Sagarin - #24

Donate To GopherHole

The Gopher defense seems to be good enough to hold serve against these teams that have average to poor defenses and the offense is good enough to score at least a moderate amount against pretty much anyone. Therefore, the Iowa and Northwestern games look more winnable now and if Michigan State was on the schedule I would be very confident, even if it was a road game.
 

WinLoseOrTy said:
First time Gophers crack the top 25 in Sagarin this season. But somehow has them behind a 4-4 Michigan State team. Wisconsin is still at #7. Might be time to revise the model...

http://sagarin.com/sports/cfsend.htm
Click to expand...

The model really hated our performance early in the season, but we've slowly been rising as we've added more blowouts recently. Just gotta keep executing and the model will eventually catch up.
 

Not sure what the model doesn't like about 20+ point wins.
 

The model is based on scores no matter how high or how low.
The model also takes into account SOS. Even though WI has two losses they won two of three games against ranked teams and the one bad loss was by one point. MN had a number of games that MN won by small numbers against unranked teams.
MN has yet to play a ranked team. It is not anyone's fault it is just how the computer spit out the schedule.
Beat PSU and the ranking will improve. It is that simple.
Sagarin feeds the numbers into the computer and the his software spits out the numbers. Sagarin is used by the CFP committee and by the betting companies.
 


Right now giving three points to MN for home field advantage the spread would be PSU -11.
 

Sagarin has prognosticated the gophers season better than me 1 of the last 2 seasons
 


Some guy said:
Sagarin has prognosticated the gophers season better than me 1 of the last 2 seasons
Click to expand...

That’s really not the point of the model. If you want to “compete” against the model, you need to compare your prognostications for all 117 FBS teams against that of the model’s.

I should also point out that 1 of 2 is statistically insignificant and no conclusions can be drawn from that.
 




BeerFueledFF said:
I'm guessing the line will come down closer to the 7-9 range when we get it next week.
Click to expand...

If it moves from 11 to 7-9 it is not because Sagarin has changed his data it is because the betting sites want more $$ on PSU so bets on MN and PSU are as close to 50% as possible.
Why is all of this so hard to understand?
 

Plato said:
If it moves from 11 to 7-9 it is not because Sagarin has changed his data it is because the betting sites want more $$ on PSU so bets on MN and PSU are as close to 50% as possible.
Why is all of this so hard to understand?
Click to expand...

This!!

Many people think that Vegas lines are predictions made by the “bookies.” When that not what they are at all. They are the collective opinion of all the gamblers, I.e. the ones willing to risk money on their opinion.
 

WinLoseOrTy said:
First time Gophers crack the top 25 in Sagarin this season. But somehow has them behind a 4-4 Michigan State team. Wisconsin is still at #7. Might be time to revise the model...

http://sagarin.com/sports/cfsend.htm
Click to expand...

Massey with over 35+ poles has the Gophers at #17 which includes the Sagarin., but Ive seen other pole, maybe CBS, has you ranked at 13.
 



Plato said:
The model is based on scores no matter how high or how low.
The model also takes into account SOS. Even though WI has two losses they won two of three games against ranked teams and the one bad loss was by one point. MN had a number of games that MN won by small numbers against unranked teams.
MN has yet to play a ranked team. It is not anyone's fault it is just how the computer spit out the schedule.
Beat PSU and the ranking will improve. It is that simple.
Sagarin feeds the numbers into the computer and the his software spits out the numbers. Sagarin is used by the CFP committee and by the betting companies.
Click to expand...

I’m not sure anyone really knows Sagarin’s formula. I always he thought it was a high minded version of the transitive property in action where margin of victory, opponent is a key factor. If anyone really understands it please post. This is why I tend (for football) to pit more stock into systems like SP+ which at least in 2019 seems to have been more accurate. We’ll see. It would be interesting to see a retrospective on projected margins for all the games this year between the Sagarin and SP+.
 

DEGopher said:
That’s really not the point of the model. If you want to “compete” against the model, you need to compare your prognostications for all 117 FBS teams against that of the model’s.

I should also point out that 1 of 2 is statistically insignificant and no conclusions can be drawn from that.
Click to expand...

Nothing sagarin does before the fact is statistically significant.

