S&P+: We are #3(?!), Heavy Favorites in Every Remaining Game

Submitted by mgoBrad on

I think by now we're all aware how much the advanced stats like our team this year, but this is... crazy? F/+ has not been updated for last week's results yet, but S&P+ has updated and we are now #3 in the country (Offense #53, Defense #3, plus excellent special teams). 

The rating has us as 16.5 point favorites this weekend (!?), 17.5 point favorites against MSU (!??!), and 15.8 point favorites against OSU (!?!?!?!?!?!). In fact, by this metric, our toughest remaining game is projected to be... drumroll please.... @PSU.

Here's a reminder explaining how S&P+ ranks teams:

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

 

Your thoughts?

Here is the relevant data: CLICK

See also: "Michigan and Florida have been the best teams in their divisions so far"

UM Fan from Sydney

October 5th, 2015 at 12:44 PM ^

We're fans, not players or coaches. We can talk and worry about any future game we want.

I really think we're going to beat Florida in Jerry World. Oh shit. I just jinxed it. I was looking ahead too much.

Newton Gimmick

October 6th, 2015 at 5:28 PM ^

he's basically in a vacuum, inside a garbage bag, on another planet in another dimension. Those of us still in Ann Arbor, however, are much closer to the vibe of the program.  If we're always bitching about not meeting ridiculous expectations, I think that can infect the program.  (In other words I agree with Gucci Mane!)

VicTorious1

October 5th, 2015 at 3:26 PM ^

The general consensus is that, as fans, we can look ahead to any game (and not only the next) without an effect on the game or a change to the outcome.  The players and coaches, on the other hand, should focus solely on the next game as they can truly affect the outcome.  

It might be more valid for fans to say, "no need to worry now about going undefeated the rest of the season and making the playoffs. we'll talk about the playoffs as we get closer to winning out."  This reasoning might make more sense as there is still plenty of ball left to play.  Nonetheless, it still would have no effect on any of the remaining games.  

TLDR:  Looking ahead to NU and MSU has no effect on the outcome of those games.

mGrowOld

October 5th, 2015 at 12:13 PM ^

My thoughts are that as much as I love Michigan I want to find the guy who created this formula and see how much he wants to wager me on the MSU & OSU games.  I'll take the Spartans +17.5 and the Buckeyes +15.8 points for all the money he can find to bet.  

Lanknows

October 5th, 2015 at 12:26 PM ^

Then the S&P line against MSU will be bettable, assuming Vegas comes in pretty close.  They typically do, since Vegas uses simlilar (proprietary) computer models to set lines.

Then you can put your money where your mouth is.

....

If the rankings stay the same  (i.e., Assuming Michigan PLAYS like the #3 team in the country against northwestern (a top 30 team) and beats them handily, and further assuming that the spartans fail to dominate the #110 team in the country (Rutgers)) then you can expect Michigan to come in as double digit favorites.  S&P may be off by a touchdown or so, due to public perceptions and polls, but I'll take the algorithm over conventional opinion more often than not.

StephenRKass

October 5th, 2015 at 12:15 PM ^

I mean, the advanced stats are nice and all, but we haven't faced a dominant OL, a dominant DL, and a dominant RB like Zeke Elliott. Sure, I'm happy. I'd say it is reasonable to be favored against most teams. But if we're favored now like that against OSU, I'm not buyin'. I think we'll play OSU tough. I think we have a good chance of winning, if we keep improving. But favored by more than two TD's? What're they smokin'?

harmon40

October 5th, 2015 at 6:34 PM ^

Against MSU and OSU, toss out the rankings and advanced statistical analysis. It will come down to match-ups and the chess match between the coaches.

However we no longer need fear ANY team. Win or lose, this team will not be embarrassed. They will be prepared and they will fight to the death. 

It is sweet to look at the toughest games on our schedule and feel optimism, rather than simple despair. 

 

 

alum96

October 5th, 2015 at 12:19 PM ^

West Virginia was #3 last week for what it is worth (and still #7).  I dont like the combined S&P+ as a "ranking tool" and I don't think its really the correct way to use it.

