has on the game through such basic statistics like shot taken, points made, etc… It doesn’t capture the dynamics of the game.