So thanks for pointing out the worthlessness of the metric before the fact



After the fact at the end of the season it gives some great analysis like the teams with more wins against harder schedules probably being better than the teams with fewer wins.

And then it gives some subjective information that is unprovable like Texas A&M at 7-5 is better than 10-2 AAC team.
 

Plato said:
If it moves from 11 to 7-9 it is not because Sagarin has changed his data it is because the betting sites want more $$ on PSU so bets on MN and PSU are as close to 50% as possible.
Why is all of this so hard to understand?
Click to expand...

I was not trying to say anything about Sangarin changing his data, I was attempting to state where I expected Vegas to set the line.

Per another post ESPN has it at 6.5 right now, which is close to what I expected.
 

Sagarin stated one of the BCS or ESPN did make him change his #1 team back in 2013 to more align with the mob. By mob I mean the rube voters and fans, not the actual criminal organizations. I’m sure it can be googled.
 


fmlizard said:
MSU lost their 3 October games 100-17
Minnesota won their 4 October games 168-41

Man, Sagarin must weight SOS by 100000%
Click to expand...

Yup.
And SOS is determined in large part by opponents SOS



In my opinion the results of games should be weighted number 1

There are 130 teams playing 12 games each...we are going to say 2 loss teams are better than unbeaten ones because the 2 loss team played 7 7 wins teams instead of 7 5-6 win teams.
 

Sagarin does not have an axe to grind and his rankings have no affect on the outcome of any game.
I have a hard time understanding why some posters feel the world is out to disrespect MN.
No ranking means anything until the last one at the very end of the season.
 

Plato said:
Sagarin does not have an axe to grind and his rankings have no affect on the outcome of any game.
I have a hard time understanding why some posters feel the world is out to disrespect MN.
No ranking means anything until the last one at the very end of the season.
Click to expand...

All models are wrong. Some, however, are useful. Sagarin, like other formula rankings uses data to mathematically rank the teams. No one things it was designed to be biased against Minnesota. But just because it is neutral, that doesn't mean it is good. The fact that a team losing as much as MSU is ranked as high as it is is leading us to question whether it is a good model.
 

bleedsmaroonandgold said:
All models are wrong. Some, however, are useful. Sagarin, like other formula rankings uses data to mathematically rank the teams. No one things it was designed to be biased against Minnesota. But just because it is neutral, that doesn't mean it is good. The fact that a team losing as much as MSU is ranked as high as it is is leading us to question whether it is a good model.
Click to expand...

MSU is the biggest outlier, but there are a bunch of 3 win teams ahead of us as well.

Hopefully Jeff takes a look at these results and does some tweaking over the off-season, as a team who has lost 3-4 out of 8 should not be considered a top 20ish team by your model.
 

WinLoseOrTy said:
First time Gophers crack the top 25 in Sagarin this season. But somehow has them behind a 4-4 Michigan State team. Wisconsin is still at #7. Might be time to revise the model...

http://sagarin.com/sports/cfsend.htm
Click to expand...

BUT, Sagarin has more than one model. You're only looking at the ranking on the left side. Sagarin has a range of values and rankings.

Sagarin also has a rating based on "Recent" performance. I don't know how far back "recent" goes but the Gophers are ranked #15 on that one which is very close to their ranking in national polls. That's one of the larger discrepancies between "rating" and "recent" where the recent is significantly higher than the rating. Conversely, Michigan State's discrepancy is in the opposite direction: their "rating" is ranked #23 while their "recent" is ranked #32. The same is true for Texas.
 

cjbfbp said:
BUT, Sagarin has more than one model. You're only looking at the ranking on the left side. Sagarin has a range of values and rankings.

Sagarin also has a rating based on "Recent" performance. I don't know how far back "recent" goes but the Gophers are ranked #15 on that one which is very close to their ranking in national polls. That's one of the larger discrepancies between "rating" and "recent" where the recent is significantly higher than the rating. Conversely, Michigan State's discrepancy is in the opposite direction: their "rating" is ranked #23 while their "recent" is ranked #32. The same is true for Texas.
Click to expand...

Do not confuse the angry ones with the facts. It is far more satisfying believing the world is out to get your team.
 

You must log in or register to reply here.



Top Bottom