ESPN FPI Says We Go 4-8 or 5-7

[ShotgunDawg]

Remember when FPI had us winning 10 a couple of years ago? FPI is worthless.

There will be growing pains because we turned over the coaching staff and are running completely new systems on both sides of the ball. But we will beat NC State, Mizzou, Arkansas, Ole Miss, and Kentucky because we are just better than they are.

That Kentucky game does worry me, but we've got the right coach to do it & our chances of winning that game are damn sure greater than 20%. Probably 40-45%

[ShotgunDawg]

They are for coaching tenure, but not necessarily in returning production -which is what the pre-season FPI is weighted toward.
Ole Miss returns all of its passing production, return 4 of the 5 top rushers, 4 of their 5 top WR, 4 of their 5 top tacklers, and the bulk of their INTS.

You can argue that those players aren't as talented as what we have returning. You can argue that they are potentially not as productive as what we have.
But, they return a very high percentage of their production and that is a huge factor in the pre-season FPI.

So we get no credit for Costello & Leach? If FPI isn't considering those two factors then it's simply inaccurate & worthless

[TrapGame]

They way college football has been effected by COVID 19 all bets are off on any of this. This is going to be one of those weird and wonderful years in college football. The quarantines and "new normal" will produce some strange outcomes this Fall.

[msugolf]

The early FPI is heavily weighted to returning production. We lost our top producing passer, our #2 & #3 rushers; our #2, 3, & 4 WR; Tacklers #2, 3, 4, 7 & 9 plus Willie Gay; and the players who produced 7 of our 11 INTS.
Off a 6-7 team.

According to Edwards in his OOB interview he said they DID actually tweak the formula to not put as much emphasis on returning production and instead replaced it with returning talent. I guess they go by their own recruiting rankings to determine that, which is another topic. Also he said the formula isn't high on Costello because it only factors in the previous year production. And another thing is that it doesn't factor in overall coaching effect like Bartoo's does.

He ended up contradicting himself a bunch with the FPI results because even according to their recruiting rankings we've outrecruited Arky, Mizz, Kentucky, NCState the last 5 years. And said it DOES factor in if you have a first year coach (typically a down year) and said that it hurts State but didn't mention that Ark, Mizz, OM all have first year coaches too.

Basically its just a really bad formula that I have a feeling will get blown apart this year.

[ShotgunDawg]

They way college football has been effected by COVID 19 all bets are off on any of this. This is going to be one of those weird and wonderful years in college football. The quarantines and "new normal" will produce some strange outcomes this Fall.

I agree with this very much. I think it's going to be a really wild year

[BrunswickDawg]

So we get no credit for Costello & Leach? If FPI isn't considering those two factors then it's simply inaccurate & worthless

It is completely worthless this time of year. But, I don't make the FPI rules, ESPN does.

[ShotgunDawg]

Also he said the formula isn't high on Costello because it only factors in the previous year production. And another thing is that it doesn't factor in overall coaching effect like Bartoo's does.

Then it's worthless.

Computers are only as good as the information that is put into them &, if they're incapable of putting in quality information, then the FPI is worthless as a source

[BiscuitEater]

The early FPI is heavily weighted to returning production. We lost our top producing passer, our #2 & #3 rushers; our #2, 3, & 4 WR; Tacklers #2, 3, 4, 7 & 9 plus Willie Gay; and the players who produced 7 of our 11 INTS.
Off a 6-7 team.

Actually, we lost #3 and #4 rushers Gibson and QB Stevens. #1, #2, & #5 return + a 4* with great potential. Rushing has MUCH greater potential than in '19.

Shrader was only 4.0 less in efficiency & <3% less than Stevens .. with Will Rogers and Costello easily > loss of Stevens.

Our 'leading' WR only caught 30, our #2 caught 30. Just see way more potential with Leech O than we got in '19.

Just don't see that much of a drop off in D. Willie Gay will be missed but only played like 2.3 full games in '19.

[Commercecomet24]

On the plus side at least when we lose with Leach the press conferences won't be boring lol.

[confucius say]

So we get no credit for Costello & Leach? If FPI isn't considering those two factors then it's simply inaccurate & worthless

Correct. You actually get dinged. Edwards said there is no scenario where you get a bump from a coaching change, you always get dinged regardless of who is replacing whom. You also get dinged if your qb is not a returning starter. That is why the fpi was so off on oh state last year-they got dinged for not have a returning qb even though fields was really good.

[WPS]

They are for coaching tenure, but not necessarily in returning production -which is what the pre-season FPI is weighted toward.
Ole Miss returns all of its passing production, return 4 of the 5 top rushers, 4 of their 5 top WR, 4 of their 5 top tacklers, and the bulk of their INTS.

You can argue that those players aren't as talented as what we have returning. You can argue that they are potentially not as productive as what we have.
But, they return a very high percentage of their production and that is a huge factor in the pre-season FPI.

Yeah I would imagine the formula helps Arkansas too since 4/5 OL are back, every WR is back, and Rakeem Boyd is back and should be the SEC's 3rd best RB.

