## Statistical Proof that the Timberwolves are Cursed

Maybe it started on April 19, 2006. That's when Mark Madsen went 0-7 on three-pointers to ensure that the Wolves would avoid Clippergeddon for another draft. Or maybe it started on January 23, 2007, when the Timberwolves fired Dwane Casey in favor of Randy Wittman despite Casey's 0.500 season record at the time.

The fact is, I don't know when it started. But now I've proven that it exists. The Timberwolves are cursed.

EXHIBIT A

Pythagorean Wins (PWins) are used to predict a team's record based on how many points it has scored and given up. As far as I know, it's supposed to work pretty well as a prediction of team performance. Table 1 shows how well it has worked for the Wolves dating back to the 2007-08 season. In each case, the Wolves underperformed their PWins. This is a little strange, but it's not quite proof of a curse.

Table 1: Timberwolves Actual Win Percentage versus Pythagorean Win Percentage

 Season Actual Win% Pythagorean Win% Difference Per-82 Games 2013-14 0.478 0.609 -0.130 -10.7 2012-13 0.378 0.415 -0.037 -3.0 2011-12 0.394 0.424 -0.030 -2.5 2010-11 0.207 0.293 -0.085 -7.0 2009-10 0.183 0.207 -0.024 -2.0 2008-09 0.293 0.329 -0.037 -3.0 2007-08 0.268 0.280 -0.012 -1.0

EXHIBIT B

Eric in Madison recently wrote a fine article about the Four Factors. It made me wonder how the each factor contributes to a team's success and whether the Wolves' odd profile (bad own and opponent eFG%, but good at most everything else) can produce enough wins to make the playoffs. So I did what Madison Dans do: I ran a regression.

I used team level data for five seasons, from 2008-09 through 2012-13. Each observation consists of one team-season. The dependent variable is the team's winning percentage for the year. The independent variables are the Four Factors (eFG%, TOV%, ORB%, and FT/FGA) that they and their opponents produce (for a total of eight variables).

Then I add a ninth variable, which I call "Curse". This variable is equal to one for the Timberwolves and zero otherwise.

Table 2: Regression of Win% on the Four Factors (and the Curse)

 Variable Coefficient Standard Error t-Stat Intercept 0.732 0.263 4.32 Own eFG% 4.261 0.188 22.68 Own TOV% -0.048 0.004 -12.88 Own ORB% 0.013 0.001 9.80 Own FT/FGA 1.129 0.134 8.43 Opp. eFG% -4.428 0.214 -20.67 Opp. TOV% 0.037 0.003 11.22 Opp. ORB% -0.012 0.002 -5.77 Opp. FT/FGA -1.261 0.136 -9.28 Curse -0.043 0.018 -2.39

Roughly speaking (statistically), when the t-Stat is higher than 2.0 in absolute value, the variable has a statistically significant effect on team winning percentages. Not surprisingly, all of the Four Factors matter.

Now look at the last row. Just being the Timberwolves, controlling for all of the other Factors (for and against) costs the team 4.3 percentage points of its win percentage, or 3.5 games per 82-game season. The Curse is statistically significant.

Where Do We Go From Here?

The Curse exists. We know this now. No sense bickering and arguing about who started it. We need to figure out a course of action. I propose that we start with what WE can do: sacrificing CH members. Since we live in so many different places, traditional ritual sacrifices are not practical. So I propose periodic lotteries to select sacrificial bans. Fear not, for we will celebrate and memorialize the "winners"!

If that doesn't work, I guess we'll just have to trade Kevin Love.

## Trending Discussions

forgot?

As part of the new SB Nation launch, prior users will need to choose a permanent username, along with a new password.

I already have a Vox Media account!

### Verify Vox Media account

As part of the new SB Nation launch, prior MT authors will need to choose a new username and password.

We'll email you a reset link.

Try another email?

### Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

### Join Canis Hoopus

You must be a member of Canis Hoopus to participate.

We have our own Community Guidelines at Canis Hoopus. You should read them.

### Join Canis Hoopus

You must be a member of Canis Hoopus to participate.

We have our own Community Guidelines at Canis Hoopus. You should read them.