Recruiting Forum Football Talk [RIP 9.3.2019]

Status
Not open for further replies.
Watching SEC This Morning getting ready for SEC Media days. They're saying this is the best SEC East we've had in a long time. McElroy really high on Vandy and UK. They don't know what to expect from UT this fall (have to see it to believe it) but they're in agreement no one is going to be an "easy out" in East for first time in awhile.

Vols have a real shot to shock some folks this fall. Man, I'm starting to get jacked up for the season. 46 days...
Why is he so high on UK? Unproven at QB, and they lost three of the best players they have ever had. And we still whipped that ass.
 
That’s a good idea, but would take awhile to go through all 130 teams. I might re-up with cfb matrix and just ask them for the data.
How many years are you using to train the model? I've got some data for 2015+ I'll drop here.
 

Attachments

  • coach_years.zip
    27.7 KB · Views: 4
Why is he so high on UK? Unproven at QB, and they lost three of the best players they have ever had. And we still whipped that ass.
It's always the same thing with these analysts from what I can tell, they base most of their predictions on what they've seen in previous season(s). Vandy and UK have been decent lately and UT hasn't. See it to believe it...

But I'm with you fellas, I think UK and Vandy both take steps back. I think it's UGA then UF/UT/Mizzou fighting for next few spots and then SC/Vandy/UK fighting to stay off the bottom in the East. Those are my predictions for this year.
 
This for
How many years are you using to train the model? I've got some data for 2015+ I'll drop here.

You’re my boy, Blue!

This first run I just went back one year. I might add another couple of years. What I did wasn’t flashy - just a straightforward multiple linear regression. I need to learn Bayesian stats to keep up with the latest trends in research, and I feel like it would be ideal to apply to this stuff.
 
  • Like
Reactions: Vols410
It's always the same thing with these analysts from what I can tell, they base most of their predictions on what they've seen in previous season(s). Vandy and UK have been decent lately and UT hasn't. See it to believe it...

But I'm with you fellas, I think UK and Vandy both take steps back. I think it's UGA then UF/UT/Mizzou fighting for next few spots and then SC/Vandy/UK fighting to stay off the bottom in the East. Those are my predictions for this year.
I think it’s
Ga
Mizzou
Fl/TN
Everyone else
 
  • Like
Reactions: JacketVol
I agree with this to a degree. They do the same thing with race. Like the last FF movie where they made Johnny Storm black. WTH? The character was created in the '60's and has always been white. I understand diversity, but why not just create new interesting characters to achieve it? Changing established characters is lazy and uncreative imo.

The thing about James Bond is he's not quite an established character. It's more an identity. Maybe you could argue there should be a different numerical designation for a Jane Bond, but I'm not sure the point. The fact Bond is a spy has always led me to believe he wasn't born James Bond. Rather Bond is the name of whatever spy carries the designation of 007. In other words, it's an alias.
The book You Only Live Twice gives his parents' names as Andrew Bond and Monique Delacroix. Yes, I've read all the original Ian Fleming novels.
 
This for


You’re my boy, Blue!

This first run I just went back one year. I might add another couple of years. What I did wasn’t flashy - just a straightforward multiple linear regression. I need to learn Bayesian stats to keep up with the latest trends in research, and I feel like it would be ideal to apply to this stuff.
Nice! I'll continue to work on my coaches and get back to you when they're ready to plug in. Basically I'm just using my own multiple linear regression to map experience/talent to expected output on the field for offense, defense, and overall ratings, then measuring the over-achievement/under-achievement. I'm also trying to account for diminishing returns on talent to some extent; it's still impressive for the most talented team to be the best team in the country, and that coach should be better represented than simply average. That's my guess anyway.
 
  • Like
Reactions: RiotVol
It's always the same thing with these analysts from what I can tell, they base most of their predictions on what they've seen in previous season(s). Vandy and UK have been decent lately and UT hasn't. See it to believe it...

But I'm with you fellas, I think UK and Vandy both take steps back. I think it's UGA then UF/UT/Mizzou fighting for next few spots and then SC/Vandy/UK fighting to stay off the bottom in the East. Those are my predictions for this year.
#WGWTFA
 
What's your favorite book and which movie was the best at adhering to the book?
For the books, I really liked On Her Majesty's Secret Service because it got it a side of Bond that you didn't really see in the other books. The Spy Who Loved Me is another favorite. For movies, it'd probably be one of the Sean Connery ones, before all the gadgets. Dr. No or From Russia with Love maybe.
 
Make her a different number and she can own that . Otherwise the character will always be compared . Heck people argue over who is the best 007. Of course George Lazenby owns that distinction.
 
  • Like
Reactions: SweetasSoda
So, my interest in making predictive models is it seems every year teams get pumped up by a few talking heads, and most media has little else to do but repeat and discuss the things a few folks said the day before. It’s basically swarm behavior. That swarm behavior affects how casual bettors evaluate where to place their money.

If you come up with a decent model, they you should be able to wreck fools on the season total wins and first week scores, where you’re picking based on data and most everyone else is listening to talking heads.

This is why I spend more time reading Chatt’s posts than what most of the media writes this time of year. He’s doing what they should be doing - combing through the data and looking for patterns.
This is why there should not be polls until week 3 or 4. They base it off of last year then hate to be wrong so they hate to make corrections.
 
Might as well take a vacation from the SECN until the season starts. The heads are going to declare Saban king of intergalactic football, talk up the usual suspects, pick a few Cinderfellas and dismiss the rest as no-shots.

Not hard to guess where UT is going to fall.
 
Status
Not open for further replies.

VN Store



Back
Top