How the Coach's Poll Reacts to B1G Results

Submitted by Mmmm Hmmm on

A consistent week-to-week topic on this board is how pollsters treat teams given the previous week's results.  Because the Coach's Poll matters for BCS standings, I thought I would look at how B1G teams were treated each week.

I note that there is a significant flaw in this analysis, in that I do not look systematically at the results from schools in other conferences, which can often exert significant influence on how much room there is to move up/down and also how much attention voters pay to other games.  Also unexplored a comprehensive look at conference perception.  Still, I hope this is somewhat informative:

Michigan (Started #8)

  • 19 (0-1) - Lost on neutral field to then #1 Alabama 14-41 (fairly large drop, probably because Michigan looked like it was not competitive)
  • 18 (1-1) - Beat Air Force 31-25 (thank you Arkansas and Wiscy falling)
  • 17 (2-1) - Beat UMass 63-13 (likely more of a result of MSU losing than destroying a creampuff)
  • NR (2-2) - Lost @ ND 6-13 (Ugh...argh...still ranked third among vote-getters and two slots behind Northwestern)
  • NR (2-2) - Wife Day (actually dropped among vote-getters to 9th)
  • NR (3-2) - Won @ Purdue 44-13 (third among vote-getters, just ahead of Northwestern and MSU, in that order)
  • 23 (4-2) - Beat Illinois 45-0 (Biiiiiiig boost from killing baby seal Illinios)
  • 20 (5-2) - Beat MSU 12-10 (fairly solid jump for eeking by an unranked MSU team at home)
  • NR (5-3) - Lost @ Nebraska 9-23 (12th among vote-getters...ouch)
  • NR (6-3) - Won @ Minny 35-13 (3rd among vote-getters--bullish on Devin?)
  • 23 (7-3) - Beat Northwestern in OT 38-31 (?)

Nebraska (Started #16)

  • 14 (1-0) - Beat Southern Miss 49-20
  • 24 (1-1) - Lost @ UCLA 30-36 (UCLA did not come into the season with much hype, so they were probably underrated at the time relative to now)
  • 22 (2-1) - Beat Arkansas State 42-13
  • 20 (3-1) - Baby sealed Idaho St. 73-7
  • 20 (4-1) - Beat Wisconsin 30-27 (Not much of a bump for beating Wisky, even if the perception was that Wisky was bad and it was a close game at home)
  • NR (4-2) - Lost @ OSU 38-63 (Bo Pelini stands there with his large mouth bass face after getting killed on the road by undefeated OSU, still unable to comprehend why his team has less votes than Michigan, Northwestern, MSU, and OHIO (!))
  • NR (4-2) - Bye (Nebraska  remained behind NW among vote-getters, 9th to 6th for Northwestern)
  • NR (5-2) - Won @ Northwestern 29-28 (Still only 5th among vote-getters, perhaps punishment for Northwestern snatching defeat from the jaws of victory in Cubs-like fashion)
  • 21 (6-2) - Beat Michigan 23-9 (apparently lots of credit for beating Bellomy)
  • 18 (7-2) - Won @ MSU 28-24 (not sure I can explain this one)
  • 14 (8-2) - Beat Penn St. 32-23 (I guess the "lucky" call did not enter voters' minds)

Wisconsin (Started #12)

  • 13 (1-0) - Beat Northern Iowa 26-21 (penalized slightly for close loss)
  • 22 (1-1) - Lost @ Oregon State (NTOSU) 7-10 (even after its victory, NTOUS was 9th among non-top 25 vote getters)
  • 24 (2-1) - Beat Utah State 16-14 (Not penalized too much for almost losing to non-BCS school)
  • 23 (3-1) - Beat UTEP 37-26
  • NR (3-2) - Lost @ Nebraska 27-30 (No votes for you!)
  • NR (4-2) - Beat Illinois 31-14 (a couple of votes, behind Nebraska, Texas Tech, and Duke (!!))
  • NR (5-2) - Won @ Purdue 38-14 (one ahead of Nebraska)
  • NR (6-2) - Beat Minny 38-13 (third-highest vote-getter)
  • NR (6-3) - Lost to MSU 13-16 (Votes? #HALOL)
  • NR (6-3) - Bye (taking the weekend off brings no votes)
  • NR (7-3) - Won @ Indiana 62-14 (2nd among vote-getters--boost for controlling destiny in Leaders?)

