What's new
AllBuffs | Unofficial fan site for the University of Colorado at Boulder Athletics programs

This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

  • Prime Time. Prime Time. Its a new era for Colorado football. Consider signing up for a club membership! For $20/year, you can get access to all the special features at Allbuffs, including club member only forums, dark mode, avatars and best of all no ads ! But seriously, please sign up so that we can pay the bills. No one earns money here, and we can use your $20 to keep this hellhole running. You can sign up for a club membership by navigating to your account in the upper right and clicking on "Account Upgrades". Make it happen!

What's the difference between these games?

skibum

Thou shalt not groom Mary Jane
Club Member
So something has been bugging me the last few seasons. Basically, in games where X is true, I thought that we performed noticeably worse. I wanted to determine if what I was noticing was an actual difference, or if I was seeing things.

So here are your basic stats for 2016 games where X is not true, vs games where X true. The third column is the difference, and it's highlighted red if games with X are actually worse, and green if they were better. I excluded the FCS game, and all stats are average per game.

2016.jpg

And here are the sames stats for the 2015 and 2016 seasons combined (again excluding the FCS games):

2 seasons.jpg
So it looks to me like my gut instinct may very well be correct.

We play measurably worse at night*.


*I defined "night" as any game that started at 7pm Mountain Time or later.
 
Have you corrected for strength of opponent? Usually, it's the bigger games, and thus the better opponents, that get moved to the late game - USC, UCLA, ASU (prior to this year).

I know not how to correct for this, but I do think you need to take it into account
 
Interesting theory skibum, and who knows, maybe you're on to something.

That, or the
increased: points allowed, offensive turnovers, defensive passing yards allowed, defensive rushing yards allowed
and
decreased: points earned, offensive passing yards, offensive rushing yards, offensive third down conversion %, defensive third down stops %, and forced turnovers
might be contributors...to the worse record for X?

I realize you are probably suggesting that the issues listed above are worse at night, and that's a pretty interesting thing to point out. However, it could just be due to low certainty with the data due to the small sample size. Or it could be skewed due change in CU team quality this year vs. past years, as we seemed to play more late games before this year. It could also easily be due to quality of opponent or just due to the factors I listed above, as they are traditionally very good correlations to loses, so I'd just see the night thing as a spurious correlation without a good reason to think otherwise.)

Fun with numbers! We can look at this stuff now unlike the past where I would've been much more interested in forgetting how the season went and instead would've dove into basketball stats. This year has been so much fun.
 
Damn - I was of the opinion that it was a quality of opponent thing, but we played Michigan, Oregon, SC, Stanford, Utah and Wazzu during the day (along with Oregon State, ASU and CSU were 5 pm kicks), whereas we played Arizona and UCLA at night (along with UW).
 
most of the night games last two years have been against PZT teams. Late in the game it's past our guys' bedtimes, while the west coasters still have an hour to go. I'm sure that's it.
 
Damn - I was of the opinion that it was a quality of opponent thing, but we played Michigan, Oregon, SC, Stanford, Utah and Wazzu during the day (along with Oregon State, ASU and CSU were 5 pm kicks), whereas we played Arizona and UCLA at night (along with UW).
My first thought was that it was the strength of opponent (and this was why I immediately excluded the FCS games) but I went through the games, and we had both tough and easy opponents late and early.

The only reason I noticed it is because it seemed like almost every ****ing time I had to stay up past midnight, the team played like **** - even when they won.

I know in years past we've heard about weekday team practices talking place at 6 and 7 in the morning - when you play as many late games as we do, that just seems like a bad idea.

Although I don't know what the answer is in terms of class conflicts (I also don't knows if early practice times are a current thing).
 
I noticed, but didn't want to say anything because it's merely a coincidence and I haven't reviewed every single kickoff on the season. But it was enough that I noticed it. Sun shining brought us touchbacks on kickoffs. As soon as the sun set...no more touchbacks. Really couldn't even get it to the goalline. So obviously our team is powered by the Sun.
 
Interesting post, but I agree that sample size makes it impossible to draw valid conclusions. Three games in the first analysis is nowhere near a large enough sample to make a valid argument about night game performance, especially when those numbers are so greatly skewed by the UW game. The decision to make 7:00 MT the cutoff allowed you to lump the two 6:00 games in with the day games. The combined score of those games was 84 - 23. Throw in the 5:30 Utah game (which was played after the DST change) and you get 111 - 45. Pretty big turnaround to move those games to the day column.
 
I'm looking at CU's schedule on the website right now and it looks like UCLA, Arizona, and Washington were the 3 games of X. 1 home, 1 away, 1 neutral. Washington was obviously one of the toughest 3 teams on our schedule (we should know shortly how tough based on how they do in the CFP). UCLA was an admitted lackluster game. Arizona was a game where apparently they did a bunch of stuff defensively they hadn't shown all year.

I'm going with too small of a sample size to draw a true conclusion on this one.
 
2016 was a small sample, which is why I decided to look at 2015 too, and same problem.

The scatter plots and regression analysis also show a correlation with kickoff time as independent variable and various other measures as the dependent: the later the game, the worse the performance.

If I get really bored this off season I'm going to go full stat geek on the data and see where it leads - this was just a quick "does the data warrant a closer look" test. And 24 games over 2 seasons is probably enough to overcome "small sample" questions (which is even an incorrect criticism on its face: I'm using the entire population, not a "sample").
 
Back
Top