I like to use the S&P+ defense and S&P+ offense separately as ranking those specific units nationally (i.e. UM defense #3, UM offense #53).  Think it makes much more sense than the combined which tries to rank a team.  By the way Louisville at 2-3 is #13, Arkansas at 2-3 is #17 and Utah at 4-0 is #19.  A pretty miserable PSU is #20 while an impressive Texas A&M is #22.

6tyrone6

October 5th, 2015 at 12:16 PM ^

only get better. Harbaugh has UM playing like UM in 1980's, great defense, efficient offense and 10 plus wins per season. How exciting is it winning 3 yards and a cloud of dust! Soon we will have NFL bound QB's and receivers and the transition back to the future will be complete.

 

stephenrjking

October 5th, 2015 at 12:16 PM ^

Season is still early. We'll know more after the next two weeks. I expect OSU to round into fine form by the time they come to visit; I am trying very hard, and mostly unsuccessfully, to suppress ecstatic visions of what we will do to MSU in two weeks. 

harmon40

October 5th, 2015 at 6:24 PM ^

Statistical analysis is all fine and good but specific match-ups matter. MSU's D has struggled, but will we be able to take advantage of their awful secondary? Will they stone our running game with 8 in the box?

That said...

Right now it feels like we ought to beat them. How is the #86 offense in the nation going to move the ball on us?? C Cook will have to face the #2 D in the nation...while playing behind a patchwork OL. 

Optimism is certainly warranted. The last couple years it felt like we had no chance to beat MSU, or even play with them. Whatever happens in that game this season, it's awesome to not feel hopeless, to wonder how long we can hang with them before they pull away, etc.  This is no "puncher's chance" we have. We can feel legitimately confident heading into that game - and every other one on our schedule. Happy times!!

Lanknows

October 5th, 2015 at 12:17 PM ^

but it's better than AP and Coaches Polls.

If you look at the point lines that the S&P estimates produce they are pretty consistent with Vegas, whose lines are pretty consistent in being right in picking winners and losers.

 

Mr Miggle

October 5th, 2015 at 3:00 PM ^

Minnesota has 3 wins against Colo St, Ohio and Kent St, all by 3 points. A 27-0 loss to NW (who's also behind them) and a close loss to TCU. They have OSU as a 3 point favorite at home. If that game was this week, I'd add at least 14 points to the line.

MSU is #34, which is good for 9th in the Big Ten.

Maybe late in the year, their predictions may line up with Vegas lines. For now, I don't believe they're anywhere close.

Njia

October 5th, 2015 at 12:29 PM ^

They've got the Germans ranked #1, Japan #2. USA team isn't even cracking the Top 10. I don't know if I'm going to put my money on the British to win the next game against Team Hitler, either. Look what just happened at Dunkirk.

You Only Live Twice

October 5th, 2015 at 12:26 PM ^

What is their logic?  That our D has not given up any points in 2 games, hence will give up no points the rest of the season?

We'd all love that don't get me wrong....

In reply to by You Only Live Twice

Lanknows

October 5th, 2015 at 1:16 PM ^

I am 90% sure that # of shutouts has no impact on the calculation.

But the low yardage totals and 3rd down success rate are a big deal.

BornInAA

October 5th, 2015 at 12:29 PM ^

Football fans + Excel + Internet 

Can make any team they like look great by playing with the weighting of statistics.

If you heavily weight defense and points against we look like unstoppable gods. 

On the offense side, not so much.

Reader71

October 5th, 2015 at 12:29 PM ^

How does S+P usually do in projections?

I think it is a great tool to rank and describe what has already happened, but is it's record as a predictive model any good?

alum96

October 5th, 2015 at 2:05 PM ^

I've never really seen it used that way as described.

What it does very well is rank units on each side of the ball and make adjustments for SOS, garbage time, etc.

So if you look at all the defensive units vs other defensive units it does a fine job ranking them.  Same for offensive units.

This "combined" score thing - I am not sure how it is compiled nor do I care when it has a bunch of 3 loss teams in the top 20.