I think you will get your wish.
You may remember a previous post wherein I showed off some shmancy graphs of third down conversions, though they lacked context and a suitable smoothing method that didn't make Baby Jesus get pissed off about data distortion. Well, more bit hammering has produced something I think is worth showing people. Without further adieu...
Third Down And What-What
A third down conversion rate is a conflation of two pieces of data: the distance you have to go and how frequently you make that distance. The varying distributions of said distance impact the end result so heavily that the stat often presented on televison broadcasts may as well be called "third and second and first down conversion rate." It's not totally useless, but it can be improved upon in a logical and fairly simple way. And perhaps you can learn something about the team you follow or football in general by taking a deeper look into the situation. If this sounds terribly dull to you, I invite you to forget all about this and check out these creepy drawings of NBA players by bizarre little Japanese girls.
Anyone left is undoubtedly hardcore like Quickdraw McGraw, so here we go. When we last left our blogging superhero, he was casting about for a logical way to smooth the wonky data points in individual team's third down rates. The problem: out towards the third-and-long boonies the disproportionate amount of third-and-tens caused an unnatural flattening effect where any smoothing would drag surrounding distances towards the third-and-ten-percentage.
The solution: take the average yardage of each set of data and place the percentage at that mark. So if you had 10 third and tens and 5 third and elevens, whatever conversion percentage you got from that would be placed at 10.33 yards out. Assuming the conversion rates are linear--a good assumption that far out--this provides non-distorting data smoothing.
So, viola(!), the Michigan defense's third down efficiency:
How To Read This Graph: the thick line in the center is the NCAA average, which is not smoothed. Jutting out from the average are the various icebergs that compose an individual team's deviation from the norm. In general you want your defense to be below the line and your offense above it. Red is bad. Green is good. I checked data out to third and 25, but it gets very thin and useless out there. 15 is about the sanity limit.
This is extremely similar to the graph produced by the first iteration of this process, but I feel much better about it with the theoretically non-distorting smoothing. It stands as clear evidence that last year's Michigan defense was in fact subpar when put in third and long situations. Message board yahoos (and cantakerous bloggers), revel in your victory. Perhaps more disturbing, however, was Michigan's far below average performance on third and short despite employing one Gabe Watson.
The second useful piece of information is the average distance faced on third down. Look nyah:
How To Read This Graph: It's the same thing. Thick line == NCAA average. Red and green == individual team's deviation. There's not really a good way to determine what's "good" or "bad" without taking the actual distance into consideration, so the green and red do not flip. Green == above. Red == below.
Note that no attempt has been made to smooth this data, since there's no line to fit it to. You gets what you gets.
Tomorrow I'll highlight some of the more interesting graphs produced by this method; Thursday you'll get a teeny app that you can use to see this data for any team in D-I. (Caveat: due to some member schools not reporting data to the NCAA, the database I've cobbled together is incomplete. This includes four games USC played, including something of minor importance called "The Rose Bowl." Shame on USC. Some negligent person in the athletic department should be fired for ignoring this task... probably updating his blog.)
Anyway, I'll leave you with the offensive graphs:
I actually thought they'd be considerably uglier.
Also: I'm taking requests. I have a database of most plays/drives that happened in D-I last year. If you've got any ideas as to how to use it, I'm listening. If you know SQL and would like to try something out yourself (Bueller? Bueller?), I can give you a public login.
Respect my authorita! And my implausible cup!
Warning: Annoying tic alert. Um... apparently my "blog ethics" post caused Kyle King to dub me "The Lawgiver," which is cool because Judge Dredd is my second favorite obviously terrible Sylvester Stallone movie* and I've always wanted a flimsy pretext to scream "I AM THE LAW" at fellow bloggers, commenters, and innocent passers-by who pick up the wrong cantelope when THE LAW is shopping. This is by way of explanation for future bizarre behavior.
I just want to try it out...
I AM THE LAW!
(* #1 == Demolition Man.)
