FOA 2012 Preview: Strategic Tendencies Tables

by Aaron Schatz

I've spent the last few days working on the "Strategic Tendencies" comments for the team chapters in Football Outsiders Almanac 2012, and I wanted to share with everyone some of the cool new information we have for you. This year for the first time, we had game charters tracking defensive as well as offensive formations. In addition, after the season was over we went through and marked each play not just for offensive "formation" (where players are lined up) but also for offensive "personnel" (who is actually in the game, which is a better guide to how teams judge things).

We'll have this table in the book for all 32 teams, but we present the Super Bowl champion New York Giants here as a sample. Personnel is listed using the two-digit standard, where the first digit is running backs and the second digit is tight ends. Subtract both from five, and you get the number of wide receivers. We've got the five most common personnel groups for each team, and the five most common defensive formations, on both sides of the ball. Some of the less common defensive formations have been grouped together, so everything with six or more defensive backs is "Dime+" and everything with fewer than four defensive backs except for 4-4-3 is "Goal Line." There will be tables in the book with leaguewide averages from every formation and personnel group.

Also, note these DVOA ratings are the new DVOA v7.0 which will be introduced in the book, now normalized so every year averages 0.0%.

Giants Offense Giants Offense vs. Opponents Giants Defense Giants Defense vs. Opponents
Pers Freq Yds DVOA Run% Pers Freq Yds DVOA Pers Freq Yds DVOA Pers Freq Yds DVOA
11 44% 7.0 19.5% 18% 3-4-4 28% 5.2 -0.1% 4-2-5 59% 6.4 5.3% 11 43% 5.9 7.3%
21 27% 5.1 6.8% 52% 4-2-5 23% 6.9 15.8% 4-3-4 28% 4.7 -1.4% 12 20% 5.9 -2.4%
12 13% 6.4 16.0% 50% 4-3-4 23% 6.2 23.3% Dime+ 6% 5.6 6.8% 21 17% 5.8 18.2%
22 4% 6.5 15.2% 77% Dime+ 9% 8.0 33.6% 3-3-5 5% 5.6 -12.0% 22 5% 3.7 -5.5%
10 2% 7.7 60.4% 4% 2-4-5 8% 8.3 60.4% Goal Line 1% 1.1 23.1% 10 3% 7.7 20.5%

The Giants, as most people know, were pretty unique in their high usage of a 4-2-5 formation with three safeties. You may notice that the same five personnel groups are listed for both offense and "against Giants defense," but these will actually be different for most teams. The Patriots' top five personnel groups, for example, include two with six offensive linemen.

Next we have the Strategic Tendencies table that we've run in the book every year since 2006. We've made a few changes this year, so I thought I would share this one with you too. Again, this one is for the New York Giants.





Pass Rush



Runs, all plays 39% 24 Form: Single Back 60% 19 Rush 3 6.7% 15 4 DB 28% 31
Runs, first half 37% 28 Form: Empty Back 3% 28 Rush 4 63.3% 18 5 DB 64% 1
Runs, first down 43% 31 Pers: 3+ WR 49% 20 Rush 5 20.8% 17 6+ DB 6% 18
Runs, second-long 43% 8 Pers: 4+ WR 3% 16 Rush 6+ 9.1% 10 CB by Sides 53% 31
Runs, power sit. 58% 16 Pers: 2+ TE/6+ OL 25% 29 Zone Blitz 2.9% 25 Go for it on 4th 1.05 7
Runs, behind 2H 33% 9 Play action 14% 29 Sacks by LB 11.5% 26 Offensive Pace 30.2 12
Pass, ahead 2H 48% 6 Max protect 11% 16 Sacks by DB 2.1% 27 Defensive Pace 30.1 8

I'm not going to go super into detail on explanation the way we do in the book's "Statistical Toolbox" section, but a few definitions:

