Moving the (Stati)Sticks: Week Four Comment Count

Adam Schnepp

21743838161_156d6f599c_z

[Upchurch]

The mark on my elbow is still there. It’s smaller now; what once was about the size of a nickel is smaller than a dime, but it persists. I felt the stinging sometime in the fourth quarter and immediately turned to Ace to show him, not because I wanted attention for my accidental self-inflicted narrative device but because this thing on my elbow happened because of that thing that we were watching on the field. I was using the elbow as a pivot point, and so many incredible things had happened for which I needed to quickly cover my mouth to preserve the thin veneer of professionalism over the roiling excitement inside that the skin just rubbed off.

The numbers from last Saturday don’t provide as visceral an experience as what we saw on the field, but they show just as much dominance. Michigan jumped 15 spots in overall S&P+ from 25th to 10th, while the defense moved up to fourth. FEI has Michigan 26th, which is 10 spots higher than they were after week three. F/+ (which combines S&P+ and FEI) has Michigan 16th.

FEI uses preseason projections that includes things like last year’s performance (/shudders) and a five-year FEI rating with diminishing weight until week 7, when they’re dropped completely. You can see why Michigan’s digging out of a hole here. Every week, though, Michigan’s a little closer to shedding the baggage of the past, and that has felt true both literally and metaphorically.

The first three weeks of the season felt like the cold open of a new restaurant; there were kinks expected and, all things considered, the operation got off to a relatively smooth start. Week four was the grand opening; the curtain was pulled back, and the conversation shifted from how much potential the new joint has to how good it is right now. This new place is great. Just remember to keep your elbows off the table.

[After THE JUMP: Less about flesh wounds, more about numbers]

The Mathlete’s Four Factors:

Once again, a quick reminder of what the factors mean:

Conversion rate = [1st Downs gained]/[1st Down plays (including first play of drive)]. A three and out is 0/1. A one play touchdown is 1/1. Two first downs and then a stop is 2/3, etc.

Bonus Yards = [Yards gained beyond the first down line]/[Total plays from scrimmage]

This is an adjustment to how I have previously calculated, to account for the plays a team runs.

Field Position = The expected point difference per game for where a team’s offense starts and where a team’s defense starts. Each drive is given an expected value based on the start of scrimmage, all of the drives for the offense and defense are totaled and compared. This accounts for all elements of field position: turnovers, special teams, drive penetration etc.

Red Zone: Points per red zone trip (TD’s counted as 7 regardless of PAT)

Offense:

  Field Pos. Conv. Rate Bonus YPP Red Zone
Week 1 21.0 73 1.52 5.7
Rank 60 30 59 27
Week 2 25.0 68 1.63 5.8
Rank 77 70 95 30
Week 3 25.3 70 2.56 5.8
Rank (B1G Rk) 84 (12) 54 (6) 43 (6) 30 (5)
Week 4 24.5 72 2.92 5.8
Rank (B1G Rk) 91 (14) 30 (4) 31 (3) 27 (4)

Defense:

  Field Pos. Conv. Rate Bonus YPP Red Zone
Week 1 27.9 73 1.64 57
Rank 47 44 20 30
Week 2 25.1 67 1.60 6.1
Rank 51 58 23 88
Week 3 24.0 63 1.28 6.1
Rank (B1G Rk) 35 (4) 38 (6) 9 (3) 100 (13)
Week 4 23.1 59 1.23 6.1
Rank (B1G Rk) 29 (5) 17 (4) 4 (1) 110 (13)

The Four Factors paint a picture of a truly dominant defense that has improved in every category every week with the exception of Red Zone. But hey, if your opponent never gets there, it’s hard to decrease that number.

Offensively, seeing Conversion Rate rebound after a bit of a dip is encouraging in terms of offensive efficiency, while Bonus YPP’s consistent increase is indicative of an offense that has started to pick up some big gains. (We’ll look at Bill Connelly’s Explosiveness stats later. Spoiler: The offense is getting more explosive.)

The weekly drop in Field Pos. was something that I was concerned about last week, but this week I was so worried/confused after Michigan put up 31 points and dropped not just nationally but also to the bottom of the Big Ten that I had to talk to The Mathlete. In his words:

So field position is a bit tricky. It really is a way of comparing what your offensive opportunity is instead of just output. A Michigan team scores 28 ppg (3.5 points above field position) would indicate a much better offense than 28 ppg from Penn St (2.6 points below field position). Michigan's low number indicates that the offense has had a bad combination of starting field position and fewer total drives in their games thus far. One note, the 24.5 is 1.4 ppg above the defensive number meaning that Michigan is winning field position in their games, even if their overall offensive number is on the low side.

