The Real Problem With Our Defense Last Year
When watching the games last year, I was often left with the impression that the biggest problem with our defense was our offense. I would love to see some stats on average starting field position for opposing offenses, because I feel like our worst moments on defense always came after the offense coughed up the ball in a bad position.
According to http://www.cfbstats.com, we were ranked 84th in scoring defense and 67th in yards allowed. Again, I wish I had stats on average starting field position, but these two stats seem to indicate that our defense was better on a down-to-down basis than it was at preventing points. Maybe you can explain the disparity in scoring defense versus yards allowed by saying we gave up tons of TDs instead of FGs, but there were 14 teams above us in total scoring defense that allowed more TDs. I feel like the most likely explanation for this disparity is the fact that our offense put us in a terrible position so often. This seems to gel with my own impression of the games last year (like the Notre Dame game, for example).
Hopefully the offense turns around in terms of moving the chains and taking care of the ball. If it does, I think we might be able to expect more improvement from our defense (and in our overall record) than your average fan is predicting.
although this has been discussed previously. I can't recall seeing the stat you mentioned on this blog before (really the only M site I regularly visit) but I am as sure as I can be without the stats in front of me that our awful offense contributed to our merely very bad defense. Another stat I'd like to see was the offense' average time of possession relative to the opponent, which is another factor that would definitely hamper a defense.
As to this year, yes, the offense will be better, and the defense may/will benefit (rapidly knocking on wood).
EDIT: I can't ever recall seeing so many diaries so rapidly posted on this blog as in the last couple of weeks. Must be a point thing.
I'd be careful asking for time of possession on this blog. Brian isn't a big fan of it... http://mgoblog.com/content/mathy-mailbag and http://mgoblog.com/content/mailbag-40
And people here (including me) tend to follow Brian's leads. Which may be why we're getting diaries like this: http://mgoblog.com/diaries/useless-stats
I do agree that a lot of diaries are being made quite quickly. This is partly because there's no point limit to post a diary like this is to post a forum topic.
brian has it backwards. the point requirement should be for diaries. Especially considering some of the garbage people post on the board (no offense). I wanted to post something today but I couldn't because it wasn't diary material. So it goes...
I believe that Brian is working on revising the point system have it do more work for him and us than it currently does. I'd be willing to bet that putting a point minimum before you can post a diary will be something that he adds.
We have to remember two things, first that Brian is already busy creating content which is the real reason that we all come here. We also need to remember that this is Brian's primary source of income, so it's in his best interest to increase the number of clicks on his site. Thus, it may not be the best idea monetarily to piss off new people by forcing them to reach 20 points before they can post. Less people means less clicks means less money means less happy Brian.
I'm sorry that you couldn't post what you wanted to post. As of right now you just need 2 more points to be able to post on the board. You'll get those soon I'm sure. And I do agree, there have been a fair few diary entries that probably should've been board entries.
It seems to me that a good metric besides the length of the entry is the point of the entry. If you merely want to post something to have a discussion then the board is probably the best place. But if you want to inform the community of something and have that be the primary purpose of the diary (the giving of information from the writer to the reader) then that's probably worthy of a dairy. In my mind, people should air on the side of posting on the board. If something is truly diary worthy then people will probably say that in the discussion on the board and it can then be rewritten and expanded based on the discussion from the board and made into a dairy entry.
But that's just my (off-topic) two cents.
If I remember correctly, diaries are also meant to be used as a user's mini-blog, so anything goes on the wild blue internet where that is concerned. So the whole idea of being "Dairy worthy" is subjective given that criteria.
I see your point in that we can put anything in a diary since its "ours" and the internet has everything man has ever thought of on it.
But, this is a Michigan sports blog. There's a reason that people label non-Michigan sports topics "OT" on the board. I understand that there is a definite difference between the board and diaries in that the board is meant for discussion whereas diaries aren't necessarily meant for discussion. And since we're all Michigan sports fans, the discussion should probably primarily be about Michigan sports, or least that is the definition of "on topic" subjects for the board. Diaries don't necessarily have that restriction.