Fleebin' Christ, man. One thing operating the BlogPoll did was teach a valuable lesson about knocking poll voters when the end result of their poll is seemingly nonsensical: don't. Because the week after you make an impassioned plea to the voters at large about Team X and their undeniable superiority, team X goes out full of vim and vinegar and goes all French Army on you. The business of polling is an exercise in learning just how uncertain college football is. So commentary like this from CFR--most notable for contributing hilariously bad photoshop banners to the CFB blogosphere--is silly:
This was the problem I had with the well intentioned but ultimately flawed BlogPoll. It had the potential to deliver weekly outcomes that differed from the highly predictable AP poll. Instead, within a few weeks it was lock-step with the AP poll, offering little deviation.
... especially when you consider that the BlogPoll's ranking methodology was like so:
They were soon thinking like and using the same logic of AP voters, mostly slotting teams by record and perceived conference and schedule worth.
It's unfortunate that voters couldn't look past those "record" and "schedule strength" red herrings. Lord knows that we criminally underrated so many teams with flashy new wave offenses because they did things like lose by thirty. We did have an advocate of Gang-of-Six football on board in He Is... Manpundit(!), but the results of the first week were so thoroughly embarrassing that instead of submitting a saner ballot with Boise State conspiciously omitted from the top ten he stopped voting altogether.
I don't really know that this was a good year to deviate from the AP poll. Texas and USC went wire-to-wire as #1 and #2--though they flipped spots in the final poll. As you progressed further down the poll during the year you found a ton of teams that were mediocre or unpredictable week to week or both. And how, exactly, is a poll with somewhere from 35-50 voters--depending on how motivated people were week-to-week--supposed to reflect the wingnut flyer picks that are by definition against the conventional wisdom? It's not like voters all decided to put Alabama #8 in the middle of the season. A snapshot of the proceedings shows disagreement that averages out to... about #8.
For the record, I had hoped that the BlogPoll would be radically different and clearly superior to the stupid MSM StupidPoll [/childish blogger snark] and was disappointed in midseason when this turned out to be wrong. I got over it, though, because it occurred to me that the thing was working as it was because of its structure, and that any selection of a goodly number of people who paid close attention to college football (and--this is important--realized teams play defense too) would spit out an end result pretty close to the AP poll. The BlogPoll followed the structure of the aforementioned exactly: once a week list the top 25 teams in the nation, in order, even though you really have no idea whatsoever if Georgia Tech is #17 or #18. Then let the average iron out the noise.
Where the BlogPoll differed was in its philosophy--talk amongst yourselves and try to understand the POV of others--and its openness--do something dumb and yes, we will find you. This had some effect--witness Wisconsin's final placement after I asked voters to slot them in front of an Auburn team they had just spanked--but in the end, a top 25 poll is a top 25 poll is a top 25 poll. Maybe in a few years a culture of payin' attention and doing something different will grow mossy from the rock shaped by the 1, 2, 3... 25 order that polls are historically locked into, but that takes time. Hopefully not as much time as the Seneca Falls Declaration.
In that vein, however, let it be known that I'm dissatisfied with the current structure of the poll and should a majority of pollsters support or at least tolerate a change to something more flexible, I am all for it. Ideas I have kicked around:
- Dropping the straight ranking thing in favor of assignment of points. You get er... (25 + 1) * 25 /2 = 325 points to distribute to anywhere from 20 to 30 teams. Max points per team: 25. One team must be denoted a nominal number one even if you give the teams at the top an identical point total. If you want to do straight up 1-25, that's available. If you'd rather not vote for five crappy teams at the end of a poll, that's doable. Etc.
- Allowing voters to have portions of their vote count extra. Hell, at the beginning of the year I didn't know what Howard's Rock was, let alone where Clemson should rank, but I knew what I thought of most of the Big Ten early. I may have been wrong, but at least I had an opinion. Maybe a team or two each week could be selected and your vote for that team would count double or triple.
I'll revisit this closer to the season's start to get a feel for everyone's opinion on the matter, but if you A) hate the idea of changing the poll format or B) have ideas for the change, please comment/email.
