Separate names with a comma.
Discussion in 'RayGator's Swamp Gas' started by g8trchamp, Oct 10, 2013.
this may be the worst and most inaccurate poll ever created.
Alabama at 37? Ok... Lol
This. Lock this thread and throw away the key. Terrible poll.
We're #2 and LSU is #12?
This poll has potential.
It's based on adjusted yards per play.
Also I see that there are 2000+ asian girls online -
Pretty sure that just shy of a few million.
and Bama at 37 -- someone needs to reset their algorithm
Even Memphis made it into the 50+ cat and they won't win three games.
Good try , but we won't bite this bs
It's based on advance statistics, not wins and losses. It's not something you'd use when you're considering bowl games, but its a decent measure of teams talent and how well they've played in comparison with their record.
its the most ridiculous thing i have ever seen in regards to polls. when you have kentucky in a top 25, i don't care what formula or advance stats your using; you should take this poll and throw it in the trash. then take the trash out. take it to the dumpster. set the dumpster on fire. put the fire out. call the garbage company to come and pick up the dumpster. the garbage company proceeds to drop the the dumpster in the atlantic ocean where it falls to the bottom never to be seen again.
Bama at #37? Ohh, well.......if they finish 37th or higher in any poll, that'll be enough for them to claim another national championship.
Wisconsin at #6 with 2 losses? Good thing they got the big wins over UMass, TN Tech, & Purdue to show what they're really made of.
It's way too early
It's a joke..
The poll makes no sense
All computer rankings need more data than they currently have.
This one is not considering turn overs, RZ performance and special teams play, so a team like Bama is undervalued, while we are overvalued. It also does not adjust for SOS, so teams playing weak schedules are overvalued, compared to teams playing tough schedules.
At a minimum, they need to add factors for SOS, TO margin, and points per RZ opportunity and figure how to weight them. Perhaps net yards per play on ST plays, too. Or if they want a simple metric with minimal factors, just go with points per possession instead of yards per play.
But this ranking does give you a pretty good look at how well teams control the flow of the game on offense and defense. FWIW.
Guys, he posts the data but its not meant to be evaluated until like 8 games in. It's going to have flaws like any computer poll regardless. It gives an idea of rankings based on fairly good metrics though. Shouldn't be used for comparing teams just yet.
The website claims to use "adjusted" yards per play, which leads me to believe that they are taking the yards/play in relation to the average yards/play generated and yielded by the opponent - and explains why the ratings come out near 1. This should act as it's own measure of SOS.
The results still seem a bit screwy, but it does solve a problem that you get with most score based models: It recognizes that Florida was better than Miami. (However, so does Sagarin, so even his score model picked this up using the rest of the games)
I think the drawback of using points/possession (which I also endorse) is that in football there are so few possessions. We've had a game with 7 possessions this year, I think, which probably lends itself to some randomness that yards/play won't display. However, you've correctly identified the drawback of yards/play, which is explosive drives that don't lead to points.
I'll be interested to see how this model does over the course of the year.
By the way, this isn't a "poll". I recognize that the OP used that term, but this is a computer model.
Their final poll from last year had FSU at 2 and Utah State at 5 so...
So this poll doesn't use SoS in the formula? That's all I need to know... Worthless.