Cowboys Stupid Mistake Rating, Week 11

[fanfromvirginia]

At the beginning of the season, I came up with an idea for a new stat, which I'm calling the Stupid Mistake rating. The formula is giveaways x 6 + # of penalties.

I've adjusted the ratings for every team based on last week's games. We had 18 points (2 TOs and 6 penalties). This brought our overall average up to 17.3, which is 20th in the league. We're second to last in penalties (ahead of only GB). We're about average in giveaways. Overall, we were dead last in the league in 2008 using this rating so we've gotten better.

Dallas' average in its 3 losses is 25 and 14 in its 7 wins.

League best (Chargers): 11.8
League worst (Carolina): 20.6
League average: 16.24
Cowboys: 17.3

This week's NFL-wide data (Team/Giveaways/Penalties/Rating):

1. SD/11/52/11.8
2. NE/10/59/11.9
3. MIN/10/59/11.9
4. KC/11/57/12.3
5. GB/8/80/12.8
6. IND/13/51/12.9
7. MIA/13/51/12.9
8. BAL/11/69/13.5
9. DEN/14/55/13.9
10. JAC/16/50/14.6
11. SF/15/64/15.4
12. ATL/17/52/15.4
13. CIN/16/62/15.8
14. NYG/16/62/15.8
15. WAS/17/56/15.8
16. HOU/16/63/15.9
17. PHI/15/75/16.5
18. SEA/18/58/16.6
19. PIT/19/53/16.7
20. DAL/16/77/17.3
21. STL/18/66/17.4
22. TB/20/56/17.6
23. NO/19/62/17.6
24. TEN/20/62/18.2
25. CHI/21/68/19.4
26. BUF/20/75/19.5
27. ARI/21/69/19.5
28. DET/21/70/19.6
29. NYJ/23/61/19.9
30. CLE/25/51/20.1
31. OAK/24/61/20.5
32. CAR/25/56/20.6

 

[CF74]

Left turn Clyde

 

[Hostile]

Left turn Clyde

Right turn. Left turn would hit Clint in the face.

 

[CF74]

Right turn. Left turn would hit Clint in the face.

To much Vino:laugh2:

 

[jackrussell]

Why is there a multiple of 6 involved?

How does this: The formula is giveaways x 6 + # of penalties = this: Cowboys: 17.3?

 

[fanfromvirginia]

Why is there a multiple of 6 involved?

How does this: The formula is giveaways x 6 + # of penalties = this: Cowboys: 17.3?

It's a per game average. They make 1.6 giveaways per game (x6 = 9.6) + 7.7 penalties/game = 17.3.

Why is it 6? Because I chose it. I'm thinking of lowering it to 5 to deemphasize giveaways a bit. There are about 3.5 times more penalties than giveaways. I wanted to weight the giveaways a little bit more than 3.5, so I made it 6.

 

[jackrussell]

It's a per game average. They make 1.6 giveaways per game (x6 = 9.6) + 7.7 penalties/game = 17.3.

Why is it 6? Because I chose it. I'm thinking of lowering it to 5 to deemphasize giveaways a bit. There are about 3.5 times more penalties than giveaways. I wanted to weight the giveaways a little bit more than 3.5, so I made it 6.

Wouldn't #giveaways + #penalties achieve the exact same results?

 

[Hypnotoad]

Wouldn't #giveaways + #penalties achieve the exact same results?

no he wants to give weight to giveaways. giveaways are more of a stupid mistake than penalties

 

[SaltwaterServr]

no he wants to give weight to giveaways. giveaways are more of a stupid mistake than penalties

To be more accurate, turning the football over should be adjusted by field position and result. A fumble on the opponent 25 yard line should in fact not be given the same weight as an interception returned for a TD.

I understand what the OP is trying to do, but it's a gross oversimplification. Penalties are the same way. A personal foul call against you when you just converted 3rd and 10 is huge compared to a false start on 1st and 10. You've got 3 shots at making up the extra 5 yards whereas the 3rd and 1/4 the field is a killer.