But, diaries are very prominent on this site. They're at the very top of the page when you first open it, you're bound to see them. And since most people are looking for Michigan related sports thing, it's somewhat odd when you see diaries discussing Dave Dombrowski (http://mgoblog.com/diaries/ot-evaluation-dave-dombrowski) or other things that aren't at all related to Michigan sports. That sort of thing can cause a fanbase to start to split and fracture (as the very issue itself is currently doing). So, I can see the value in defining what "should" go in diaries and what "shouldn't". I admit that my argument holds almost no water if the diaries weren't so prominent.
I'd think of the diaries as less "user's mini-blog" than a "user's mini-MGoBlog". Otherwise why would Brian have forcefully changed the date on the "OT: On Florida" (http://mgoblog.com/diaries/ot-florida) diary to get it off the front page of diaries? If anything goes, then why didn't that? Just because it was inflammatory and most of the users hated it? Although that definitely was part of the reason, I think a big reason that a lot of people hated it was because it didn't belong here on this blog in that form.
Eh, now I've started to rant so I'll just stop now. But I will repeat that I think the diaries are mini MGoBlogs rather than just mini blogs. Again, that's just my opinion and I'm nobody.
Although I respect the hell out of Brian (which is why most if us are here I'd imagine) your link and his previous comments on time of possession (that I recall) are far from definitive IMO on that subject. Though I do agree that TOP as a stat is generally overrated, it's incorrect IMO to deny that in general, other things being equal, the longer your defense is on the field, the greater the chances are that they will be scored on, and the lower the chances are that your offense will score.
On TOP, its a double-edged sword. Yes, the offense could help by not going 3-and-out, but the defense could have helped their cause, by, you know, not letting up 99.6% of 3rd down conversion.
potato salad
after that mini-thread.
they talk about this a little bit in the hail to the victors
From what I remember, our worst moments on defense came on 3rd and long. Bad field position can account for a minor decline in defense, but when all you have to do is stop them from getting 17 yards and they get 17 yards, that's not due to field position.
I'd like to make excuses for them, but the secondary underperformed in a major way last year, and I don't see our safeties getting better this year.
I am on the fence as to whether our third down conversion defense in general is a huge factor in our wins (if any amateur statisticians want to take a crack at some sort of regression analysis, be my guest).
For context, our 38.78% defensive 3rd down conversion rate was 60th nationally and 7th in Big Ten (PSU was tops in the big ten at 32.36%, FSU at 25.75% was tops nationally). The mean national conversion % was 39.04% (37.60% in the big ten). We were very close to the median both in conference and nationally.
In our three wins, we allowed Miami to convert 42% of third downs (and there were a lot of third downs), Wisconsin 32%, and Minnesota 29%. Our percentage against Minnesota was good, obviously, compared with our season average of 39%, as was our percentage against Wisconsin. Miami actually converted an above-average percent of their 3rd downs but couldn't find the end zone.
The four-loss stretch from Toledo through Minnesota was certainly pretty ugly (44, 50, 50, and 47%), but we held Toledo to only 13 points. I am just not feeling a strong connection.
I also cobbled together some stats on 3rd and long based on the defense's situational stats for rushing/passing (links below). On 3rd and 7-9 yards to go, we let up 12 conversions in 41 attempts (29%). On 3rd and 10+ we let up 12 conversions in 50 attempts (24%). I don't have a lot of context for those numbers, but PSU (which led the conference in overall 3rd down defensive conversion %) was 7/33 on 3rd and 7-9 (21%) and 5/48 (10%) on 3rd and 10+. We are fairly far off the league-leading pace on those numbers, so maybe you have a point about third and long.
http://www.cfbstats.com/2008/team/418/thirddown/defense/gamelog.html
Game Result Attempts/Converted (%)
2 Utah L 23-25 5 of 19 (26.32)
Miami (Ohio) W 16-6 8 of 19 (42.11)
@ Notre Dame L 17-35 3 of 12 (25.00)
Wisconsin W 27-25 6 of 19 (31.58)
Illinois L 20-45 6 of 15 (40.00)
Toledo L 10-13 8 of 18 (44.44)
@ 8 Penn St. L 17-46 7 of 14 (50.00)
24 Michigan St. L 21-35 9 of 18 (50.00)
@ Purdue L 42-48 8 of 17 (47.06)
@ Minnesota W 29-6 4 of 14 (28.57)
Northwestern L 14-21 6 of 17 (35.29)
@ 9 Ohio St. L 7-42 6 of 14 (42.86)
TOTAL 196 of 76 (38.78)
http://www.cfbstats.com/2008/team/418/passing/defense/situational.html http://www.cfbstats.com/2008/team/418/rushing/defense/situational.html
Also, if someone could let me know how to paste an actual table into a post, that would be greatly appreciated.
