I was watching game 5 of the NBA Eastern Conference Finals between the Heat and the Pacers where Lebron James picked up 2 early fouls and was immediately benched. This strategy, which I have seen numerous times, on most NBA teams, is puzzling to me. In this instance, you have arguably the best player in the game and you are electing to not use him for the maximum amount of time possible?
Lebron James is one of the leading scorers in the NBA, and has been for the past 10 years. He has been named to the NBA All Defensive First Team for five consecutive years. This means that his impact on the game on both ends of the floor is enormous and that whomever comes off the bench to replace him is going to be a significant downgrade. Therefore, when you elect to put him on the bench and to replace him with an inferior player, there had better be a compelling reason.
The reasoning generally given in situations like these is that you want ensure that he does not foul out so that he can play at the end of the game in case it is close. I believe this is faulty logic. Firstly, James has only fouled out 5 times in his 10 year NBA career, for an average of once every other year. On average, he picks up 2.1 fouls per 48 minutes, meaning that even having picked up 2 early fouls, he is still extremely unlikely to foul out of the game. It is extremely likely that he will be available at the end of the game, even having picked up 2 early fouls.
Secondly, sitting him early in the game costs his team points in the near term. Every minute that James is not on the floor and is replaced by a lesser player, he is theoretically costing his team points. In game 5, James played only 24 minutes instead of the 38 he has averaged during the playoffs. Since he is averaging 34 points per 48 minutes, and his replacements Udonis Haslem and Shane Battier average 4 and 10 points per 48 minutes, his sitting for an extra 14 minutes cost his team approximately 6.6 points of offense. Since it is unlikely that they can make up this point gap defensively (given that James was First Team All Defense), it is a good assumption that the cost to his team was greater than 6.6 points.
Thirdly, what about the concern about having him available at the end of the game? Well, as addressed above, it is still extremely unlikely that James will actually foul out of the game (for more in depth analysis, you can look at the times he picked up 2 early fouls and do analysis of the times he has fouled out, though its a very small sample size). This means that it is extremely likely that he will be available at the “crucial time.” Even if James were not available for the final 6 minutes of the game because he had fouled out, that loss would not be more than the expected 6.6 points per game the Heat lost by sitting him(that would mean that the difference between James and his replacement was worth 52.8 points per game).
So if sitting James because of 2 early fouls is a poor decision, then why do coaches do it? My guess is that since that is what every other coach does, it keeps them out of the line of fire. Its also much harder to be criticized for losing with your star in the game, rather than losing when he has fouled out.
The best winning strategy should be to have the best team on the floor for the maximum amount of time, and this clearly includes having Lebron James on the court as much as possible. Given the work done by other analysts like Nate Silver, Stephen Dubner and Steven Levitt, who have proven that there is no such thing as a clutch player or a hot hand, it should be clear that the goal is to create the most point differential throughout the game, not just at the end.
I try not to be results oriented, but the Pacers won by 3 points, 93-90, and that is well within the amount of points that sitting James cost the heat.