Ah, The Mathlete. The cool, calm voice of reason in a storm of angry, aggressive numbers.

Advanced Box Score:

Things BYU can hang their hat on: Having the same number of turnovers as Michigan (which was zero). Things BYU did worse than Michigan, statistically: Every other category in the adv. box score. There’s one draw, and that’s in number of drives; both teams had 12. With those 12 drives, Michigan averaged 6.05 yards per play, had six scoring opportunities, and averaged 5.17 points per scoring opportunity. BYU, on the other hand, averaged 2.27 yards per play, had one scoring opportunity, and averaged a nice, round 0.00 points per scoring opportunity. BYU was also starting from farther away (their 22.8 yard line compared to Michigan starting at their 27.9), for what that’s worth. Success Rate is basically a measure of how often you’re on track (i.e. favorable down and distance), and BYU’s was 17% in this game. It’s fair to say Michigan derailed them.

Five Factors and Other Stats:

The story on offense is essentially the same as last week, though Michigan’s moving up in most categories except efficiency (they dropped five spots); they’re now 89th in Explosiveness, 32nd in Success Rate, 23rd in Field Position, and 79th in Finishing Drives.

Defensively, Michigan is third in Explosiveness, 19th in Efficiency, 15th in Field Position, and eighth in Finishing Drives. The ones that stand out to me are Explosiveness and Finishing Drives. Michigan’s defensive line has been so good that they can rely on a six-man front while still stopping the run, and the corners and safeties (and HSP) have been able to cut off anything deep. Michigan hasn’t been gashed consistently or in a specific way, and that lack of a run/pass weakness has been a chief reason they’re able to limit big plays. They’re also only allowing 3.17 points per trip inside their 40, which is unreal. While it’s true that looking 20 yards outside the traditional red zone should keep numbers down a bit, teams are barely averaging more than a field goal when their drive gets that far.

BYU was supposed to be the pass defense’s first big test (especially as far as stifling big plays), and they passed with flying colors. That unit is now ranked 21st in Passing S&P+, 35th in Success Rate, and 9th in IsoPPP. IsoPPP is Bill Connelly’s measure of explosiveness, so that was the prime target for a stat that could have suffered against BYU.

I’ve already droned on about Michigan’s ability to limit opposing big plays, but there’s one more item of interest to look at if we zoom in a bit more. The defense has been good at stopping big plays regardless of the situation; Michigan is third in Standard Downs IsoPPP and 21st in Passing Downs IsoPPP.

One of the weirder things about the offense is that they are definitely more explosive in a certain situation, and it seems counterintuitive. Michigan’s offensive Standard Downs IsoPPP ranks 108th, but their Passing Downs IsoPPP is 15th. If you think you have an idea why that might be hit the comments, because I’ve thought about it for a while and haven’t found a good answer.

One thing that might play into weird passing downs stat is that Michigan hasn’t been in those situations all that often. Michigan’s Passing Success Rate is 46.2% (35th nationally), while their Rushing Success Rate is 49.1% (28th). There’s a great deal of balance there, as Michigan’s well above average at staying in favorable down-and-distance situations whether they choose to either run or pass.

An interesting trend to keep an eye on offensively is Michigan’s increasing S&P+ by down. They go from 99th in the country on first down to 30th on second down and 21st on third down. Defensively, Michigan gets out to a good start (9th in the nation on first down), dips down to 40th on second down, and finishes an impressive second on third down.

Yeah, man, I do read the comments:

This was requested last week and I think it’s a good idea. You can quickly glance through and do a little advanced scouting while also looking at who we beat, which is helpful if you’re trying to keep the season in context. I understand that’s not easy right now.

Opponent Off. S&P+ Def. S&P+ Overall S&P+
@ Utah 43 36 29
Oregon State 94 50 76
UNLV 111 80 97
BYU 56 39 42
@Maryland 73 68 70
Northwestern 105 10 38
Michigan State 19 26 15
@ Minnesota 67 11 20
Rutgers 86 105 102
@ Indiana 25 101 69
@ Penn State 68 13 22
Ohio State 33 6 7

You can see from the chart above that Maryland isn’t very good, but Jake Troch made a matchup visualization tool that really drives the point home. If you click on the “Advance- Play” or the “Advanced- Drive” tabs you’ll see that Michigan has the advantage in every defensive category he tracks, and it’s usually not even close. The charts are great because they compare offense to defense, so you can get a good feel for overall matchups. I’ll leave you with this, from the “Advanced- Play” tab:

adv maryland

Yes, there are some categories where Michigan’s defense is almost 100% better than the NCAA average when compared with the Terps’ offense. If this plays to the numbers, the weather won’t be the only thing in Maryland that’s ugly on Saturday.

