fanfromvirginia
Inconceivable!
- Messages
- 4,014
- Reaction score
- 164
As you can see from my sig, I'm worried about stupid mistakes. I'm hopeful that the Cowboys can cut down on these given the changes in personnel, the increased discipline, our new special teams coach, etc. But I'm not counting on it. And I think this is the most likely contributing factor were we to continue to underperform once again.
I've decided to try to keep track of stupid mistakes this season. I've created what I'm calling the Stupid Mistake stat, which I will calculate for each game and as a season-long average. I'll post this after each game, time and sanity permitting. I'm open to suggestions on how to calculate this but I'm also a lazy git so you're welcome to chime in with your own logic and numbers as the season goes along if I decide to stick to the following formula. I have erred on the side of simplicity because I don't want to put that much time into it and I want people to be able to calculate it without too much effort.
I have decided to use number of giveaways x 10 plus the number of penalties for a number of reasons. First, as, if memory serves, Yakuza Rich has pointed out, net turnovers correlate much better with winning than penalties. And the number of turnovers per game, and the margins between the teams, is quite a bit lower than with penalties. So I'm multiplying by 10 to adjust. I chose ten because it seems close enough for government work and it's easy. I'm using giveaways per game because this is a better measure of stupidity than net turnovers, which is no doubt better correlated to winning. I'm using penalties per games, rather than penalty yards because it's simpler and the two seem to be pretty closely correlated. Of course there are in theory all kinds of ways to be stupid during a football game but they're tough to quantify so I'm sticking with these two.
Here's what happened in 2008. The Cowboys were 31st in the league in turnovers per game at 2.1. The league median was 1.6. The Cowboys were 32nd in the league (last, of course) in number of penalties with 7.4/game. The league median was 5.5. Using the formula my SM factor (yep, that's what I'm calling it) we see that the league median was 1.6 x 10 + 5.5 = 21.5. The Cowboys were 2.1 x 10 + 7.4 = 28.4. The league best NE Patriots were 0.9 x 10 + 3.6 = 12.6. I'll be using these as benchmark numbers in my updates. I'll use the following data up front in each update:
League best, 2008: 12.6
League median, 2008: 21.5
Cowboys, 2008: 28.4
Cowboys, 2009: 0.0/TBD
It would be interesting to run regressions using this stat but I am W-A-Y too lazy for that.
I've decided to try to keep track of stupid mistakes this season. I've created what I'm calling the Stupid Mistake stat, which I will calculate for each game and as a season-long average. I'll post this after each game, time and sanity permitting. I'm open to suggestions on how to calculate this but I'm also a lazy git so you're welcome to chime in with your own logic and numbers as the season goes along if I decide to stick to the following formula. I have erred on the side of simplicity because I don't want to put that much time into it and I want people to be able to calculate it without too much effort.
I have decided to use number of giveaways x 10 plus the number of penalties for a number of reasons. First, as, if memory serves, Yakuza Rich has pointed out, net turnovers correlate much better with winning than penalties. And the number of turnovers per game, and the margins between the teams, is quite a bit lower than with penalties. So I'm multiplying by 10 to adjust. I chose ten because it seems close enough for government work and it's easy. I'm using giveaways per game because this is a better measure of stupidity than net turnovers, which is no doubt better correlated to winning. I'm using penalties per games, rather than penalty yards because it's simpler and the two seem to be pretty closely correlated. Of course there are in theory all kinds of ways to be stupid during a football game but they're tough to quantify so I'm sticking with these two.
Here's what happened in 2008. The Cowboys were 31st in the league in turnovers per game at 2.1. The league median was 1.6. The Cowboys were 32nd in the league (last, of course) in number of penalties with 7.4/game. The league median was 5.5. Using the formula my SM factor (yep, that's what I'm calling it) we see that the league median was 1.6 x 10 + 5.5 = 21.5. The Cowboys were 2.1 x 10 + 7.4 = 28.4. The league best NE Patriots were 0.9 x 10 + 3.6 = 12.6. I'll be using these as benchmark numbers in my updates. I'll use the following data up front in each update:
League best, 2008: 12.6
League median, 2008: 21.5
Cowboys, 2008: 28.4
Cowboys, 2009: 0.0/TBD
It would be interesting to run regressions using this stat but I am W-A-Y too lazy for that.
