A comparative look at Cal's defense

by HolmoePhobe on Nov 1, 2009 10:18 PM PST Comment 15 comments

There's been a fair amount of discussion recently regarding Cal's defense, including much citing of statistics, and while the discussions have been very interesting, I feel like we might be approaching some of these statistics in the wrong way. The usage of statistics seems to fall mostly into the following two categories, both of which are problematic:

1. Looking at statistics in a vacuum - e.g. "Wazzou passed for 350 yards against us, so our secondary is doing badly".
--> Hypothetically, Wazzou could be passing for 400+ yards against everyone else. We know that this isn't the case, but the point is that the frame of reference is somewhat arbitrary/subjective.

2. Looking at statistics in terms of Pac-10 ranking - e.g. "Cal is second in the Pac-10 in points per game".
--> Hypothetically, the next eight Pac-10 teams could just have really terrible offenses. Also, this fails to account at all for strength of schedule, which is definitely relevant to a team's statistics.

For the purposes of this fanpost, I decided to use a slightly different method of evaluating Cal's defense: comparing opponents' performance against us to their average performance this season. In other words, instead of looking at whether our defense performed better or worse than average against each opponent, I'm looking at whether our defense made opposing offenses perform better or worse than they usually do. This isn't a particularly original idea, and it has its own flaws (again, it doesn't take schedule into account at all), but it has the virtue of demonstrating in what manner opposing offenses are able to perform well against us, compared to their other performances.

Continue reading this post »

Data is from ESPN and cfbstats.com. Note that a few of the categories do not include last week's games. The categories are also somewhat arbitrary. My goal is to get a sense of whether any statistics in particular jump out at us. If there is interest, I can put together a more detailed version of this post later on.

The index below is the ratio of how an opposing offense performed against Cal to how they normally perform, i.e. a higher index means they did better against us than they usually do; a lower index means that they did worse against us than they usually do.

MARYLAND	Average	vs. Cal	Index
Points	22.83	13	0.57
Total YPG	328.83	303	0.92

Rushing YPG	90.43	136	1.5
Rushing YPA	2.73	3.16	1.16

Passing YPG	237.97	167	0.7
Passing YPA	6.97	5.6	0.8
Completion %	59.1	56.7	0.96
Passer Rating	124.42	103.43	0.83

First Downs	16.2	17	1.05
3rd Down %	0.4	0.24	0.6
Red Zone %	0.87	1	1.15
Turnover Margin	-1.43	-2	x

(This game was the reverse of all the others in terms of rushing/passing indices, but Maryland is a horrible team, so I wouldn't read too much into it.)

MINNESOTA	Average	vs. Cal	Index
Points	23.7	21	0.89
Total YPG	322.09	270	0.84

Rushing YPG	112.38	37	0.33
Rushing YPA	3.76	1.76	0.47

Passing YPG	209.71	233	1.11
Passing YPA	6.68	6.9	1.03
Completion %	53.79	64.7	1.2
Passer Rating	110.98	133.75	1.21

First Downs	15.24	14	0.92
3rd Down %	34.86	33.33	0.96
Red Zone %	0.83	1	1.2
Turnover Margin	-0.19	-3	x

OREGON	Average	vs. Cal	Index
Points	34.69	42	1.21
Total YPG	380.91	524	1.38

Rushing YPG	232.91	236	1.01
Rushing YPA	4.92	4.54	0.92

Passing YPG	148	288	1.95
Passing YPA	6.49	9	1.39
Completion %	57.99	78.1	1.35
Passer Rating	117.79	184.67	1.57

First Downs	19.14	26	1.36
3rd Down %	31.05	46.67	1.5
Red Zone %	0.91	1	1.1
Turnover Margin	0.95	-1	x

(This is one of those cases where the stats don't show how horrible the beat-down was...also Oregon has been beating a lot of other teams down as well, which probably explains the relatively low indices.)

USC	Average	vs. Cal	Index
Points	30	30	1
Total YPG	421.91	457	1.08

Rushing YPG	193.43	174	0.9
Rushing YPA	5.63	4.14	0.74

Passing YPG	228.49	283	1.24
Passing YPA	9.13	8.1	0.89
Completion %	61.1	57.1	0.93
Passer Rating	147.56	119.35	0.81

First Downs	18.63	20	1.07
3rd Down %	32.8	40	1.22
Red Zone %	87.3	100	1.15
Turnover Margin	-0.63	1	x

UCLA	Average	vs. Cal	Index
Points	19.03	26	1.37
Total YPG	285.6	448	1.57

Rushing YPG	97	137	1.41
Rushing YPA	3.29	5.27	1.6

Passing YPG	188.6	311	1.65
Passing YPA	5.57	7.4	1.33
Completion %	54.11	50	0.92
Passer Rating	100.86	107.44	1.07

First Downs	15.86	17	1.07
3rd Down %	29.25	31.25	1.07
Red Zone %	80.95	100	1.24
Turnover Margin	0.78	-2	x

WSU	Average	vs. Cal	Index
Points	14.83	17	1.15
Total YPG	260.34	440	1.69

Rushing YPG	79.34	55	0.69
Rushing YPA	2.27	1.96	0.86

Passing YPG	181	385	2.13
Passing YPA	5.91	9	1.52
Completion %	52.66	67.4	1.28
Passer Rating	101.29	158	1.56

First Downs	14.29	20	1.4
3rd Down %	22.04	35.29	1.6
Red Zone %	73.53	50	0.68
Turnover Margin	-0.3	1	x

(Those passing numbers are kind of embarrassing...)