You may have seen an embryonic copy of the 2007 recruiting board when I inadvertently hit "publish" instead of "save draft," but please put it out of your mind. Shoddy work unfit for human eyes... until a week or two from now when it'll make its debut.
Of note: Michigan's always been a school that fires upon any and all targets of opportunity, no matter where they may be. In 2004 Michigan pulled the top three recruits out of Kentucky. Last year they got top 100 kids from California(Jonas Mouton), Florida(Greg Mathews), and South Carolina (Adam Patterson--and we're recruiting two more kids from Patterson's hood--Richland Northeast high school--in '07). The interesting thing about 2007 is that this scouring for horizon-minded kids may take a back seat next year due to the unprecedented level of talent within Michigan. Three kids are on the initial Rivals 100 list (up from one last year, though one has already committed to MSU) but the guys who keep an eye on these things think that will be at least five once the evaluations are done and no fewer than 30 players will end up at BCS schools. Michigan will probably end up with close to ten instate kids if all goes well. And the last time that happened? I dunno. Readers?
It's also good to have Herrmann gone. I am of the opinion that he was a yes-man co-conspirator who shared Carr's defensive philosophy. ... All is not forgiven or right, though. ... English has no coordinator experience and UM again failed to conduct a national search to land the best-possible candidate. It is important for a coach to feel comfortable with his subordinates (obviously), and there is something to be said for continuity in an industry that requires strong relationships (think recruiting) to succeed. But it is baffling that Michigan has not done more to bring in new, industry-leading ideas by way of better coaches.
; Vijay takes the high road:
...instead of "good riddance" or "finally", I think I'll send Jim Herrmann off with a "Thank you and best of luck" and the hope and honest expectation that he will distinguish himself as a good coach at the professional level.
I will offer an opinion anyway. It would be disingenuous for me to adopt Vijay's all-class attitude toward the departure of a man who caused me disproportionate misery over the years, but all told Herrmann is probably less responsible for the bunched-panties phenomenon than the man in charge. Hell, one of the strangest things about this whole drama has been the repeated assertions of Herrmann's "defensive genius" in the paper and from a couple of friendly insider emailers. And he did get hired by an NFL team to coach linebackers despite apparently not having bothered to undertake such a task at Michigan. Either this is just coaches taking care of each other--not an uncommon phenomenon in an industry with the job security of Dude Third In Charge of Al-Qaeda--or Herrmann really isn't the the begin and end of Michigan's problems these last eight or so years. It could be either. I vote some of both.
The hope represented by English is at least half a hope that Carr's no longer that guy and no longer wants a close, personal, suspicious-if-you-know-what-I-mean-NTTAWWT-I'm-not-prejudiced- except-towards-defenses-who-never-ever-hold-bowl-opponents-under-28-points friend as his defensive coordinator... but the comedy of errors that preceded English's ascension casts doubt on that. All the excitement over Ron English will evaporate right quick if Carr glares at his new defensive coordinator every time a cornerback lines up within spitting distance of a wide receiver. The proof, as they say, will be in the pudding. We need aggressive, clever pudding. Or something. This metaphor died.
In the end, Herrmann got exactly what he deserved: an anonymous, well-paid position where he can't hurt anyone except New York Jets fans--who deserve it--in easy harassing distance of Joey*. Now we have to find someone else to blame, but even there, Lloyd has provided in the form of new/old offensive coordinator Mike DeBord. He giveth, and he taketh away. Yea, verily.
I'm not ramping my expectations up to unreasonable levels quite yet, but at least if we fail next year it will be in new and interesting ways.
*Expect a string of strange incidents were Herrmann is embarrassed by loud flatulence at press conferences, consumes hilariously foul gum, and regards any and all jars of macadamia nuts with suspicion, wary of yet another pacemaker-testing fake snake explosion.
Vorified. You know this already. English to DC:
Ron English is the Wolverines' new defensive coordinator, a school source confirmed.
You don't know, or like, this paragraph:
Herrmann's future has not yet been decided. He could still remain at Michigan, possibly as special teams coordinator, which DeBord has handled the last two seasons since returning to Michigan after being Central Michigan's coach for four years.
