The 1-7 Dallas Cowboys: An Interim Coach’s Dream?
A reader recently raised a point I’ve heard echoed by many media outlets around the country–that interim head coach Jason Garrett is inheriting a terrible situation in Dallas. The team is 1-7, the starting quarterback is out for the reason, and the team seems to have quit.
I understand the view and of course the situation in Dallas is a difficult one. But let’s not forget that Garrett doesn’t have to go 7-1 from here on out. To have a chance at retaining the head coaching job, he simply needs to show the ability to rally the troops and have them improve.
Improving on 1-7 isn’t that difficult. As bad as things are going, even if the Cowboys retained Wade Phillips as head coach, you’d expect them to be better than 1-7 in the second half of the season. It’s simple regression to the mean.
There’s a pretty interesting win expectation formula known as “Pythagorean Win Formula” that tells you (to a surprisingly accurate degree) how many games a team “should” win based on their points scored and yielded. There are variations of the formula, but a basic version of it is below:
(Points Scored)^2/((Points Scored)^2 + (Points Yielded)^2)
The formula has been tested again and again in real-life situations and holds up year after year. Score X points and give up Y, and you can expect to win Z games more times than not. It’s pretty radical.
Well, as poorly as the Cowboys have played, when you plug their numbers into the formula, their winning percentage “should” be .325. Not good. But almost three times as good as their current .125 winning percentage.
So even if the Cowboys play exactly like they have in the first half, the most likely outcome is they’d win three of the final eight games. There isn’t much Garrett needs to do for this team to improve–it’s simple regression to the mean.
Think of it like this. . .you can take a completely random event, such as the flip of a coin, and it will exhibit “regression to the mean” behavior. We know a coin “should” lands on heads 50 percent of the time. So if we flip it 10 times and only get one head, what are the chances that we’ll improve on that number in the next 10 flips? Really, really good, even though the act is totally random and the first 10 flips have no bearing on the next 10.
So if the Cowboys are a .325 team, the chances they’ll improve on their one-in-eight win rate are really good, despite the current circumstances.
And if Garrett can at least get the team to play hard again, that number should be even higher. Remember, when things are going poorly, they’re never really as bad as they appear. This was a nightmare season and not much more more could have gone wrong.
Which is exactly why now is the perfect time to take over in Big D.