ASU	Average	vs. Cal	Index
Points	26.7	21	0.79
Total YPG	354.6	329	0.93

Rushing YPG	132.8	82	0.62
Rushing YPA	4.48	3.2	0.71

Passing YPG	221.9	247	1.11
Passing YPA	6.1	8.2	1.34
Completion %	53.6	0.57	0.01
Passer Rating	112.31	134.49	1.2

First Downs	18.9	15	0.79
3rd Down %	29	0.33	0.01
Red Zone %	86.21	0.75	0.01
Turnover Margin	1.14	-1	x

Finally, here are the index "averages" - note that these are averaged across games, not across plays (i.e. so the average YPA would be the average of each game's YPA, not the average of all attempts over the season).

OVERALL	Index
Points	1
Total YPG	1.2

Rushing YPG	0.92
Rushing YPA	0.92

Passing YPG	1.41
Passing YPA	1.19
Completion %	0.95
Passer Rating	1.18

First Downs	1.1
3rd Down %	0.99
Red Zone %	0.93
Turnover Margin	x

Again, this is a pretty unscientific way of looking at the numbers - but I think it provides some context to the discussion.

Thoughts? Criticisms? Suggestions? Honkings?

The opinions expressed in a FanPost are, in every way, reflective of the opinions of every California Golden Blogs Marshawnthusiast. Moreover, they are reflective of every employee of SBNation, including Tyler "Blez" Bleszinski.

6 recs | Comment 15 comments

Story-email Email Printer Print

Comments

This is really cool, HP. And I think that the stats you come out with at the end mesh with how I’ve perceived the defense.

We’re a very average unit, which is disappointing based on pre-season expectations but perhaps unsurprising considering the losses and lineback (and Rulon, who we seem to be missing more than I anticipated).

We’re better against the run than the pass, but not by a huge margin.

Also, I found these numbers interesting: Passing YPG – 1.41 vs. Passing YPA – 1.19. This indicates that Cal, on a per-attempt basis is slightly below average, but not nearly as bad as everyone thinks. Our YPG stats are inflated because in 5 of Cal’s wins we’ve jumped out to quick leads and forced our opponent to throw to catch up. But on a per-play basis Cal isn’t disastrous. Still, I’d like to see improvement over the course of the season, which can be possible with Hill getting more experience in his first year.

The #1 greatest threat to America: BEARS

by norcalnick on Nov 1, 2009 10:35 PM PST reply actions 0 recs

I feel like at the end of this I should be taking some kind of quiz.

CGB: Resisting Hope Since 2007

by Spazzy Mcgee on Nov 2, 2009 12:22 AM PST reply actions 0 recs

Thanks for doing this. This is interesting. So Cal Defense turns in some good games, and some not so good games. Not wholly unexpected. To me, the most concerning stat is the indices for YPA (pass) and Passer rating. I guess it isn’t totally unexpected though as since Cal stops the run first, knowingly incurring the risk of giving up on pass; hence why the YPA (pass) and passer rating are a bit higher.

www.californiagoldenblogs.com

by HydroTech on Nov 2, 2009 12:29 AM PST reply actions 0 recs

I thought the fact that passing YPG was substantially higher than YPA/rating/completion % was interesting…as it would seem to indicate that opposing QBs are putting up “big numbers” against us at least partially because they’re throwing more, rather than throwing better.

dboneisloose

by HolmoePhobe on Nov 2, 2009 9:53 AM PST up reply actions 0 recs

It was my understanding there would be no math in this debate...

Seriously, this is a very interesting post. From the indexes, the numbers appear to bear out that we give up an inordinate amount of passing yards. All in all, my reaction is that this shows we’re an average defense.

Praise be to Tedford!

by Ohio Bear on Nov 2, 2009 5:35 AM PST reply actions 0 recs

Yeah, that’s pretty much what I came away with. The pass D is a little weak, but passing numbers against us may be inflated by a higher number of attempts. Maybe I’ll do a post on run-pass balance against our defense next.

dboneisloose

by HolmoePhobe on Nov 2, 2009 9:56 AM PST up reply actions 0 recs

Nicely done. The index calculation makes it easy to see what is going on.

This basically validates the perception that we are a bend-don’t-break defense (average yards > 1, points = 1) and that we give up most of those yards through the air (1.4!).

I’m struggling to understand how what was supposed to be a strength, our DBs, has turned out to be a liability. Obviously pass D is a team effort and the LBs and DL have to take some of the blame, but it seems like all of last year’s starters have regressed.

Syd has been very good, but perhaps slightly below last year, and also injured.

Hagan has been a total disappointment after a great season last year.

Ezeff, a 2-year starter, has lost his job.

