Debunking the Bye Week Myth

Submitted by Soulfire21 on

It seems reasonable to bring this up every year, particularly when our next opponent is coming off of a bye week.

Logic Says:
Teams that get a week of rest and an entire extra week to prepare have an advantage over their opponent because their opponent didn't get the extra rest or time to prepare.

Reality Says:
Bye weeks seem to hurt more than they help.  Since 2002 (to 2010*), teams of the six BCS conferences have an overall win pct of 0.480 when coming off of a bye week.  The Big Ten teams in particular struggle when coming off of a bye.  From 2002-2010* Big Ten teams are a combined 17-32 when coming off of a bye.  This is good for a 0.35 win pct.

Notables:
Ohio State went 56-10 in Big Ten play from 2002-2010* but was 1-3 when facing a conference opponent after a bye.  This includes their 2004 loss to Northwestern, their only loss to the Wildcats since 1971

Penn State and Iowa are also 1-3 when coming off of bye weeks and playing conference opponents

37 of the 65 major-conference teams have a losing post-bye week record.

Edit:
Under Mark Dantonio, MSU has a 1-1 record coming off of a bye week

In 2008, they lost to PSU 14-49
In 2010, they beat Purdue 35-31
 

[*This information was summarized from this article, read the full version from the Wall Street Journal; info current through Oct. 14, 2010]

 

BrickTop

October 10th, 2011 at 12:53 PM ^

my only question regarding the validity of this information would be whether or not teams tend, on the whole, to schedule better opponents for the week when they are coming off of a bye. In this case, it might be true that the winning percentage would be lower as a function of better competition being scheduled at that time rather than the bye. Under these circumstances, it might also be true that the bye actually increases the winning percentage albeit not above 500 but better than it would have been. Just wondering, data integrity is one of my research interests.

 

As a side note, WOW we are the WORST conference when coming off of a bye. Crazy stuff.

joeyb

October 10th, 2011 at 1:02 PM ^

This is a great point, which I missed. Just because they have a losing percentage after a bye week does not mean that it did not increase their chances to win. I think what you would have to do is look at the record against the spread after a bye week compared to playing in consecutive weeks. You will see a little less than a 50% chance of winning (ties taken into consideration if you go back far enough) when playing in consecutive weeks, but you might also see that after a bye week, it increases to 55-60%, which would be significant enough to try to place a bye week before a good opponent (more specifically, an opponent that is slightly better than you).

jmblue

October 10th, 2011 at 10:10 PM ^

would be whether or not teams tend, on the whole, to schedule better opponents for the week when they are coming off of a bye.

For most schools this is a moot point; their conference sets the schedule, not them.

If Big Ten teams are terrible after a bye, that would seem to be evidence that they are not helpful, because almost all of those post-bye games are coming against other Big Ten schools.

jmblue

October 10th, 2011 at 11:57 AM ^

I can't confirm this, but someone said recently that MSU under Dantonio is 1-5 after a bye.  A bye has the big advantage of getting a team healthier, but it seems to leave teams rusty as well.

MichiganStudent

October 10th, 2011 at 11:59 AM ^

This is some great information. I would have thought it would have been the opposite. Thanks for making me feel better about our chances. 

jg2112

October 10th, 2011 at 12:00 PM ^

Bye weeks occur in the playoffs when you automatically advance to a future round.

Michigan State did not get a bye this past weekend from anything - they're still playing 12 games this year.

Michigan State, in fact, is playing this Saturday after an "OPEN WEEK."

jg2112

October 10th, 2011 at 12:18 PM ^

The dictionary definition of a "bye" is completely at odds with what the NFL and college football calls a "bye." Just because the NFL and MgoBlue tells one what something is doesn't mean it's right.

No one claims teams in baseball, basketball, soccer or hockey have "byes" when they do not have games on dates when other teams are playing. When Michigan has a weekend off this season in hockey or basketball, it won't be called a "bye," it'll be called an "open date," or probably the most appropriate title, an "off day."

joeyb

October 10th, 2011 at 12:38 PM ^

I went to wikipedia to see what it said, because it usually explains how or why a term gets used improperly. It didn't do that, but it did mention round robin tournaments with an odd number of teams will give each team a bye week. I'd imagine that's how it came about because a lot of conferences used to play round robins, and some still do. With out of conference open weeks and with larger conferences, that no longer applies, but I think the term sticks around due to familiarity.

Mr Miggle

October 10th, 2011 at 12:54 PM ^

That one dictionary definition is hardly exhaustive of all accepted usages of the word "bye" in sports.

Different sports use different terms for what are essentially the same things. There's nothing unusual about that, nor is it limited to sports.

Look up the definition of "basket". Does it say anything about a made field goal in basketball?  If not, does that mean all of us that use the term that way should stop?

 

Brhino

October 10th, 2011 at 12:07 PM ^

Michigan State's bye-week results in the Mark Dantonio Era:

 

2010 (11-2):

Bye week Nov 13

Next week: Def. Purdue 35-31

Comment: way-too-close game against terrible team

 

2009 (6-7):

No bye week

 

2008 (9-4):

Bye week Nov 15

Next week: Lost to #7 Penn State 18-49

Comment: crushed by a superior team

 

2007 (7-6):

No bye week

 

Conclusion: not much to go on here. 

