We have a stat that measures the difference between you being in a game vs you being out of the game. That stat is far more reliable and takes out so much more of the noise than a stat that of how the team fared (with a player or without a player actually playing the game). The author himself admitted the Bulls were better defensively without Rodman. Does that sound like logic to you? It doesn't to me either (whether he was better than Jordan or not doesn't matter, we can both clearly see that Rodman was an extremely good defender).

The fact that Rodman missed so many games is obviously a knock against him. Perhaps Rodman's game differential of with or without you is better than Shaq's because Rodman would wait until he was 100% and Shaq would play as soon as he could. There are a million reasons why that could be the case or you could subscribe to his "X factor" that somehow the team played better when Rodman played. "Just seeing Rodman on the court made Michael Jordan want to try harder."

The author conveniently used stats from 1986, because using his model guys like Wilt and Russell would have destroyed his "advanced stats" methods. While Rodman rebounded equal to them, he couldn't touch either of them in scoring, assists, blocks, steals etc.

He mentioned PER is a flawed stat. Yes, it absolutely is. So are the stats that he came up with. As Charles Barkley said "I could bet 18 rebounds per game if I wanted to as well". I believe he could. The difference? Well Charles was actually trying to score and pass and facilitate his team scoring. Rodman simply wanted to get a rebound, nothing more.

Was Jordan the best player because he scored the most? Well, yes but that is just a part of it. The author decided to completely ignore offensive efficiency, assist rates, defense, etc etc. Those narratives would not have fit his argument.

The author purported that rebounding is a more important stat than we think. Yet there was no compelling evidence that this was true. He compared individual rebounding percentage to individual ppg as a method of predictor. I have no doubt rebounding percentage is higher (and it was). Now compare that to many of the other advanced stats out there and you will find that it isn't close to being a great predictor. It isn't the points per game that is important, but how you arrived at that many points (again efficiency).

Quite frankly the author liked Dennis Rodman from an early age and found statistics that fit his narrative. Somehow the author failed to mention that the Bulls/Spurs simply weren't that much better when Rodman was in the game (again, box score plus minus! The author kept saying impact on the game and ignored the one stat that tells you the actual point effect the player had on the game).