The time involved though in adjusting each penalty and turnover would take quite a bit of time to figure out the algorithm. I cannot remember the site off the top of my head, but they're the one's who came up with DVOA metrics. They adjust their formulas a little bit each year it seems like.

Anyway, a fumble or interception immediately returned for a TD should be weighted the greatest.

Turnovers inside your own 30 should be second heaviest since the opponent is immediately in field goal range. Turnovers on your side of the field from the 31-50 get a 3rd tier weight.

50-35 on the opposition get a 4th tier.

From their 35-15 you bump it back to a 3rd tier because you've taken away a field goal scoring opportunity. However, they've got a long way to go to get their own points and the onus is partially back on the defense so it isn't as bad as a second heaviest.

Now the turnovers inside the 15 can go either way to a second or a third. You've got a sure fire field goal but you've gotten well within the red zone and the defense holding you to a field goal is a partial victory on their part. That again is mitigated by the fact that the defense now has kept you from scoring completely but their offense now has a a very long field to go to get their own points. So that's a conundrum right there. A kick-off return that makes it to the 12 is horrible field position and great special teams play but a fumble recovery defending your red zone is a great play. I guess it's a 2nd tier because it takes so much wind out of the offense's sails, but I could argue it for a 3rd tier.

Now all of that needs to be weighted against penalties that ended drives. If the penalty is committed and you convert the first down, it's mitigated to a degree. However, if you just do convert the first and gain no more yards afterwards say on a holding call you have sacrificed field position by having to conquer a large chunk of real estate twice.

Enough.

 

[fanfromvirginia]

To be more accurate, turning the football over should be adjusted by field position and result. A fumble on the opponent 25 yard line should in fact not be given the same weight as an interception returned for a TD.

I understand what the OP is trying to do, but it's a gross oversimplification. Penalties are the same way. A personal foul call against you when you just converted 3rd and 10 is huge compared to a false start on 1st and 10. You've got 3 shots at making up the extra 5 yards whereas the 3rd and 1/4 the field is a killer.

The time involved though in adjusting each penalty and turnover would take quite a bit of time to figure out the algorithm. I cannot remember the site off the top of my head, but they're the one's who came up with DVOA metrics. They adjust their formulas a little bit each year it seems like.

Anyway, a fumble or interception immediately returned for a TD should be weighted the greatest.

Turnovers inside your own 30 should be second heaviest since the opponent is immediately in field goal range. Turnovers on your side of the field from the 31-50 get a 3rd tier weight.

50-35 on the opposition get a 4th tier.

From their 35-15 you bump it back to a 3rd tier because you've taken away a field goal scoring opportunity. However, they've got a long way to go to get their own points and the onus is partially back on the defense so it isn't as bad as a second heaviest.

Now the turnovers inside the 15 can go either way to a second or a third. You've got a sure fire field goal but you've gotten well within the red zone and the defense holding you to a field goal is a partial victory on their part. That again is mitigated by the fact that the defense now has kept you from scoring completely but their offense now has a a very long field to go to get their own points. So that's a conundrum right there. A kick-off return that makes it to the 12 is horrible field position and great special teams play but a fumble recovery defending your red zone is a great play. I guess it's a 2nd tier because it takes so much wind out of the offense's sails, but I could argue it for a 3rd tier.

Now all of that needs to be weighted against penalties that ended drives. If the penalty is committed and you convert the first down, it's mitigated to a degree. However, if you just do convert the first and gain no more yards afterwards say on a holding call you have sacrificed field position by having to conquer a large chunk of real estate twice.

Enough.

Sounds good. Go for it. My main thing on this is the data needs to be available in a box score or elsewhere on a stats site so the info is available from every team on a weekly basis.

 

[fanfromvirginia]

no he wants to give weight to giveaways. giveaways are more of a stupid mistake than penalties

Well, it's not just that. It's that if you don't weight them at all then almost 80% of your stat (and therefore variance) is made up of penalties. So multiplying penalties by 3.5 just evens things out. You're exactly right about the extra weight.

 

[THUMPER]

Sounds good. Go for it. My main thing on this is the data needs to be available in a box score or elsewhere on a stats site so the info is available from every team on a weekly basis.

