Vela: A. Jones, CB Rankings and Ken Hamlin’s Best Position: K.C. Joyner Returns, Pt.1

[AdamJT13]

Which of these defenses performed better, statistically?

Defense X's possessions faced
1 play, 0 yards, INTERCEPTION
5 plays, 65 yards, TOUCHDOWN
3 plays, 4 yards, INTERCEPTION
7 plays, 53 yards, FIELD GOAL
3 plays, 3 yards, PUNT
4 plays, 60 yards, TOUCHDOWN
13 plays, 49 yards, TURNOVER ON DOWNS
4 plays, 40 yards, TOUCHDOWN
3 plays, 0 yards, PUNT
7 plays, 32 yards, MISSED FIELD GOAL
3 plays, 3 yards, PUNT
5 plays, 20 yards, INTERCEPTION
3 plays, 9 yards, PUNT
3 plays, 6 yards, PUNT
7 plays, 16 yards, TURNOVER ON DOWNS
3 plays, 3 yards, INTERCEPTION


Defense Y's possessions faced
6 plays, 21 yards, FIELD GOAL
5 plays, 49 yards, TOUCHDOWN
3 plays, 5 yards, PUNT
9 plays, 51 yards, FIELD GOAL
3 plays, minus-10 yards, PUNT
13 plays, 71 yards, FIELD GOAL
9 plays, 93 yards, TOUCHDOWN

 

[percyhoward]

Remember, I eliminated all possessions when the offense either didn't have time to score or had no intention of scoring. The vast majority of games had 8-11 possessions for each team.

I know you only counted the possessions that "should" count. According to these charts from Football Outsiders, the number of such possesions in an average game is a little over 11. I understand that may not be typical of your sample, but I'd still like to see the data for 12-possession games, and you still haven't shown it. That's fishy.

With more or less than that, the sample size was smaller and the stats fluctuated wildly.

How exactly? What were the numbers, if you don't mind. Maybe it's supposed to fluctuate. It's only fluctuation in the mind of someone who is assuming there's a points-to-possessions correlation to begin with.

In order for your "expected" point totals to happen, the number of possessions in a game would have to be unaffected by the performance of the offenses and defenses involved.

That's right. As I have always said, you can't project a 10-possession performance into a 15-possession game. It changes the nature of each possession. It means it's shorter. Which usually means the defense is playing better. That's word-for-word what I've been saying for...too long.

And their not MY expected point totals, they're the ones you throw around every time this topic comes up. You say (said?) 25% more possessions means 25% more points--now you're amending that to say "IF performance level is the same." My whole point has been that it's about performance, not counting possessions. An offense that scores more points on more possessions does so because of how it performed, not because of how many possessions it had.

You've gone from saying that number of possession affects scoring, to saying that "performance affects how number of posessions affects scoring."

The average offense in the 134 games I looked at (10 teams, 16 games each, minus 26 duplicates) gained 30.0 yards per possession.

Two problems with that. First, you based the whole thing on another possession-based stat (yards per possession). It's only natural that, what would be defined as an "average" team using yards per possession would end up in the same place using points per possession. Why not use 2007 rank in total offensive points (the middle three or four teams, for example)?

Second, why throw out anything? If this is the logic you're using, it's shaky...

Better defense and/or worse offense leads to more possessions, while worse defense and/or better offense leads to fewer possessions. That means worse offenses get more chances to make up for performing poorly, and better offenses get fewer chances.

That's a strange interpretation of a natural occurence. That's why points don't go up with more possessions--the possessions are shorter because you still have to do the whole thing in 60 minutes. Essentially, that's football. It's not something that should be compensated for, just to make your favorite stat work better. If you want to, go right ahead. Just know that after it's been cut up and shaped and carved so much, it may not resemble anything that approaches the sum total of what actually happened on the field.

If number of posessions affects scoring (for all teams, not just the ones you choose), then shouldn't you have been able to show how by now?

 

[percyhoward]

Which of these defenses performed the worst, statistically, in 2007?

Cowboys
Chiefs
Titans
Bears
Commanders
Eagles
Packers
Vikings
Stealers
Colts
Bucs
Patriots
Chargers
Seahawks

According to points per possession, they are in inverted order (worst-to-best).

 

[AdamJT13]

Which of these defenses performed the worst, statistically, in 2007?

Cowboys
Chiefs
Titans
Bears
Commanders
Eagles
Packers
Vikings
Stealers
Colts
Bucs
Patriots
Chargers
Seahawks

According to points per possession, they are in inverted order (worst-to-best).

In terms of stopping the opposing offense from scoring, we performed the worst on that list (not accounting for field position or strength of opponent). If you think that points per game is more important, then you must agree that 11 of those teams were better than we were.