MSU (Started #13)

  • 11 (1-0) - Beat then #23 Boise St. 17-13 (quality win at home, but looks like MSU was penalized somewhat from a perception of Boise State looking overrated at #23)
  • 10 (2-0) - Won @ Central Michigan 41-7
  • 20 (2-1) - Lost  to ND 3-20
  • 18 (3-1) - Beat Eastern Michigan 23-7
  • NR (3-2) - Lost to OSU 16-17 (fifth among vote-getters despite losing to undefeated OSU by a point)
  • NR (4-2) - Won @ Indiana 31-27 (see above)
  • NR (4-3) - Lost to Iowa 16-19 (No votes for you!  The Fall comes before the Threat!)
  • NR (4-4) - Lost @ Michigan 10-12 (Pride comes before the lack of votes)
  • NR (5-4) - Won @ Wisconsin 16-13 (Still no votes)
  • NR (5-5) - Lost to Nebraska 24-28 (All your votes are belong to somebody else)
  • NR (5-5) - Bye (Hide yo couches)

Northwestern (Started NR)

  • NR (1-0) - Won @ Syracuse 42-41
  • NR (2-0) - Beat Vandy 23-13 (ranked well below Oregon State)
  • NR (3-0) - Beat BC 22-13 (still 5th-ranked among those receiving votes but not in top 25)
  • NR (4-0) - Beat South Dakota 38-7 (I know all three BCS team wins were bad, but 4-0 with only one seal clubbing!)
  • 22 (5-0) - Beat Indiana 44-29 (Finally...some respect!)
  • NR (5-1) - Lost @ Penn St. 28-39 (just behind Michigan, good enough for 4th among vote-getters)
  • NR (6-1) - Won @ Minnesota 21-13 (6th among vote-getters, which is a backslide from the previous week)
  • NR (6-2) - Lost to Nebraska 28-29 (No votez iz sad)
  • NR (7-2) - Beat Iowa 28-17 (9th among vote-getters)
  • NR (7-2) - Bye (5th among vote-getters; better study up for the next game rather than celebrating those votes...)
  • NR (7-3) - Lost @ Michigan 31-38 (8th among vote-getters)

Conclusions based on limited information:

  1. Ranking among those receiving votes is a crap-shoot.  There seemed to be some strange sequencing among B1G teams if looking at record-to-date, as far as I could tell.
  2. Northwestern was underrated early on.
  3. Wisconsin was punished for disappointing.
  4. B1G is not very good, but the voters are not doing it very many favors.  Aside from a few nicer-than-expected bounces, a lack of ranked teams has probably lead to depressing the rankings of the teams that have been winning.
  5. The early season dumpster fire at Penn State and sanctions at OSU hurt the B1G profile in the polls.
  6. And, most important of all: The B1G would probably benefit from beating a few good eams OOC early and then riding on voter assumptions into league play.

Thoughts? Criticism? Vehement disagreement?

(P.S. Apologies in advance for errors; I will try to correct them as they are identified)

Belisarius

November 11th, 2012 at 4:30 PM ^

#6 is what I've been thinking about most. But it actually brought me to the conclusion that the Big Ten should abandon the lock-step pre-conference game schedule. I notice, if you lose games early to opponents from the same conference, ieven if it's an upset, it creates the perception of depth in the conference rather than of dissapointment. That helps with perception later on.
In other words, winning the first few weeks seems to matter most, but if you lose, it's better to lose in-conference that out.

Mmmm Hmmm

November 11th, 2012 at 4:33 PM ^

I have not analyzed it, but I would not be surprised if the SEC's pattern of having the second game of the year be a conference game helps them somewhat, perception-wise.

Also, to be a little clearer, much of the voting (including the unranked vote-getters) cannot to my eye be explained by B1G conference results alone.  Of course, that is what we would expect--over 100 other teams are playing, many of which have realistic shots at being ranked or getting votes at various times of the year.  And that is even assuming that voters have and utilize perfect information about this season's performance alone, and can take into account injuries/bad calls/etc., which is a mighty unrealistic assumption.

justingoblue

November 11th, 2012 at 5:16 PM ^

which I doubt you could get a good enough sample for in one season is name recognition. If #8 Northwestern got blown out by Alabama in the first week, would the drop be similar? If that same team beat AFA and pounded UMass while losing to ND, do they get back into the polls?

I tend to think it's less of an issue than a lot of people seem to, but I'm sure it has some bearing on the votes received by each school.

corundum

November 11th, 2012 at 4:39 PM ^

This is an interesting post that I am glad you put together. I think you should do another one at the end of the regular season and maybe after the bowl games as well. The information isn's too telling at the moment, but future analyses could be made concerning preseason perceptions / initial poll accuracy, the most underrated / overrated teams in the conference, and why teams overachieved.

Mmmm Hmmm

November 11th, 2012 at 5:10 PM ^

Thank you--I might try to do that by the end.  Also, if PSU is ranked at the end, I will add them in.

I thought about doing the Harris Poll instead because I anticipated finding some really off-the-wall results (Lou Holtz is/was a voter foergodsakes).  It took more time than anticipated to put together, though, so I might only do an update.

PurpleStuff

November 11th, 2012 at 4:45 PM ^

When compared with the SEC and the Pac-12 especially, the Big Ten's resume in big games (BCS bowls, big non-conference games) just sucks and has for a while.  OSU did a nice job in BCS games overall under Tressel, but had some high profile embarassments (blown out by Florida and LSU in title game, got crushed at USC in 2009).  After them, where do you go for a big win?  Wisconsin lost the last two Rose Bowls (their last win is from the 1999 season).  PSU beat a 4-loss Texas team in the Fiesta Bowl in '96 and a 4-loss FSU team in the Orange Bowl in '05 and a 3-loss Oregon team in the Rose Bowl in '94 in the last 20 years, but lost to SC in their only other big bowl game against an actual high quality opponent.  Michigan edged a shaky Virginia Tech team and lost to SC twice and Texas in the Rose Bowl, and to Oregon twice in the regular season.  You basically have to go back to Tom Brady's tenure for anything close to an impressive win on the national landscape (we haven't finished in the top-5 since that 1999 season).  Purdue lost decisively in their one Rose Bowl appearance with Drew Brees and MSU and everybody else haven't done dick in the last 15 years.