That's what I was going to say too. Maybe Adam has access to all those stats/variations but the average fan does not.

But if SaltwaterServr wants to spend the time going through each "gamebook" on NFL.com and look up the situation for each play and extrapolate the variations and calculate the stats, I for one would be thrilled to see it!

STATS.inc probably has all of that but it isn't available for free.

 

[fanfromvirginia]

Wouldn't #giveaways + #penalties achieve the exact same results?

You need to weight these to make them comparable. Here's a more extreme example that will help illustrate the principle. You want a stat, for whatever reason, that includes rushing yards and TDs. If you just add the rushing yards and TDs together the importance of TDs will become meaningless as they'll be wildly outnumbered by rushing yards. Every one-yard gain would equal every rushing TD.

 

[SaltwaterServr]

That's what I was going to say too. Maybe Adam has access to all those stats/variations but the average fan does not.

But if SaltwaterServr wants to spend the time going through each "gamebook" on NFL.com and look up the situation for each play and extrapolate the variations and calculate the stats, I for one would be thrilled to see it!

STATS.inc probably has all of that but it isn't available for free.

If I hit the lottery, I might sit down and do it. This isn't exactly something you'd want to start in the middle of the season.

Now if you started with the beginning of the season, it would be so much more manageable. Discussing and hashing out how someone would do it is something that could be done now.

In the meantime, I'll stick with the OP's single tier weighting system.

 

[lspain1]

This is an interesting stat and provides some backup to this feeling I have that the Cowboys seem to shoot themselves in the foot more than most. I also believe that the Cowboys seem to draw slightly more than their fair share of tick-tack penalties (notice I said slightly and no, I'm not into conspiracy theories).

 

[Wrangler87]

This is an interesting stat and provides some backup to this feeling I have that the Cowboys seem to shoot themselves in the foot more than most. I also believe that the Cowboys seem to draw slightly more than their fair share of tick-tack penalties (notice I said slightly and no, I'm not into conspiracy theories).

And on the other side of the spectrum is the Patriots, who to no ones suprise, are number two. (no, I'm not into conspiracy theories either...but)

 

[fanfromvirginia]

This is an interesting stat and provides some backup to this feeling I have that the Cowboys seem to shoot themselves in the foot more than most. I also believe that the Cowboys seem to draw slightly more than their fair share of tick-tack penalties (notice I said slightly and no, I'm not into conspiracy theories).

Yeah, that all sounds about right to me. In terms of the rating, I've avoided the whole cause and effect deal thus far but am beginning to think that it is likely a pretty big factor. There are five times we've had a score of 18 or higher. Three of those five games were our losses and the other two were our squeaker wins against KC and Washington.

 

[fanfromvirginia]

If I hit the lottery, I might sit down and do it. This isn't exactly something you'd want to start in the middle of the season.

Now if you started with the beginning of the season, it would be so much more manageable. Discussing and hashing out how someone would do it is something that could be done now.

In the meantime, I'll stick with the OP's single tier weighting system.

On my part, I may just rebrand this after the season. If you wanted to truly capture 'stupidity' you would need to beef this up significantly. However, I think that this metric is capturing something important AND it's really, really easy to keep track of and compare against other teams.

I think it just needs a less ambitious title.

 

[Phrozen Phil]

I note that KC is 4th and NO is 23rd, so I'm wondering how that translates in comparison to the other teams. It's an interesting series of stats and I commend your effort, but these two team's stats present a bit of a head scratcher.

 

[fanfromvirginia]

I note that KC is 4th and NO is 23rd, so I'm wondering how that translates in comparison to the other teams. It's an interesting series of stats and I commend your effort, but these two team's stats present a bit of a head scratcher.

There's no question that this metric doesn't fully explain success -- thus the NO, KC, and, to a lesser extent, Dallas exceptions. But it does, IMO, generally track well with success and failure so I think the KC, NO, Dallas exceptions become very interesting in terms of this metric rather than deadly to it. I definitely think this has something to do with both relative success and measuring 'stupidity' or something like it.