The only two defenses on that list that allowed more points per game than we did were Kansas City and Chicago. The Chiefs' defense allowed a whopping 0.38 more points per game than we did -- or six more points all season. They faced nine more possessions than we did (according to Football Outsiders). They finished 14th in points allowed per game, 13th in yards allowed per game and ninth in yards allowed per possession, so being 13th in points allowed per possession shouldn't be surprising. They also finished No. 1 in third-down percentage allowed and No. 2 in red zone touchdown percentage allowed (37 percent, compared to 50 percent for us).

Chicago, on the other hand, was a classic case of a pretty good defense being saddled with a horrific offense. The Bears faced 196 legitimate possessions (we faced 169) and allowed 1.67 points per true possession (we allowed 1.72). We had only one game when our opponent had more than 12 legitimate possessions (one game of 13) -- the Bears had six (two of 13, two of 14 and two of 16). Our range of possessions faced was 8 to 13, and we allowed 18.19 points per game. In the Bears' 12 games when the opponent had 8 to 13 possessions, they allowed 17.83 points per game. In the four games when they faced 14 or 16 possessions, they allowed an average of 28.25 points per game. There's no doubt that they would have allowed a lot fewer total points and points per game if their offense had been able to sustain drives.

And by the way, our defense finished 13th in points allowed per game.

 

[AdamJT13]

That's right. As I have always said, you can't project a 10-possession performance into a 15-possession game. It changes the nature of each possession. It means it's shorter.

Not necessarily. They're only shorter if the defense forces them to be shorter.

And their not MY expected point totals, they're the ones you throw around every time this topic comes up. You say (said?) 25% more possessions means 25% more points--now you're amending that to say "IF performance level is the same."

I haven't amended anything. I've been saying the same thing all along.

My whole point has been that it's about performance, not counting possessions. An offense that scores more points on more possessions does so because of how it performed, not because of how many possessions it had.

So, you would look at the examples I posted above -- Offense A, Offense B, Defense X and Defense Y -- and determine that which ones performed better?

Your answers will tell me all that I need to know.

In case you missed them the first time --

Offense A's possessions
5 plays, 14 yards, PUNT
3 plays, 20 yards, TOUCHDOWN
3 plays, 8 yards, PUNT
3 plays, 15 yards (plus 13 in penalties), TOUCHDOWN
3 plays, 36 yards, TOUCHDOWN
3 plays, minus-4 yards, PUNT
6 plays, 50 yards, TOUCHDOWN
5 plays, 44 yards, TOUCHDOWN
3 plays, 6 yards, PUNT
7 plays, 52 yards, FIELD GOAL
3 plays, 4 yards, PUNT
3 plays, minus-1 yards, PUNT
6 plays, 14 yards, INTERCEPTION
6 plays, 25 yards, PUNT


Offense B's possessions
4 plays, 96 yards, TOUCHDOWN
13 plays, 74 yards, FIELD GOAL
5 plays, 65 yards, TOUCHDOWN
10 plays, 78 yards, TOUCHDOWN
10 plays, 69 yards, TOUCHDOWN
3 plays, minus-7 yards, PUNT
13 plays, 69 yards, FIELD GOAL

Defense X's possessions faced
1 play, 0 yards, INTERCEPTION
5 plays, 65 yards, TOUCHDOWN
3 plays, 4 yards, INTERCEPTION
7 plays, 53 yards, FIELD GOAL
3 plays, 3 yards, PUNT
4 plays, 60 yards, TOUCHDOWN
13 plays, 49 yards, TURNOVER ON DOWNS
4 plays, 40 yards, TOUCHDOWN
3 plays, 0 yards, PUNT
7 plays, 32 yards, MISSED FIELD GOAL
3 plays, 3 yards, PUNT
5 plays, 20 yards, INTERCEPTION
3 plays, 9 yards, PUNT
3 plays, 6 yards, PUNT
7 plays, 16 yards, TURNOVER ON DOWNS
3 plays, 3 yards, INTERCEPTION


Defense Y's possessions faced
6 plays, 21 yards, FIELD GOAL
5 plays, 49 yards, TOUCHDOWN
3 plays, 5 yards, PUNT
9 plays, 51 yards, FIELD GOAL
3 plays, minus-10 yards, PUNT
13 plays, 71 yards, FIELD GOAL
9 plays, 93 yards, TOUCHDOWN

Which offense performed better, and which defense performed better?

 

[Alexander]

Instead of dueling banjoes, this is dueling slide rules.

 

[percyhoward]

Your answers will tell me all that I need to know.

All you need to know about those four scenarios, you mean. The defenses that performed better were the ones that per-possession stats says performed better--in those examples, which don't really put the stat to the test.

My question is, on average, in REAL games, do teams score more points in games when they get more possessions?

Just take the middle 10 teams (12th-21st, whoever they were) in offensive scoring from last season, and compare how many points and possessions they got in each game. If per-possession stats really work, the points should go up in proportion to the possessions.