In the SEC, Florida (twice), Tennessee, LSU (twice), Auburn, and Alabama (twice) have all won national titles in the last 15 seasons.  In the Pac-12, SC has won the national title (twice), Oregon has won two BCS bowl games and gone to the title game to finish in the top 4 three times since 2000, Washington won the Rose Bowl and finished #3 in 2000.  Stanford won the Orange Bowl and finished #4 in 2010.  Oregon State won the Fiesta Bowl and finished #4 in 2000.  Arizona State was a play away from a national title in 1996.  Utah has had two undefeated BCS bowl winning seasons since 2004 finishing at #4 and #2 in the AP Poll. 

Until someone other than OSU does something impressive, those conferences are going to be justifiably looked at as superior.

Mmmm Hmmm

November 11th, 2012 at 5:13 PM ^

Fair point--one I acknowledge but could not really test for without doing this for all conferences.  That was a little much for one afternoon but that would be interesting, especially to compare Big 12 BCS vs Poll (BCS always seems to love the Big 12 for reasons I do not know/understand), and to look at SEC vs Pac-12 and ACC.  I have a feeling that the conference perception would play out, but I am thinking there may be other noise going on depending on who is voting and their agendas.  My further guess is that the results of a wider study on a full season of information is that there is still something other than records going on week-to-week but that it would still be hard to fully identify.

JohnnyV123

November 12th, 2012 at 4:40 AM ^

I agree with your opinion but don't think you're being fair to the B1G either.

"Michigan beat a shaky Virginia Tech team" last year. I know a lot of people thought they were overrated last year but they lost to exactly one other team. Yeah, they got completely outclassed in those games but regardless, one team. I remember on here people thought it was dumb for me calling that they were going to get an at large BCS bid. It was a nice win against a good team and great program with something like 9 straight 10 win seasons.

Michigan State beat a really good Georgia team as well. Yeah, in both games a lot of luck was involved but there was also a lot of good play.

Michigan over Florida....yeah yeah knock it all you want with Florida having three losses by then but this was the team that experienced The Horror versus the Heisman trophy winner in a game that both teams played well (okay maybe not on defense). This meant something nationally at the time it was just forgotten about quickly by the next Michigan season.

Ohio State over Oregon in the Rose Bowl. Ohio State over #2 Texas in the regular season the year they went to the champions game....and they finished the year ranked #2. Their national championship regardless of the ridiculous luck they had still happened.

How about the actually happened but vacated wins from Penn State who had a solid last decade of bowls including a win against LSU.

And about the Pac 12? They have won zero BCS national championships, one behind the B1G. If you're counting Arizona State's almost national champion in 1996 (even though you could count many of the B1G's impressive almost wins like Michigan vs. Texas, Northwestern vs. next year's national champions Auburn, etc) I would like to include Michigan's 1997 national championship. Though I don't think you should include a team like Utah who was not a Pac 12 team at the time doing the equivalent gives the B1G a third national champion in Nebraska.

I'm completely with you that the B1G is seen as and deserves to be seen below the SEC but not as much sold on the Pac 12. I'll give it to the Pac 12 if they manage to win the BCS championship this year.

polometer

November 11th, 2012 at 4:55 PM ^

coaches know jack about teams that they don't face.  I know this sounds a bit harsh, but does anyone honestly think a head football coach, after all his game planning, after all his practices, after all his scheming and preping, after all his recruiting and booster duties--really goes home and watches film from teams that he will never face this year?

 

I just don't understand the rationale behind the coaches poll.

TheTruth41

November 11th, 2012 at 9:09 PM ^

Stating the exact thing you just said.  Coaches don't watch games other than those from their own team and teams they'll face (but probably not up to the week they actually play them).

I'm sure there's some friendly adjusting as well (not dropping a team you play too low but low enough where you'd be above them).  Seems like it creeps on insider trading.

One Inch Woody…

November 11th, 2012 at 5:16 PM ^

The Big 10 was not very good at all in the OOC season, but right now, the traditional powers are all playing pretty well... that OOC season is making the conference look worse now than it actually is.

SHub'68

November 11th, 2012 at 10:15 PM ^

By Stewart Mandel covers the coaches poll and says basically what's said in comments here - coaches can't evaluate teams they don't see, rank teams according to their own agenda, but mostly just fill it out at the end of the night based on final scores. Also, it mentioned Purdue coach Joe Tiller, who had his son fill out his pre-season ballot; his son who was a junior at Wyoming at the time. See "Bowls, Polls and Tattered Souls Tackling the Chaos and Controversy That Reign Over College Football" pages 43 & 44, 2007 by Stewart Mandel.