Michigan Fan Satisfaction Index - Cinci Results

Submitted by OneFootIn on September 13th, 2017 at 9:37 AM

The results are in. On a scale from 0 to 100, where 0 is the worst you ever felt about a game and 100 is the best you ever felt, our 93 respondents averaged a 59.9. There was a reasonable amount of variation in responses – the standard deviation was 16.5 – but the most common response was 60, with only a couple handfuls of folks straying below a 40.

In technical terms, the crowd ranked this victory somewhere around warm milk – better than losing, but not much to get excited about, especially given how sloppy and uninspired the offense and special teams looked.

Figure One Cincinnati Feelings

The hive mind is more positive about the young season – the average feeling about the season this week is a 77.2. The clustering here was tighter; the standard deviation was 10.2 the median was 80, and only one wag reported a season rating of less than 60.

This seems pretty reasonable. On the one hand, Michigan is undefeated, has what will clearly be another great defense, and appears to have a plethora of young talent emerging on offense. On the other hand, Speight’s struggles, continuing concerns about the offensive line, and the knowledge that Wisconsin, Penn State, and Ohio State will present stern challenges probably keeps this figure from getting much higher for now.

Figure Two:  Feelings about the 2017 Season (for now)


Themes, Thoughts, and Trends

One of the things I’ve noticed over the years is how the tenor of board discussions often shifts after the post-game columns go up and the weekend’s full slate of games has been digested and analyzed. Translation: Brian has to talk a lot of us off the ledge when the torch and pitchfork brigades start forming, but he also uses, like, numbers and analysis, to provide a serious and thoughtful take on what’s happening.

This makes sense for many reasons. Most obviously, taking a reading of your emotional state right after a game is likely to lead to higher highs and lower lows than you would see after a few days, when logic kicks in. Secondly, as sociologists love to remind us, the interpretation of events is a collective process. We all look to others, especially experts and authorities, to help us make sense of things.

To test this hot take/cold take idea I compared the responses collected on Sunday to the responses from Monday and Tuesday. Confirming the strong emotions hypothesis, the hot take respondents recorded both the highest high for the game (100 compared to 90) and the lowest low (20 compared to 30).

The average distance between the hot and cold takes was also surprisingly large – at least to me. Sunday respondents (n = 53) averaged 57.8 for the game and 75.3 for the season, while later respondents (n = 40) averaged 62.5 for the game and 79.9 for the season. It appears that with a day or two of cooling off, the game didn’t seem as bad as it had in the immediate aftermath. It will be interesting to see if this hot/cold differential holds throughout the season and what direction it tends to take.

Figure Three: How Do You Take Your Takes?

Sadly, I didn’t do a survey last week. But I think we can all agree that the collective feelz was a lot higher than 59.8 for the Florida game and that the irrational exuberance stoked by that curb stomping would have resulted in a season rating higher than 77.2 for most of us (it would have for me, certainly). In following weeks I will start reporting game-by-game comparisons and seasonal trends.



Bo Glue

September 13th, 2017 at 9:46 AM ^

I looked at that first graph and thought it strange there were no zeroes. Then I realized you just skipped them from the axis. Makes more sense now.

Just out of curiousity, who is it you survey to get these numbers?


September 13th, 2017 at 10:01 AM ^

The data comes from a survey I posted in a diary last Sunday morning after the game. Presumably most of the respondents are regular mgoblog readers.

And yes, good eye - there are no zeros in that chart simply because no one rated the game that poorly. 

EDIT: And just as a note: I will be posting the next survey on Sunday morning after the Air Force game, so be sure to take it.


September 13th, 2017 at 9:48 AM ^

This work certainly validates my personal assessment that the combination of high raw talent, youthful inexperience yet great coaching will likely make this a wild ride. I shudder to think what the past two regimes would have done here. They certainly would have dragged the whole chart to the left.


September 13th, 2017 at 1:18 PM ^

I was a 70 rating on Cinci and a 100 rating on the year. 

Basically Cinci being a 20+ point victory in a let down game isn't that big of a surprise. Sure we didn't blow them out of the water. But that was never a game like it could have been from 2007-2014. 

Tell me 14 days ago we would be 2-0, with both not being generally close, and I wouldn't be blown away, but I would believe our season is 100% on track. I still don't see us losing before PSU and I have faith Speight can get it together by/before then. To me that it a season on track. 

Swayze Howell Sheen

September 13th, 2017 at 3:44 PM ^

buckets are your friend.

here is a bucketized fig 2, for example:

<= 50 *
     60 ************** 
     70 ******************************
     80 ************************************
 >=90 ************
Much easier to read.
thanks for getting some data!


September 13th, 2017 at 10:20 PM ^

Very interesting, thank you for the good read. One thing occurred to me while reading. Is it possible that instead of logic kicking in as days go by, that maybe our fandom kicks in, and our eternal optimism seeps in, like what happens during fall camp and the start of the season? Could it be that our in-game emotions and post game gut feelings aren't totally off base?


September 13th, 2017 at 11:42 PM ^

That's a good hypothesis that probably can't be answered until after the next more-outrageous-than-expected victory.  If Michigan beats Staee 79-0, going for two on the last two scores to outpace last year's Rutgers total, that should give you a good test case to see if the values are always increasing after a game or if it's more of a case of regression to the mean.