This is not diary worthy.
It seems very few "diaries" have been worthy lately.
Let me continue to parse the UFR data that was given yesterday, and I will try to answer some of your questions.
Uh, even if field position played a role, 67th in yards allowed is still pretty bad.
That's certainly average, at best, but part of that is also the amount of possessions afforded opposing offenses. If our offense was frequently off the field in 3 downs (or close to it) or never saw the field because we fumbled a punt, that leaves more opportunities for the opposing offense to wrack up yards. I am not saying we had an elite defense last year, but I still think there's some hope for next year if our offense can even approach mediocrity.
It's not really "average". If you take it as where we ranked nationally, it's middle of the pack, but if you look at BCS-conference schools, only a handful were worse:
Louisville
Texas Tech
Arkansas
Minnesota
Stanford
Colorado
Oregon
NC State
Oklahoma St.
Kansas
Missouri
Syracuse
Washington
That puts us in the 22nd percentile of BCS conference teams in total defense in terms of total yards allowed. It's an important distinction because the number that compares us to all of D-I compares us to a whole host of teams whose only function, really, is to be a doormat.
Further, in the Big Ten, our defense was 9th of 11 in yards allowed - again, showing a severe lacking compared to our peers.
The average start on the opponents scoring drives was their own 45 yards line. They do have the play by play pdfs at mgoblue if someone wants to calculate it for the whole season.
Starting field position on the 45 is brutal (I wonder what the national average was). What is field goal range for your average college kicker? 40 yards? 45? That means you only need to get 27-32 yards to be in field goal range. I sure hope that Brian's theory about lost fumbles being essentially random is correct. We could really use a break towards the mean this year.
was never on the field long enough, which tired the defense. As a result the D gave up too many big plays.
The defense giving up big plays has been a problem for a few years now. Not disagreeing that the defense might have been tired from being on the field to long, but the defense giving up big plays didn't just become a problem last year.
I think our D was average as a whole, but the offense kept them on the field too much. This gave the opposing team more chances to expose our weak links and wear our D down. Thus, making them look worse then I think they actually were.
is important as well. Not only because it gives the opposing side more time to score but it puts the defence in a horrible position as far as getting time to breath between a given series. I don't care how good Barwis is, last year it wouldn't have made a difference if he was on the field himself playing defense, he would have been winded. It is a heck of a lot easier to score on a tired defense than a rested one obviously.
I can't count how many times I saw the defense make a nice stand only to come off the field to watch yet another three and out. Not only does that add to their exhaustion level, but it is quite disheartening when you play your ass off and make some nice plays on D and then watch that offense go out and make a mess of things again.
I won't say our defense was great last year, but I think it was much better than the stats indicate.
Our average time of possession ranking from last year was 110th nationally. The only teams worse than us were:
Record Team
5-7 Florida International
5-7 Temple
8-5 Houston
3-9 Syracuse
8-5 North Carolina
3-9 Indiana
1-11 SMU
2-10 San Diego State
10-4 Missouri
10-3 Oregon
http://www.cfbstats.com/2008/leader/national/team/offense/split01/categ…
Obviously that's a pretty good mix of awful, average, and good. Two very average teams - Arkansas State (6-6) and Memphis (6-7) - were both in the top five.
I wish they tracked time of possession by either your number of plays on the field or the actual time you posses the ball (as opposed to the game clock time). I feel like incomplete passes distort the statistic quite a bit. It might be a useful metric for games between two teams that run a lot (or have similar pass/run balance), but it doesn't seem to be that reliable of an indicator of success in general.
Again, though, TOP is not a purely offensive stat. Our offense didn't help, but our defense could have helped themselves by getting off of the field.
Comments