Johnson, a hard hitter but seems to take bad angles.

At least Hill seems to be improving and is getting valuable reps. Cattouse also looks very good.

GOLD OUT MOZAMBIQUE!

by OskiMonsta on Nov 2, 2009 6:38 AM PST reply actions 0 recs

This basically validates the perception that we are a bend-don’t-break defense

Or, alternately, a team that’s built a lot better for red zone defense than full-field defense.

I think the discrepancy comes from the geometry of the game. When an offense has the full field in front of it its options include the short pass/rush, the intermediate pass, and the deep pass, only two of which Cal has the personnel to cover effectively (the rush/short pass and the deep pass). Opposing offenses take advantage Cal’s inability to defend the intermediate pass until they get to the red zone where the offense only has as many options as Cal can cover.

It also helps that we’ve played a bunch of games against shitty red zone defenses.

"Let me tell you a story. I was a political prisoner for two years. The instant I was released I ran to McDonald's. I had a Big Mac and a Coke.

It was fantastic."
-Toyama Koichi, US Presidential candidate from Japan
http://www.youtube.com/watch?v=uGZqOkeYbB0

by AERose on Nov 2, 2009 12:16 PM PST up reply actions 0 recs

nice breakdown!

the passing thing is really a two fold quasi-issue. 1) in 6 out of 8 games we’ve scored first and built up good leads in the first quarter, forcing the other team into a bit (if not full) disaster catch up mode.

and 2) we’ve usually always focused on stopping/controlling the run first.

The yardage challenge and interception challenges we’ve seen from last year are I think mostly related to pass rushing. we’ve really struggled to put much rush on with only 3/4 rushers. Should we change and rush 5 more? I don’t know.

Go Bears Go

by Rocksanddirt on Nov 2, 2009 8:26 AM PST reply actions 0 recs

Beat me to it

100% hit the nail on the head here – while Cal has the exact same number through eight games as it did last year (22), 17 have come against Maryland (6), E. Wash (3), Minnesota (3) and WSU (5). This would indicate to me that when Cal can “out-athlete” its opponent we are very capable of creating pressure, but when the offense is capable of either 1) Matching up 1-1 or 2) Scheming up sufficient defenders, we do not have the same ability to make plays as we did with Follett (who contributed a ridiculous 10.5 sacks last year). Also, having a threat like Follett around means you have to scheme to stop him which leaves players like Alualu and Cameron Jordan free to rush 1-1, which is much tougher on opposing o-lineman.

Also, so far this year we have generated 9 QB hurries in eight games compared to 20 last year for the season, so there is room for improvement.

While it would require massive amounts of film viewing, we could generate an idea of how many interceptions, pass break-ups and incompletions were the result of pressure in the backfield – remove that pressure and we may see the same level of defense as we did last year.

Alternatively, it bears (zing!) examining that last years numbers may have been inflated as the result of playing bad QB’s (or in the case of UCLA – really, REALLY, epically bad QB’s). Looking at the numbers, WSU (4), Colorado St. (3), UCLA (4) and Stanfurd (3) generated 14 of the teams 24 INTs last year.

by CalFanNY on Nov 2, 2009 1:25 PM PST up reply actions 0 recs

In the 3-4, you need at least one of the LBs to become a serious pass-rushing threat if you are going to use it on passing downs. Follett was that guy last year, and his play meant that we didn’t have to bother substituting a lot — he could rush the passer and also hold up decently in coverage and run support.

None of the LBs has stepped up as a consistent threat this year; too often they just run into an O-lineman’s chest when coming on a blitz instead of finding a hole. A few have shown flashes here and there, but no one has really stepped up to be The Guy. It’s looking like Gregory is slowly adjusting to this season’s reality; first he tried shuffling the LBs from inside to outside and vice versa, experimenting to see if any of them clicked. Now, he’s trying to go with 4 DL in nickel packages to generate more pass rush. It seems to be getting a little better.

It does look to me like the secondary has also regressed a bit. Soft zones are a good explanation for why teams are able to check down and dunk-and-dunk us for long drives, but the long pass plays given up in the last few games are fully on the players for blown assignments. BBDB is based on the idea that if you force a college offense to sustain a long drive, they will make a mistake before scoring a TD most of the time. That’s pretty much out the window, though, if you’re giving up 80-yard TD passes.

by sycasey on Nov 3, 2009 12:02 PM PST up reply actions 0 recs

Wasn’t Price supposed to be that pass rush threat as a hybrid down lineman, pass rushing linebacker?

by LeonPowe on Nov 5, 2009 6:25 AM PST up reply actions 0 recs

Problem is that he’s a little small to play every down. He wouldn’t help much in run support, as Follett could.

Also, based solely on observation, Price looks pretty good as a speed rusher trying to get outside the tackle. However, if that’s his only trick, it’s pretty easily stopped. Follett would speed rush to the outside sometimes, but could also power through the inside if the tackle tried cheating out against him; Price doesn’t quite have the recognition yet (in fact, none of the LBs seem to have it).

by sycasey on Nov 5, 2009 11:56 AM PST up reply actions 1 recs