Started By
Message

re: Predicting every SEC Teams schedule using ESPN's FPI Matchup Predictor

Posted on 8/7/19 at 10:35 pm to
Posted by AUFan2015
Oneonta, Alabama
Member since Oct 2013
1853 posts
Posted on 8/7/19 at 10:35 pm to
If a team had the higher percentage I counted that as a win for that team. Do I think this will end up being each team's record?

No, the matchup predictor will obviously change as each game gets closer and, then there are things like injuries, upsets, etc. that will determine a team's record.

This was a just for fun thread since it's still the off-season, to get a conversation started on each teams expectations for the upcoming season.
Posted by mulletproof
Shambala
Member since Apr 2013
4672 posts
Posted on 8/7/19 at 10:44 pm to
I read your op and all the responses. It is just dumb in my opinion. Sorry I hurt your feels, if you think it is smart, more power to you.
Posted by fibonaccisquared
The mystical waters of the Hooch
Member since Dec 2011
16898 posts
Posted on 8/7/19 at 10:48 pm to
quote:

quote:
(3-5 SEC)

So long, Gus!


I called this weeks ago...

LINK
Posted by AUFan2015
Oneonta, Alabama
Member since Oct 2013
1853 posts
Posted on 8/7/19 at 10:50 pm to
Didn't hurt my feelings at all. Just think it's dumb to post in a thread on something you think is stupid. I'd just skip over it. To each his own. Let's leave it at that.
Posted by fibonaccisquared
The mystical waters of the Hooch
Member since Dec 2011
16898 posts
Posted on 8/7/19 at 10:50 pm to
quote:

Feel like Tennessee is a bit inflated


Most of the algorithms like them due to favorable schedule and returning like 90%+ of their production from last year... New coordinators in year 2 on both sides of the ball definitely seems to be getting ignored.
Posted by TrueLefty
St. Louis County
Member since Oct 2017
14980 posts
Posted on 8/7/19 at 11:06 pm to
quote:

So, UGA has a better chance of winning against Notre Dame than winning at Tennessee. That should tell the national talking heads enough.


Do not forget that Missouri beat Tennessee big the last 2 years! and this year is going to be a loss? No way!
Posted by mutigerz
Ballwin, MO
Member since Nov 2013
787 posts
Posted on 8/8/19 at 11:53 pm to
LOL at Florida with a 56% chance of winning at Mizzou
Posted by chawk195
Spartanburg, SC
Member since Feb 2015
1174 posts
Posted on 8/9/19 at 12:10 am to
I don’t see SC at 5-7, not with everything we have coming back. This is also the same FPI that says we’re a top 20 team saying we’ll go 5-7. I see 7-5, maybe 8-4. 9-3 and Muschamp is coach of the year with this slate.
Posted by UKWildcats
Lexington, KY
Member since Mar 2015
17203 posts
Posted on 8/9/19 at 7:33 am to
The lone takeaway is that ESPN'S FPI is a terrible algorithm.
Posted by CoachDon
Louisville
Member since Sep 2014
12409 posts
Posted on 8/9/19 at 7:36 am to
quote:

KENTUCKY (5-7) (1-7 SEC)


I'll perma-ban bet anyone on that prognostication. It's asinine and is the same ole same ole LAZY and uninformed reporting that happens every season.

Posted by Realistic Ag
Member since Jun 2014
1896 posts
Posted on 8/9/19 at 8:14 am to
quote:

The lone takeaway is that ESPN'S FPI is a terrible algorithm.

Nah. But, treating a 51% and 98% as the same 1 win doesn't make a lot of sense.
Posted by CrabInMyShoeMouth
Member since Jul 2016
2486 posts
Posted on 8/9/19 at 8:29 am to
quote:

So, UGA has a better chance of winning against Notre Dame than winning at Tennessee.


I don't think ND would finish any better than 8-4 with TN's schedule. Brian Kelly is 23-16 on the road as the head coach at ND, and 4-10 on the road against ranked opponents.

Not really a stretch when you consider UGA has ND at home and TN on the road.
Posted by Arch Madness
Charleston
Member since Jan 2018
1059 posts
Posted on 8/9/19 at 9:03 am to
You know that’s not how statistics work? You’ve got to take all probabilities into consideration and factor that into a team’s chances using a cumulative probability distribution. Otherwise it’s flawed, ie, a team with 10 games to win at 51% would be considered undefeated in your model
Posted by UKWildcats
Lexington, KY
Member since Mar 2015
17203 posts
Posted on 8/9/19 at 11:56 am to
quote:

I'll perma-ban bet anyone on that prognostication. It's asinine and is the same ole same ole LAZY and uninformed reporting that happens every season
Seconded. I will join Coach and take another ban bet with a second taker. UK is looking at 7 to 9 wins.
Posted by AUFan2015
Oneonta, Alabama
Member since Oct 2013
1853 posts
Posted on 8/9/19 at 1:41 pm to
Here's the methodology I used.

0-40% chance to win is considered a loss

40-60% chance is a toss up so I gave the team with the highest percentage the win

60%+ is considered a win in my opinion.

Again this is a just for fun thread mostly meant to encourage discussion on how each team will do this season.

Like I said to another poster who had an issue with this, the FPI predictor will change as the games get closer, and also there are things like upsets, injuries, team isn't as good as predicted or vice versa, etc. that will change the results; so right now a game that is a loss might turn into a win and vice versa.
Posted by RatRodDawg
UGA & USC alum/Los Angeles, Calif
Member since Nov 2018
2494 posts
Posted on 8/9/19 at 1:47 pm to
I hope I'm wrong, but it seems our chances of winning per %-wise should be a bit lower on some of these...Notre Dame, UT, UF, AU and A&M are all going to be tough for us.
Posted by VFL1800FPD
Nashville, TN
Member since Aug 2012
9058 posts
Posted on 8/9/19 at 1:53 pm to
quote:

If Pruitt goes 9-3, he should get NCAA CFB COTY Award.


We don't agree on much but this is one thing I will give you

I had us at 6 or 7 wins but with that DL getting hurt im down to 5-6 wins
Posted by Prof
Member since Jun 2013
42652 posts
Posted on 8/9/19 at 1:58 pm to
quote:

Wow, Tennessee's schedule is an absolute joke.


IIRC, it's ranked at 7th in the nation but it is easier than what we normally play.
Posted by Prof
Member since Jun 2013
42652 posts
Posted on 8/9/19 at 2:01 pm to
quote:

I would be pretty surprised if Tennessee actually turns out to be our third toughest/closest game of the year.

Seems like these metrics are propping the Vols up a bit




I don't know what they're using, maybe they're looking at series history as a factor, but there's no reason for UF to be rated a harder game than UGA other than the fact that losing to UF is what we do no matter how good or bad either team is.
This post was edited on 8/9/19 at 2:02 pm
Posted by Tiger_Claw
Little Rock, AR
Member since Nov 2013
5136 posts
Posted on 8/9/19 at 2:03 pm to
There is no way we lose to Tennessee.
Page 1 2 3
Jump to page
first pageprev pagePage 3 of 3Next pagelast page
refresh

Back to top
logoFollow SECRant for SEC Football News
Follow us on Twitter and Facebook to get the latest updates on SEC Football and Recruiting.

FacebookTwitter