MATHS: Mistakenly predicting Derek Fisher's coaching using his playing career

In case it isn't abundantly clear, this is a deliberately quackish use of maths for summertime pleasure and it is not meant to be taken seriously. UNLESS YOU SEEK THE TRUTH -Seth

Lest we forget, Phil Jackson recently hired Derek Fisher to coach the Knicks. In all his inimitable wonder, the Zen Master decided on a just-retired role player who, along with Billy Hunter, ran the NBA Players' Union right into the ground. Players want a coach who they trust to not completely screw up, yet they've already watched him get trounced by the owners in CBA negotiations. Fisher has no official experience as a coach, though he maintains that's irrelevant. I think he'll be a failure, and I can prove it. To do so, I decided to do what I do best: use numbers to form tenuous connections that I'll spin as fact.

Below, you'll see a table depicting what former Knicks coaches accomplished during their playing days. This only includes stats from the NBA and ABA, so any international stats (such as Mike D'Antoni's from Italy) are not factored in. Coaches who played prior to the addition of the 3-point line were included, but I didn't call their 3FG% 0. Instead, I left those stats out of any calculations I did for 3FG%. Also, I stopped prior to Pat Riley, because fuck that guy. Finally, since Jeff Van Gundy never played in the NBA, he's not involved.

Next, here's a table listing those same coaches by record as Knicks Head Coach.

See? The Knicks suck. There are two coaches on there with winning records. Two! Derek Fisher is, as I expected, royally screwed.

Regardless, let's soldier on. I plotted each stat as a player against wins as a coach, and used a regression line to determine how the data correlates. The R² number on each graph tells us how much of the win percentage as a coach is explained by each stat as a player. Normally, you'd want a number kinda close to 1, but that wasn't the case with any stat here, so instead, I included the graphs for the three stats with the highest R² values. Those three stats are FG%, 2-point FG%, and Assists, though none of them have high R² values at all, so take this with a huge grain of salt.

First, let's take a look at what Derek Fisher's FG% as a player suggests his winning percentage as Knicks Coach will be. Using the equation above, we can do a mostly terrible job of that! Plug in Fisher's FG% of 0.399 for x, and you get 2.5444*0.399 - 0.6982 = 0.317, or 26 wins per season. Sounds about right.

Next, 2-point FG%, which correlates even worse than overall FG%. Plug in Fisher's 2FG% of 0.413 for x, and you get 2.029*0.413 - 0.4801 = 0.358, or 29 wins per season. Things are looking up!

Finally, assists, which correlate better than 2FG% but worse than FG%. Plug in Fisher's 3.0 AST for x, and you get -0.0177*3 + 0.5051 = 0.452, or 37 wins per season. You know who won 37 games last season? The Knicks! Well, I guess stagnation is better than regression.

The Knicks have sucked for years. There's no reason to think that a perpetual screw-up like Fisher will change anything. Correlation may not imply causation, but be warned: if we listen to the numbers, the Knicks are fucked.