SIAP: UM Opens -11 against NW, Up to -13.5 on Some Books

Submitted by FauxMo on

Rarely do I see a Michigan line where I don't think either "that's about right" or "Michigan should be favored by less," but this is one of them. I think opening at -11 is nuts, and even -13.5 is still too little. Given that NW lost at home to both Duke (by 14) and Akron (by 5), and that the stadium will probably have more Michigan fans than Wildcats fans, I think this one feels like -17 or -18. 

Is NW getting a player or players back from injury? Is this just skepticism because Michigan has been crap on the road the last few years (decade), or what?? 

SugarShane

September 24th, 2018 at 9:25 AM ^

It’s a road game against a team that’s typically well coached. You’re overrreacting to the Nebraska game if you think a 2 touchdown spread isn’t generous enough 

ak47

September 24th, 2018 at 12:02 PM ^

Its actually generally more right relative to Vegas earlier in the season. Early season lines are generally skewed by off season hype and over rating teams which impacts vegas betting lines but not S&P numbers (i.e. ND and Michigan are always over rated by their fans, etc.). By mid to late season Vegas and the bettors have honed in on team quality and spreads are more reflective of reality which reduces the success rate of S&P.

Clarence Boddicker

September 24th, 2018 at 10:33 AM ^

Don't get why more people don't understand this. Bookies aren't predicting outcomes. The lines are set to attract equal betting on both sides and changes based on betting patterns. It reflects what gamblers collectively think the outcome will be, not the analysis of a well-informed football guru.

ak47

September 24th, 2018 at 10:07 AM ^

Actually it beats the spread by about 51-54% per year and was at like 57% through 3 weeks this year. So yes it is.

However I've seen too much weird shit between Michigan and Northwestern and NW is coming off a bye so I'm not touching that bet even if I did bet.

Bluelaf

September 24th, 2018 at 10:23 AM ^

Beating the spread isn't a good measure of accuracy, unfortunately.  If S&P+ were setting the lines, Vegas would beat the spread the majority of the time.  This happens because no matter how accurate a predictive system is, it will never be able to put its spreads at the exact median of outcomes.  So if you flip a coin every game, you will probably beat the spread.

The question you want to ask is: whose line is closer to the final margin on average?  And it's Vegas; wisdom of the crowds is powerful!

 

Edit: FWIW I think the spread should be about -8.

J.

September 24th, 2018 at 10:38 AM ^

> So if you flip a coin every game, you will probably beat the spread.

This makes no sense.  There is no way to assign meaning to those words such that this is a true statement.  The best you can hope to do is to beat the spread half the time by flipping coins.

Bluelaf

September 24th, 2018 at 11:20 AM ^

Consider a system that sets lines: X% of the time, the line will be below the outcome, and 100%-X% of the time, the line will be above the outcome (ignore pushes for the moment.)  If X% is anything other than 50%, then flipping a coin will beat the spread more often than not.  And no matter how good Vegas is (and they're pretty damn good) X% will never be exactly 50%.  So flipping a coin will beat the spread on average (of course, you won't make any money because of commissions).

 

J.

September 24th, 2018 at 12:02 PM ^

OK, I see what you're trying to argue, but you're still wrong, and I can prove it.

Imagine two bettors; A and B, and take commission out of the equation for a second.  A likes your logic, so he decides to flip a coin and bet on the outcome: heads to take the favorite and tails to take the underdog.  B is a contrarian sort who thinks your logic is silly, so he watches A flip his coin and bets the opposite: heads to take the underdog and tails to take the favorite.

By your logic, both of them will beat the spread on average, despite the fact that they're taking opposite sides of the same proposition on every wager.  That's clearly impossible, so your hypothesis must be incorrect -- QED. :)

Your general point is correct; Vegas can't always predict the margin.  However, the error is not systemic.  If it were, you could take advantage of it, and you wouldn't need a coin flip to do it.  For example, if underdogs covered the spread 60% of the time, you'd make a killing by betting every underdog.  However, since you can't tell in advance whether the spread is high or low, it doesn't really help you.

Now, sharps do exist, and they make their money by exploiting the public's willingness to bet at poor odds.  The best way to make money as a sports gambler is to be a contrarian; in this case, it would mean waiting to see how far the line moves in Michigan's favor and then taking Northwestern, because (a) Michigan is a more popular team and (b) Michigan is coming off an epic beatdown whereas Northwestern is not.  The public money is going to favor Michigan heavily (as you can see by the line movement so far).  The books are counting on the sharps coming in on Northwestern, or they're going to have way too much exposure to a Michigan blowout.

Bluelaf

September 24th, 2018 at 1:19 PM ^

Ok, you're right.  Coin flipping will fail half the time.  And that's because the center of the distribution for coinflipped picks is the line, not the average outcome.

