Menu
What's new
By:
Filters
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!

College Football Countdown: #101 Colorado

'Tini said:
Sorry, not buying that a subjective approach is the best way to approach year+1 projections
Click to expand...

You understand that any statistical model that you write down is inherently subjective, right? How are you going to measure all the relevant variables? Is the standard linear model the right one to use (i.e. is there a linear relationship between the objects that you're modeling?) What belongs in the regression and what doesn't? Are there important unobservables?

In projecting the quality of college football teams year-over-year probably the majority of relevant variables are unobserved. Good luck making sense of that with straightforward data analysis. In-season most of these things are static, so they don't matter, but I'd guess that even the most sophisticated computer rankings of teams come with caveats about injuries, etc.
 
buffs04 said:
You understand that any statistical model that you write down is inherently subjective, right? How are you going to measure all the relevant variables? Is the standard linear model the right one to use (i.e. is there a linear relationship between the objects that you're modeling?) What belongs in the regression and what doesn't? Are there important unobservables?

In projecting the quality of college football teams year-over-year probably the majority of relevant variables are unobserved. Good luck making sense of that with straightforward data analysis. In-season most of these things are static, so they don't matter, but I'd guess that even the most sophisticated computer rankings of teams come with caveats about injuries, etc.
Click to expand...
So we should just never attempt to model reality for the future? Should Vegas stop using models to predict the future season and just say "well, I think the x team win o/y should be y games"? Surely that will help the bookies make a lot of money. Of course there are inherit flaws in any model, but there are a whole helluva lot more flaws in someone just listing teams based on what they think based on no (or very few) quantitative stats.
 
Yeah I never understood why these preseason rankings always have to be so flawed, if you're not going to predict the rankings accurately then don't publish one at all.
 
TimmyDUBs said:
My statistical analysis shows we are accurately ranked 83.5
Click to expand...

My statistical analysis of your statistical analysis shows that you shouldn't have just objectively averaged the other two statistical results without subjectively adding in some Kool-aid factors such as Sink girl hotness and Mascot speed to slobber ratio. Adding those factors result in a solid improvement to 83.37.
 
DexBuffsteen said:
My statistical analysis of your statistical analysis shows that you shouldn't have just objectively averaged the other two statistical results without subjectively adding in some Kool-aid factors such as Sink girl hotness and Mascot speed to slobber ratio. Adding those factors result in a solid improvement to 83.37.
Click to expand...

Hold up...what are the p values?
 
You must log in or register to reply here.
Back
Top