No offense to Herrmann, but I'd be creeped out if English was looking over his shoulder at him when attempting to forge a steely-eyed asskicking defense. Still, this makes last week's nattering nabob of negativism explosion look silly in retrospect ("NNN" HT: Spiro T. Agnew(!)).
Le sigh. Right. Basketball team has flitted earthward once more. Reaction from RBUAS and NKOTB Maize n Brew. Wonk says that the dang-diddly-ang three point marksmanship being displayed by Michigan opponents is...
...almost exactly equal measures of both [bad luck and bad defense]. Ohio State has played worse defenses than Michigan's and not shot as well as they did last night. That much is luck. Still, the Wolverines do indeed have some defensive liabilities. Screens, to pick one mundane example, seem to work about 50 percent better against Michigan than what is normal, particularly against non-Graham Brown Wolverines. Much of the first-half damage from long range was done on simple ball screens. Daniel Horton (who got the rare points-assists dub-dub last night with a 15-10), Dion Harris, Chris Hunter--all struggled with screens. (Horton was also too eager to sag on the weak side last night and was burned on a nice skip pass to Foster from Sylvester.)
Personally I didn't think all that poorly of the perimeter defense, but I defer to the infinitely more qualified to answer Wonk on this matter.
Grant Bowman's back in the boiler room, but his brief detour to the Steelers' practice squad got national attention, including a seven-minute interview on the Bob & Tom Show. Dangerous Logic captured the wild radio beast and set it down in data for your listening edification. Enjoy.
Must be something in the gumbo. From Gatorsports.com's Florida recruiting recap, DBs section:
ONE THAT GOT AWAY: Jai Eugene, the No. 1 cover corner in the nation, told Tebow the night before the Army All-American Game that he was committing to the Gators. He picked Michigan instead, then signed with LSU.
Jeez. I envision Eugene at a local McDonald's, paralyzed with indecision when asked if he wants to supersize: "Yes... no. Can I say Michigan? I can? LSU."
Chance of Sports Guy mocking this particular thing he doesn't understand: 100%. Boston College is going to play a hockey game at Fenway. Why I say thing about?
''We're going to do it," said a BC official yesterday. ''It's just a matter of finding an opponent and a date that will work. But we're going to play a game in Fenway."
The details will be worked out over the next several weeks, and they could include the Boston University team as part of a doubleheader. The original idea called for such a doubleheader, with BU and BC against teams from the Midwest, possibly Michigan and Michigan State.
This would be cool, and Red is generally up for wacky things and tough opponents. Plus we might get a "Red Berenson face," which is him staring at you, wondering exactly how many fingers he would have to use to kill you, and settling on three. Maybe three and half because you're a wiry bastard.
There are three ways to be noticed on the Internet.
- Work very hard to create something interesting and valuable.
- Say something dumb.
- Display gazongas. (Recommended)
I mention this because apparently Jen is getting an advice column at SI on Campus:
Do you need some relationship advice? Having a problem with a friend? Want to know what really goes through a girl's mind. Email Jenn at email@example.com and she'll do her best to answer your questions in her weekly "Dear Cowgirl" column which will debut next week.
I forsee some poor intern at SI dealing with a lot of stuff like this:
Will you sleep with me?
-Joey, New York
Beleaguered SI Intern: No. Love, Jen!
Will you sleep with me?
-Paul W, Atlanta
Beleaguered SI Intern: No. Love, Jen!
I WOULDNT MIND COWING YOU'RE GIRL IF YOU KNOW WHAT I MEAN. IF YOU DONT IT MEANS SEX. WITH ME.
-Brian, Ann Arbor
Beleaguered SI Intern: I have no idea what you mean. You should seek help. Love, Jen!
How about now?
-Joey, New York
Beleaguered SI Intern: No. Love, Jen!
I HAVE A BLOG DOES THAT MAKE YOU HOT
-Brian, Ann Arbor
Beleaguered SI Intern: Please stop emailing me.
