Do Heavy Minutes Prevent Championships?

As a Bulls fan, I've been thinking quite a bit lately about how whether the heavy minutes Tom Thibodeau plays Luol Deng and Derrick Rose will come back to bite the Bulls in the playoffs.  There has been quite a bit of talk about how stars' extended minutes could hurt teams this year with the compressed schedule, and I've been wondering if there have been any studies concluding that less rest hurts players' production.

This morning I had an exchange on Twitter (@TheTeamRebound) with Henry Abbott, inspired by an article he wrote for TrueHoop on this very subject entitled "Heavy minutes hurt title chances."  The basis for the article was Henry's observation that no team since the 2004 Pistons has won a title with a player exceeding 3,000 minutes in a regular season, which is 37 minutes per game over an 82 game season.  This was true despite the fact that many of the NBA's luminaries the last 7 seasons exceeded 3,000 minutes.

Henry likewise noted that no player has averaged more than 39 mpg and won the title since Tim Duncan in 2002-03.  He also provided some anecdotal evidence from NBA insiders that the game has evolved so that it's simply harder to play as many minutes now.*

*I'm pretty sure I agree with this assessment.  If you look at games back in the 80s, even the Finals, teams just didn't play anywhere near the type of pressure defense that is played today.  Also, nobody shot 3s back then so the defense didn't have as much ground to cover.  Even in the 90s, as defensive intensity markedly improved, isolation ball became more prevalent so defensive players could take a lot of possessions off.  One caveat though is that the pace was quicker back then, and there were less--and shorter--TV timeouts in a lot of games.

One other fact lends credence to Henry's theory: Of the 56 teams to make the conference semifinals since 2004, 23 are represented on the 3,000 minutes played list.*  Assuming, in a vacuum, a 1 in 8 chance for each conference semifinalist to win a championship, one would have expected approximately 3 of the last 7 championships to have been won by this group.  Instead, these teams won zero.

*The 2010-11 Bulls and the 2004-05 Suns each had two qualifying players.

All of this begets the question of whether those players didn't win titles because they played long minutes.

On reading the article, I tweeted that two additional factors might contribute to the fact that teams with heavy minute players haven't won championships of late.  One was that really good teams have more blowouts and/or meaningless games at the end of the season, and thus less need to play their key guys so many minutes.  John Hollinger later made this same point, which he clearly stole from me.  *sarcasm font*  The second reason is that older teams tend to win championships, and veteran players are not driven as hard by their coaching staffs.  This would certainly seem to be the case with the recent vintage Spurs, Celtics, and Mavs champions in particular.  Tonight I thought of a third potential factor: Every team that has won a title since 2004 has done so with a star big man who was one of the top 20 players in the NBA, and big men in general play less minutes.



Henry suggested in response that I look at the list of players who had exceeded 3,000 minutes, so I did.  If it were the case that their stars' heavy minutes was the death knell for these teams, one would assume this would manifest in those players' decreased production in the playoffs.

Using John Hollinger's PER, which weights all of a player's statistical contributions, I went through the list to see whether the heavy minute players in fact suffered larger than expected declines in performance during their teams' unsuccessful playoff campaigns.  Previous research by Hollinger (which I wasn't able to find a link to for some reason) has shown that the typical playoff PER declines by more than a point due to the better competition in the playoffs.

First though, I subjectively eliminated those players whose teams were not considered realistic title contenders at any point that season.*  These teams were so far from winning the title that there was no conceivable way an incremental increase in the 3,000 minute player's performance would have won these teams a championship, especially when the team in question missed the playoffs entirely.

*Feel free to quibble with my decisions on this one.  I tried to err on the side of overinclusiveness.



The high minute players' playoff PERs declined by 0.85 points on average, less than the typical playoff decline.  If these players were tired from playing so many minutes, there is no evidence that it meaningfully affected their playoff statistics.  For every Dirk Nowitzki 2005 (-6.00) there was a...Dirk Nowitzki 2010 (+5.40).

**One of these things is not like the other...

Another thing that struck me was that 64 of the 96 3,000 minute player-seasons were eliminated because their teams were not even remotely title contenders.  This was the case despite the fact that I was extremely charitable in which teams qualified as "title contenders."  For example, I included Elton Brand from the 2006-07 Clippers because they'd lost in Game 7 of the second round the year before, despite the fact they ended up going 40-42 and missing the playoffs.  Teams like the 2006-07 Bulls (Luol Deng), 2005-06 Nets (Richard Jefferson), 2004-05 Sonics (Ray Allen), 2004-05 T'Wolves (Kevin Garnett), 2008-09 Hornets (Chris Paul), and 2007-08 Nuggets (Allen Iverson) were at best fringe title contenders that I nonetheless included.  The distribution of these 3,000 minute seasons provides a lot of evidence that there is selection bias in this sample; 3,000 minute seasons tend to come from teams that are not legitimate title contenders.*

*Then again, most seasons tend to come from teams that are not legitimate title contenders.

It is true that teams whose stars play over 3000 minutes in a regular season do not win championships, at least during the last 7 years.  But since those players' heavy workloads do not translate into a reduction in playoff performance relative to their lower minute peers, it is difficult to conclude that the stars' heavy minutes cause their teams to lose in the playoffs.  Instead, perhaps the conclusion should be that the much- derided coaches are capable of identifying players that can handle long regular season minutes without a dropoff in playoff performance.