Brhino

October 10th, 2011 at 12:19 PM ^

going back even further:

2004: State gets a bye in the week prior to Braylonfest.  So, they lost, but argueably a 5-7 team taking a ranked 9-3 team to OT demonstrates a better-than-expected result.

2003: Another bye before Michigan.  This time #9 Michigan State loses to #11 Michigan 27-20 in East Lansing.  You might call this a bit of a letdown.  State would go on to lose 3 of its next 4 while Michigan would finish 10-3.  So in hindsight, again, this was a close game against a superior Michigan team.

2002: This 4-8 team gets a bye week before #17 Iowa pastes them 44-16.  Bad team gets whupped by good team, not much to conclude there.

2001: TWO bye weeks for this 7-5 team!  First, state beats a Notre Dame team that will finish 5-6 by a score of 17-10 after a bye.  Next, a 31-28 win over fellow 7-5 team Iowa.  Three average teams in two close games.  Not sure what to conclude there.

1998: Nick Saban takes the spartans to a 6-6 record, including two losses after bye weeks: a 29-17 loss at Michigan (who finishes 10-3), and then a 19-18 squeaker against Minnesota (5-6).

 

Mr. Yost

October 10th, 2011 at 12:31 PM ^

Me neither...but this game is the "tell all" game. This is the one that we're going to say "are we back?"

We don't have to win pretty, we don't have to win big. IMO, we just need a W. If Denard has to run 45 times to do it, so be it. We need to win and that's it. After that, I'll know we can roll with anyone on our schedule.

Remember those days, those years where you'd look at the schedule and go "what do we have to do to be 12-0?"

We seem to have forgotten that...we now look and say "8-4 seems right, 9-3 would be awesome"...we win this game and I'll be a lot closer to that "what do we have to do to go 12-0?" mentality.

M-Wolverine

October 10th, 2011 at 12:15 PM ^

Going INTO a bye week, the game before always seems to be abnormally brutal. It's the extra week off, and two weeks to stew about a loss, with no game for redemption the next week that seems to happen more often to Michigan (a program that doesn't lose that much to begin with).

WolverineHistorian

October 10th, 2011 at 12:19 PM ^

Situations are different for any team during any year.  Their circumstances vary.  Remember when Joe Paterno insisted that his teams get bye weeks before playing Michigan and Ohio State as part of the agreement to join the Big Ten?   Before the RichRod years, Paterno was 3-1 against Michigan with a bye week and 0-9 against us without one.  It's not always doom and gloom. 

Like I mentioned in the other thread, MSU has a history of bye weeks before us (93, 98, 03, 04).  We were 3-1 in those games but those 4 MSU teams weren't exactly world beaters going 6-6, 6-6, 8-5 and 5-6.  I think it's safe to assume this years MSU team is better than those other 4 that got the extra week.  So, yes.  It does worry me.  They may have barely escaped an awful Purdue team last year but they're obviously going to be much more fired up to play us than the Boilers.   

FreddieMercuryHayes

October 10th, 2011 at 12:21 PM ^

I think bye weeks kill momentum. It can be bad, but it can also be a good thing. Basically, if you're coming of a loss, a bye can help you re-focus and regroup. But If you come off a win, it can stunt that momentum. It won't be a huge help, and anything can happen, but this bye week may help kill the momentum MSU had coming off their program win against a .500 team.

Jivas

October 10th, 2011 at 12:32 PM ^

I did a quick Google search to see what this data would look like for the NFL; while not perfectly analogous, the advantages a team has coming off a bye week should be similar for an NFL team than for a college team.

I found an article from 2009 here.

Most relevant bit: "Since the N.F.L. introduced bye weeks in 1990, the overall record of teams playing after them (309-276-1, 52.7 winning percentage) suggests the advantage is modest at best."

So, with respect to MSU's bye week advantage, perhaps what we are fearing is fear itself.

joeyb

October 10th, 2011 at 12:47 PM ^

Another interesting thought is that the extra planning for this game might hurt MSU. They have to practice defending our 2 QB formations, but if they overthink it too much during the game, they are going to freeze when something they didn't practice shows up. I expect us to drop a new formation or two on MSU and some new plays out of the other formations. I could see us using the entire playbook if necessary, then just reload with new formations in the bye week.

madtadder

October 10th, 2011 at 12:52 PM ^

I think a more telling stat, though more difficult to research, would be how B1G teams perform against the spread after their bye week. If Michigan were playing Minnesota after their bye week that would have a huge difference on the record versus if we were playing Wisconsin after the bye week. The ATS would normalize it. Jamiemac to the rescue?

turd ferguson

October 10th, 2011 at 3:43 PM ^

I almost agree with you on this. It would help tell us whether Vegas/gamblers overvalue the bye.
<br>You're right that scheduling makes a big difference. I remember Penn State consistently getting bye weeks before they played us, and I could see teams tending to get tougher matchups after a bye. The problem with just looking at records against the spread is that the Vegas line could internalize whatever advantage/disadvantage comes from having the previous week off.