Okay, okay, I realize that not even a day after posting that I won't make references to music groups that stopped being relevant in 1987 I make a reference to a music group that stopped being relevant in precisely 1987. The Smiths don't count.
2/9/2006 - Michigan 85-94 Ohio State - 16-5, 6-4 Big Ten
You could file any number of events from last night into the folder labeled "Akin To Joe Theismann's Horrific Leg-Breaking"--gruesome things that are repeated ad nauseum despite no one wanting to see that particular atrocity again. To wit:
- Another team launching bad three after bad three and seeing them ALL GO IN.
- Another loss to Ohio State--only a fortunate opening four minutes of the latest hockey game's third period prevented a wholesale season sweep (to date, anyway) at the hands of OSU.
- Another Michigan player collapsed, clutching at an ankle suddenly more foe than friend.
- Another season of wilted promise.
Are you there, God? It's me, Michigan basketball. We're wondering when our training bra is going to fill out and our program will return to its buxom ways of years past--without the sin sin sin(!!!), of course. It would help if you ceased smiting us. Sure, the plague of broken safeties was a good one. But I think we're all in agreement here on Earth that maybe your time would be better spent making stuff like world peace or a Norm MacDonald late night talk show happen.*
Right, right, ineffable plan and all that, and I want to make it very clear that no one here is questioning the plan. It's a great plan. It's maybe the best plan I've ever seen. It's just... if I had to choose between the plan and not clawing my eyes out, I'd have to choose the not clawing my eyes out.
*(Eh! It's the Norm show! Our guest tonight are... eh... get me a drink. I've got this cake here. It's my girlfriend. Yeah, my girlfriend. I f***ed last night.)
Bullets of Depression
- How different does this season look if Iowa and OSU don't shoot totally impossible percentages from three? Reduce Iowa and OSU's three point percentages to a damn-good 40% and Michigan drops one narrowly in Carver-Hawkeye and wins going away against the Buckeyes. Is this a reflection of poor perimeter defense? Perhaps in some way, but against OSU the Buckeyes just shot a bevy of terrible threes; all of them went in.
- Michigan was outstanding offensively. Sims destroyed everything. Harris, Coleman, and Horton played well, though Horton's shot was off.
- I don't see how you can blame Amaker for opponents shooting 70% from three or his best players getting injured consistently, but the problem is that this team is going to start going backwards next year. Two posts graduate and DeShawn Sims, who is 6'7", is the nearest thing to a replacement. Petway and Sims will be seniors in '06-'07, and then there will be nothing. Jerrett Smith is going to have to transition from being a meh backup point guard to the full time starter. Amaker's recruited two types of players: guys who would come to Michigan even if Hitler was coach and guys no one else wanted.
Ron English May Possibly Be The Defensive Coordinator Maybe I Promise Nothing In Fact I Suggest You Refrain From Reading This And Enjoy Some Lemonade
Right. Hope you enjoyed the exceedingly brief DC Jon Tenuta Era. Enjoy the whiplash.
Rivals premium board headline:
We just spoke with the Chicago Bears ... something is brewing* - MichaelSpath 2/9 3:44 PM (show all)
Post from sudden insider and professional fatty-hater Paul on The Victors:
RE to DC*
Posted on February 9, 2006 at 04:07:59 PM by Paul
(RE == Ron English, obviously.) More bullcrap rumor mongering!
BTW my source is in the NFL. This is solid* - By Paul February 9, 2006 at 04:16:40 PM
Oh. My bad.
Personally, I know a guy who knows a guy who was a football manager recently--said manager was described as "freakin' out." Just check it out... this is the grand mother of all rumbles. I ain't sayin' nothing. I'm just sayin' that if this particular rumor independently spawned from multiple places is false, I will nail Laura Quinn.
That's right. Laura Quinn.
Update: Scout front page. Reports that it was announced on WTKA.
Update Update: Rivals front page. Everyone is being very circumspect about saying what exactly English is going to be doing at Michigan, but do the math: co-DC == Bears. English's New Role > co-DC.
Update Update Update: It's officially official.