It's also probably taking into account Feleipe Franks' good numbers through 3 games last year before he got hurt, which skews things quite a bit. Costello only played in 5 games so he had a shortened season as well but his completion % and QBR are worse since he played tougher competition overall (Franks went 25/27 against UT-Martin which skyrocketed his completion percentage).

Not real sure how Arkansas didn't get a big hit for returning production on defense since we lost possibly our best player from each level of the D, but I guess when you're as bad as we were on defense then losses don't matter as much lol

[BrunswickDawg]

Actually, we lost #3 and #4 rushers Gibson and QB Stevens. #1, #2, & #5 return + a 4* with great potential. Rushing has MUCH greater potential than in '19.

Shrader was only 4.0 less in efficiency & <3% less than Stevens .. with Will Rogers and Costello easily > loss of Stevens.

Our 'leading' WR only caught 30, our #2 caught 30. Just see way more potential with Leech O than we got in '19.

Just don't see that much of a drop off in D. Willie Gay will be missed but only played like 2.3 full games in '19.

Yeah - I was off on a couple, but it's still indicative of the amount of production we are losing.

And I agree with your points - but it is also not what they are using to formulate FPI. I'm not making an argument that we are going to be 5-7 like the FPI claims, only that the way FPI looks at us on paper is essentially only calculating one factor right now - returning production.

[Todd4State]

We have too many things that skew us.

Tutor Gate limited the production of guys like Marcus Murphy.

Extremely poor S&C making our players look not as talented.

Completely new systems on both sides of the ball.

Moorhead's personnel usage limited production from the total roster making guys like Javonta Payton look worse than they actually probably are.

[ShotgunDawg]

We have too many things that skew us.

Tutor Gate limited the production of guys like Marcus Murphy.

Extremely poor S&C making our players look not as talented.

Completely new systems on both sides of the ball.

Moorhead's personnel usage limited production from the total roster making guys like Javonta Payton look worse than they actually probably are.

Dang. I forgot about Tutor gate effecting this but it absolutely does.

I guess I understand how FPI got us to 4-5 wins, but again, it's worthless because reality shows that their are significant reasons to bet the over.

[FISHDAWG]

Dang. I forgot about Tutor gate effecting this but it absolutely does.

I guess I understand how FPI got us to 4-5 wins, but again, it's worthless because reality shows that their are significant reasons to bet the over.

If Bama had all new starters they wouldn't predict them this low ... there has to be a human element in all of this somewhere - the thing about statistics is there is always an exception somewhere and these guys are just relying on math because there's no way they could fully understand over 120 teams ...... like you said - they have ignored the reality factor here

[dawgday166]

According to Edwards in his OOB interview he said they DID actually tweak the formula to not put as much emphasis on returning production and instead replaced it with returning talent. I guess they go by their own recruiting rankings to determine that, which is another topic. Also he said the formula isn't high on Costello because it only factors in the previous year production. And another thing is that it doesn't factor in overall coaching effect like Bartoo's does.

He ended up contradicting himself a bunch with the FPI results because even according to their recruiting rankings we've outrecruited Arky, Mizz, Kentucky, NCState the last 5 years. And said it DOES factor in if you have a first year coach (typically a down year) and said that it hurts State but didn't mention that Ark, Mizz, OM all have first year coaches too.

Basically its just a really bad formula that I have a feeling will get blown apart this year.

Yea it is. Will say tho that with Covid and all the changes ... I have no idea what we'll put on the field this year. But I also think OM is in same boat and not sure how they're ahead of us. But ... they were always ahead of us in old FPI too.

ETA: And according to ESPN recruiting rankings, it seems like there are several teams that should have more "returning talent".

[Liverpooldawg]

Assuming we actually play.....that's about right.

[R2Dawg]

That Kentucky game does worry me, but we've got the right coach to do it & our chances of winning that game are damn sure greater than 20%. Probably 40-45%

UK is tough about half the time but historically we own UK. No way that is 20% but more like 50% at worst.

Ark being that low, joke. We play them at home and own Ark too not to mention that have not been good for a decade.

OM that low off too. About like UK, I'd say 50%. On road, rivalry, they are probably due. How many times in recent history has anyone won egg bowl three years in a row? Since 1980?

[Commercecomet24]

UK is tough about half the time but historically we own UK. No way that is 20% but more like 50% at worst.

Ark being that low, joke. We play them at home and own Ark too not to mention that have not been good for a decade.

OM that low off too. About like UK, I'd say 50%. On road, rivalry, they are probably due. How many times in recent history has anyone won egg bowl three years in a row? Since 1980?

We won the Egg Bowl 2009,2010 and 2011, but it hasn't happened very often, true.

[Pipedream]

Dang. I forgot about Tutor gate effecting this but it absolutely does.

I guess I understand how FPI got us to 4-5 wins, but again, it's worthless because reality shows that their are significant reasons to bet the over.

If the over was 4.5, it would be getting hammered, but it's 6.5 and I wouldn't touch it. I have no clue what we will look like.