Comments

jmblue

October 2nd, 2015 at 9:18 AM ^

I'm surprised MSU's offensive S&P+ ranking is that high when they've put up pretty ordinary numbers through four weeks.  Is this still partially based on last season?

Minnesota and PSU's overall rankings also seem odd.

 

alum96

October 2nd, 2015 at 9:49 AM ^

I think people's expectations for their offense are a bit out of line.  They still have scored 30 in every game.  Yes some of it has come late but there was a stat I read last week as they have the longest streak in CFB scoring 30 a game.  I think it goes back to the 2013 season when they "only" scored 28 v Stanford.

I think Cook has not taken the next step, he looks like Cook 2014 and maybe the running game is not typical MSU yet but they still are churning out points.   Their 2014 offense was top notch and I think people are comparing 2015 to 2014 and saying "disappointing" when they indeed lost 2 of their 3 top playmakers in Lippett and Langford.

That said, as I posted mid week they have played on average the 88th ranked S&P+ defense in the country and its only going to get easier the next 2 weeks ... the average will actually drop to 90 in 2 weeks!   UM's D will be a shock to their system after playing zero quality defenses thru week 6.

  • W. MI 112
  • Oregon 89
  • Air Force 73
  • C. MI 77
  • Purdue 84
  • rutgers 105
  • UM 4

alum96

October 2nd, 2015 at 10:05 AM ^

Hmm, damn sports outlets!

Image result for pitchforks and torches gif

 

Looking at their schedule I dont see a defense holding them below 30 that is not named UM or OSU. Neb D is not very good.  And they dont play Minn or NW.  Maybe PSU can if Hack doesnt do 4 INTs.  So its still a good offense, esp in Big 10 terms.

 

BrownJuggernaut

October 2nd, 2015 at 9:24 AM ^

I'm pretty sure you're not supposed to be literally rubbing elbows in the press box, Adam.

To me, the fact that some of these stats contain a preseason or historical component is what makes me extra excited. Michigan has been okay and sometimes good, but nowhere near this level and the trend up is very encouraging. At least on a statistical level.

alum96

October 2nd, 2015 at 9:45 AM ^

If you want to cheat and save some work I post the opposing teams (and some other selected teams) S&P+ and FEI data weekly starting around week 6 in the Mgoboard.  FEI wont be available until week 7 this year but they had it up earlier last week.  Thanks for adding it within your story as I think it is important to post context.   To that end I expect UM D to remain a top 10 unit all year in advanced stats as most of the offenses in the future are...well crappy.   OSU MSU and Indiana being the obvious 3 exceptions.

JeepinBen

October 2nd, 2015 at 9:36 AM ^

My guess is it's because Michigan has been content to take their foot off the gas and run into 10 man boxes. In the UFR Brian claimed he would have RPS minused lots of stuff that ends up as "meh" because it's 31-0 and you're working on  Power no matter what the defense is doing.

Whereas when we need to have a big play (passing downs) we're better at doing it.

Just a guess

dragonchild

October 2nd, 2015 at 9:48 AM ^

I'm thinking the same thing (I just posted below), but the thing about big leads is, if that was all there was to it, then Michigan would be routinely seeing passing downs in the "scrimmage" part of the game as the defense sold out on power.  The stats don't reflect that.  They definitely seem to punt more in the second half, but they haven't exactly been going backwards.

The playcalling definitely eases up when we're up 3-4 TDs in that they're not calling double fake screens or WR reverses or other crazy stuff.  But they're still throwing to the outside and sending blocky-catchy guys downfield to move the chains.  So instead of a 1st and 10 becoming a Hoke-esque 3rd and 9, they might get occasionally stuffed but they'll throw a few wrinkles to keep the run D on edge and create as many 3rd and short situations as possible.  The main difference is, to convert those 3rd and 2s in the second half they'll use more conventional stuff instead of reaching into the bag of tricks, which results in more punts and worse standard downs stats.

dragonchild

October 2nd, 2015 at 9:37 AM ^

Michigan’s offensive Standard Downs IsoPPP ranks 108th, but their Passing Downs IsoPPP is 15th. If you think you have an idea why that might be hit the comments

That's reflective of the playcalling, which has been both high-volume and brilliant.  I've noticed they'll play to their identity on standard downs (and late with a big lead) but when they're in a passing down they'll start playing cards the opponent isn't ready for.  They'll also try out the long ball on early downs with little success.  In terms of results, when they're in standard downs they often try for and are happy to get 2-4 yards and occasionally run high-risk plays that should pick up chunks but -- unfortunately with Rudock's poor play -- often result in zero yards.  Also, they're still working on consistency so they've messed up some perfect playcalls with botched execution.  So at a glance their standard downs play -- especially compared to spread teams -- looks pretty awful.