  • In the second column of stats, the first two lines represent "formation" and the next three represent "personnel." This is different from previous years where similar data was given only based on formations.
  • "Max protect" means that blockers outnumber pass rushers by at least two, with a minimum of seven blockers.
  • "CB by sides" is new for this year. It measures how often a team's top two cornerbacks play on one side of the field compared to the other side. It should give you a sense of which teams move cornerbacks around (like both New York teams) and which teams (Seattle, Atlanta, Philadelphia) like to leave cornerbacks always on the same side.
  • "Go for it on 4th" is our Aggressiveness Index stat. Higher means more aggressive on fourth downs, and 1.0 is the league average over the last few years, although the actual average in 2011 was .92.
  • "Pace" is based on situation-neutral pace, the number of seconds of game time that pass between plays when the score is within certain guidelines in the first three quarters.

I hope everyone enjoys discussing these and they get you excited to see information for all 32 teams in FOA 2012, which will be out in early July. It's just a taste to get you hooked, especially if you want to see some team that isn't the Giants. In the meantime, we also may run some articles with all 32 teams ranked in some of these stats, or in some of our other stats which are not listed here.


16 comments, Last at 17 May 2012, 6:29pm

1 Re: FOA 2012 Preview: Strategic Tendencies Tables

I like the idea of DVOA being normalized to 0 every year. However, some information will be lost if this is all you report. Can you have a "historical" DVOA as well that uses the old baseline?

2 Re: FOA 2012 Preview: Strategic Tendencies Tables

Well to put it bluntly, that is football porn. If it's OK to make a request, can we get access to more that top tens in the appendixes? Even if it has to be published on a webpage that's only available to purchasers of FOA can we see the full stats tables? We know you must have them, give us more.

(And invest in some kindle optimisation, it's the future.)

But still, oooohhh!

5 Re: FOA 2012 Preview: Strategic Tendencies Tables

Does the Pers for offense consider roster position or alignment? When a RB lines up in the slot (or at X or Z), is he a RB or WR? Does motion affect this?

I'd think alignment should be the way to go with RB and roster position with TE, but this seems to be very muddled on many teams, particularly spread offense teams.

6 Re: FOA 2012 Preview: Strategic Tendencies Tables

"Some of the less common defensive formations have been grouped together, so everything with six or more defensive backs is "Dime+" and everything with more than four linemen except for 4-4-3 is "'Goal Line.'"

That last part about "Goal Line" doesn't make sense. Why is the word "except" in there when 4-4-3 is not more than four linemen?

7 Re: FOA 2012 Preview: Strategic Tendencies Tables

This looks great! I look forward to reading through the breakdowns for the other teams as well.

I wonder if you could include things like DVOA by field position, Quarter, Run/Pass distributions in wins vs. losses per down, and DVOA Run and Pass stats (split out) for each personnel group? Not important, just curious.

Keep up the good work, and I look forward to purchasing FOA 2012!

12 Re: FOA 2012 Preview: Strategic Tendencies Tables

Is the Giants use of 5+ DBs a trend that holds with the entire league? Are nicklebacks playing more than 3rd linebackers on every team? Should we start thinking of nicklebacks as starters?

13 Re: FOA 2012 Preview: Strategic Tendencies Tables

Did the Giants absolutely never run a five-wide? Is that why there's no 00 personnel row?

How do you classify plays with 6OL and 7OL sets, i.e., fewer than five eligible receivers (disregarding the QB)?

Click my alias for a link.

16 Re: FOA 2012 Preview: Strategic Tendencies Tables

It's interesting that the Giants seem to do the unexpected well above average in almost all the situations given.

Runs, first down 43% 31
Runs, second-long 43% 8
Runs, behind 2H 33% 9
Pass, ahead 2H 48% 6

They throw on first down, run on 2nd and long, run when behind and pass when ahead.

Seems much like second-level poker strategy, where you know your opponent is expecting an optimal play, so you do the opposite (bet big when you have it, bet small when bluffing).

The one exception is running in power situations, where they were exactly average.

It'll be interesting to see which teams were even more contrarian, which were the most predictable and how strongly that correlates to success.