 

I maintain that beating the spread >50% does not mean you're more accurate than Vegas.  My intuition for why was wrong, so let me take another stab.  If you have an unbiased predictor, no matter how large its variance, you're going to be on the right side of any other imperfect predictor more than 50% of the time, as long as your errors aren't (too) correlated.

Ali G Bomaye

September 25th, 2018 at 2:12 PM ^

I have no idea what you mean by "betting Austin's public teams." That said, obviously betting sports using casino lines is a net loser on average because of the vig. My point is simply that S&P+ attempts to be a predictive metric, whereas Vegas lines attempt to make the casinos money. Two different goals.

FauxMo

September 24th, 2018 at 9:37 AM ^

Well, I am on record here somewhere saying that I am not drinking the Kool Aid after the Nebraska game. Good performance, but against a team that looks both under-talented and skeptical of their coach. Still, Northwestern lost to Duke and a MAC team at home already this year, and the crowd will make this basically a neutral field game. Like someone says vis-a-vis S&P+ below, going just on the numbers, UM should win this game going away pretty easily. 

Newton Gimmick

September 24th, 2018 at 12:41 PM ^

This metric suggests that he might be a good game-day coach, since his win totals overperform relative to expected win totals based upon advanced stats.  It also may suggest that going strictly by S&P may be less predictive in this case.

https://www.footballstudyhall.com/2018/6/27/17509234/college-football-coaching-underachievers-overachievers-2018

Mr Grainger

September 24th, 2018 at 9:27 AM ^

I thought that seemed low as well. If everything is working like it did last week, Michigan should win this one running away. The only way NW hangs around is if Michigan plays poorly and/or makes a lot of mistakes.

Mike Damone

September 24th, 2018 at 10:18 AM ^

"Michigan should win this one running away"

I am betting Virginia Tech, LSU and Oklahoma said the same thing last week.

I have been in that high school stadium in Evanston too many times to make such statements.  It is on the road at Northwestern.  Play well, win the game by any margin, and get your ass back onto 94 heading east after the game as quick as you can.

Go Blue!!!

Mike Damone

September 24th, 2018 at 9:30 AM ^

I do not disagree that we should kick their ass.

But for the Mgoblog gamblers:  BEWARE! 

I have learned many things winning and losing at the Mirage sportsbook on Vegas trips, and one of them is this - NEVER bet against Northwestern in Evanston.  I have made more money getting Northwestern to cover at home than any other "general bet".

You have been fairly warned...

Go Blue!

heyyoujesson

September 24th, 2018 at 9:34 AM ^

-11 is pretty reasonable considering Michigan's road woe's. They should win this one going away but they still have a bit more proving to do before the road spreads get wider against the team's Michigan should beat.

umich1

September 24th, 2018 at 9:37 AM ^

Don’t care, just win.  Hopefully by a margin that doesn’t take days off of my life from high blood pressure.

I don’t care what the line is, I wouldn’t touch it.  This Northwestern team is grossly underperforming relative to expectations and Michigan is the kind of school that might result in Northwestern deciding to wake up.  We get everybody’s best shot.  Always have, always will.

Perkis-Size Me

September 24th, 2018 at 9:50 AM ^

That's a very fair line. I also could've even understood if it was, say, an 8-9 point spread. Michigan still has not proven that it can play well on the road. Not consistently anyway, and certainly not to teams that have a pulse. 

I know Northwestern doesn't appear to be good this year, but they're usually a very well-coached team and with two straight losses to bad teams, Fitzgerald and his team will look at this as a "save our season" kind of game. They'll pull out the stops. Don't make the mistake of assuming they'll be a pushover. 

I still have Michigan winning this game, but I won't be surprised if its a one score game at the half. 

DCGrad

September 24th, 2018 at 9:58 AM ^

Isn’t there a rule that if something seems like a sure thing you bet the other way? My first thought was I would take UM -11 every day of the week in this game. But something makes me not want to take this one. 

freelion

September 24th, 2018 at 10:01 AM ^

I think 13.5 just shy of 2 TDs is about right. Northwestern can try to muddy this game up and make it difficult for Michigan to blow it wide open with big plays. I could see this ending up anywhere from a 7 point game to a 30 point blowout if things go the right way for Michigan.

Perkis-Size Me

September 24th, 2018 at 10:38 AM ^

Chicago weather forecast calls for rain on Saturday. Still five days away, granted, but if that weather holds, we could be looking at a very close and literally muddy game. Both teams will probably stick to running the ball in that situation. Northwestern has a few playmakers on its defense, particularly at LB, so if they're stacking the box I don't think Michigan will be able to run the ball that well. 

I really hope the rain stays away on Saturday. For one, I'm going to the game, but two, I think we all have bad memories of watching this team play in the rain last year and how it didn't end well.