However, they never bail on a drive.  When they need 8+ yards on 3rd down they have the plays and go get it.  In a sense, in passing downs they're much more "serious" about getting yards.  Well, that's not quite right.  They're always serious, but they want to be high-volume; they relish third-and-short and efficiently manage the routine yardage situations.  But they also understand the importance of each drive and getting first downs and they've been very, very good at getting x+1 yards when they need x.  Subconsciously I'm starting to build excitement on 3rd downs because regardless of the distance, that's when Harbaugh's most likely to call something wicked.

The playcalling is very, very good.  It's downright masterful.  I mean it should be obvious from his 49ers tenure but I'm really starting to see why Harbaugh is one of the best coaches in the country.  Borges for all his faults could be a pretty good schemer on a good day but he looks like a child compared to this.

blueblue

October 2nd, 2015 at 11:07 AM ^

That interpretation also supports the progresion in success rate from first down to third down: on first and second down, you're trying to be predictable, because getting the other team to make assumptions about what you're going to do is how you surprise them on third down. 

Loved the point about how they never bail on a drive. Very Harbaugh--where strategy meets attitiude. 

Quag77

October 2nd, 2015 at 10:17 AM ^

Problem with statistics is that you can twist and turn and show what you want to show in so many ways by icluding/excluding variables in any category.    Justify a loss or substatntiate a win all you want but bottom line... A win is a win and a loss is a loss!  Just win!

Ecky Pting

October 20th, 2015 at 8:33 AM ^

I thought I'd posted this yesterday in a comment on alum96's Opponent Stock Watch, but then the Internet broke and yada-yada-yada.  So, I'll try again here!

It's nice of Bill Connelly to post updates to the win probabilities on the cool Adv. Stats Profile pages he's putting out now on FSH. As mentioned in other places U-M is the favorite in all it's remaining games at this point on that basis. We'll see how that holds up...

But the reason for this note is that since it appears Mathlete is on some sort of hiatus, and I'm not finding any other reference to an expected Total Wins Probability density function, so I'll just have to do one myself.  So after locating a clever recursive solution to a Poisson Binomial Density function, here ya go (click to embiggen):

U-M Football Total Wins PDF, 2015w4

Mr. Obvious: "Holy hell! It looks like the most likely number of wins is 9!"

Other things worth noting:

  • The nicest warm fuzzy IMHO is when I look at the likelihood for a repeat of last year's 5-7 record. Yup, that's right.  It's a statistical impossibility (at least to 2 decimal places)!
  • It's equally likely that M wins 11 games as it is to win 7. I'm just sayin' ... but I'm not going to even think it.
  • A more subtle aspect is that 10 wins is ever so slightly more likely than 8 wins. I will go there... Me likey!

Whether you're a glass-half-full or a glass-half-empty type of person, you have to admit, the glass is filling up, and it's not looking too damn big at this point.

teamteamteam

October 2nd, 2015 at 11:26 AM ^

Really like this post every week. Can you add field position per drive? If I understand these stats right this would give us the expected points per drive and a better understanding of actual field position since my guess is we are on the lower end of total possessions each week this skewing our ranking in the field position category

alum96

October 2nd, 2015 at 12:16 PM ^

you link works

this is a more laymans explanation

The Fremeau Efficiency Index (FEI) considers each of the nearly 20,000 possessions every season in major college football. All drives are filtered to eliminate first-half clock-kills and end-of-game garbage drives and scores. A scoring rate analysis of the remaining possessions then determines the baseline possession efficiency expectations against which each team is measured. A team is rewarded for playing well against good teams, win or lose, and is punished more severely for playing poorly against bad teams than it is rewarded for playing well against bad teams.

The S&P+ Ratings are a college football ratings system derived from both play-by-play and drive data from all 800+ of a season's FBS college football games (and 140,000+ plays).

The components for S&P+ reflect the components of four of what Bill Connelly has deemed the Five Factors of college football: efficiency), explosiveness, field position, and finishing drives. (A fifth factor, turnovers, is informed marginally by sack rates, the only quality-based statistic that has a consistent relationship with turnover margins.)

howejunofe

October 3rd, 2015 at 10:44 AM ^

  Start   working at home with Google! It's by-far the best job I've had. Last Wednesday I got a brand new BMW since getting a check for $6474 this - 4 weeks past. I began this 8-months ago and immediately was bringing home at least $77 per hour. I work through this link, go? to tech tab for work detail

---------------------- _+_+_+_+_  w­w­w.b­u­z­z­n­e­w­s­9­9­­.­c­o­­m