Notre Dame has 75% chance of beating GT...per ESPN

[gt2690b]

CPJ wants to kick 100% of VanGorder's @ss. That's all I need to know.

CPJ wanted to kick 100% of cutcliffe's @ss last year.. hopefully he will this year

 

[floridajacket]

Their FPI is highly dependent on recruiting rankings. Considering their FPI rankings have us at 23 and ND at 7, the 75% figure is not a surprise. Basing anything on recruit rankings is retarded.

FPI also downgrades GT for getting so many turnovers. Turnovers generally don't have the same dependability as yards.

FPI and other preseason computer models have to use recruiting rankings because there's nothing better to use.

Computer models have to target the best R^2, which means trying to predict the best results for the average team. They can't do a special line of programming which says "do something different for GT and TCU."

538 should at least use rankbyoffers.com, which at least has an open algorithm based on offers. Some day I may look into the algorithm to see if it can address the issues which consistently rank TCU poorly. For example, the algorithm can take into account the timing of offers and commitments. Some GT commits have been too underrated because they committed early and CPJ doesn't allow them to still shop around after commitment.

 

[33jacket]

priceless....SECPN

 

[WracerX]

Football schools don't lose to basketball schools.

 

[beej67]

Oh that's true. I'd forgotten they're a basketball school now. Nice.

 

[ScionOfSouthland]

FPI looks like it is basically a 10,000 monte carlo simulation that gets updated every week to incorporate a tighter sensitivity analysis. Teams then get placed where they fall into the distribution to create a likely to beat other opponents on their schedule.


Right now the only data it has is last year's data since there is no week-over-week data to pull from as well as Strength of Schedule, which it looks like is both an input and an output in the model. I gander that it also includes overall recruiting rankings as a weighted element. So yes, it makes sense that ND would beat GT 75% of the time (on what is likely a 90% confidence interval) using last year's stats and the strength of schedule. The real question we have to face is whether our defensive performance was an aberration and we are an interloper in being near last in the nation in defensive statistics, or whether we have improved (dramatically, or at least markedly) .


Our "power index" right now is in the high 20s while Notre Dame is in the teens. That will likely change as we get closer to the game - incorporating ND's SOS vs. our and margins of victory/loss in week 1 and 2 - in order to get to a truer number.


What I perceive to be a gap in the results is BVG's ability to stop the option and his experience against option teams as well as the unknown element at QB for Notre Dame. If you had posed the same FPI of us vs. MSU, I'm sure their inflated run-stop stats against pro-set SEC teams would skew their perceived ability to stop the run against CPJ in their favor.

Consider this: Alabama is #7 on the FPI rankings, 1 slot behind Texas A&M and 3 slots behind #4 LSU. Which of those teams is likely to be the better power this season? My bet is Alabama. Likewise, Arkansas and Tennessee are ahead of Auburn, FSU, and Clemson which would seem to me to be a toss up on a neutral field. Auburn, as you know, is widely predicted to win the SEC this year (an assessment I disagree with).

 

[GSUGTATL]

Their FPI is highly dependent on recruiting rankings. Considering their FPI rankings have us at 23 and ND at 7, the 75% figure is not a surprise. Basing anything on recruit rankings is retarded.

Basing anything on FPI is retarded.

 

[TechSBP]

What was our odds of winning last time we went into Sound Bend and ROFL-stomped the öööö out of the Irish and human garbage disposal they had coaching them?

Charlie Weiss paradox---Can God create a man so fat that God can't create a coffin big enough to hold him?

 

[gthog61]

Charlie Weiss paradox---Can God create a man so fat that God can't create a coffin big enough to hold him?

I dunno but if he chooses cremation if will be a fire hazard, lot of oils and fats and greases

 

[CiraldoForever]

We need more of this kind of negative publicity. We're always better when people write us off. I hope every ND player reads it and believes it, and that every Tech player reads it and gets back down to earth and is totally motivated.

If ND had gone 11-3 last year, beaten Va. Tech, Miami, Clemson and Georgia and clobbered Miss State in the Orange Bowl, they'd probably be pre-season top four.

The truth is that right now nobody knows anything about who's going to do what this coming season. I'm nervous about it. But you take the good with the bad. We might go 0 - 12, or we might go 15 - 0.

All I know is that I'll be a Tech fan when the season starts, and I'll be a Tech fan when the season ends. To Hell With Georgia!

 

[southendzoneBEE]

33-3

 

[gtphd]

Their FPI is highly dependent on recruiting rankings. Considering their FPI rankings have us at 23 and ND at 7, the 75% figure is not a surprise. Basing anything on recruit rankings is retarded.

GT sat between #40 and #50 in FPI for virtually the entire season last year

 

[wesleyd21]

We need more of this kind of negative publicity. We're always better when people write us off. I hope every ND player reads it and believes it, and that every Tech player reads it and gets back down to earth and is totally motivated.

If ND had gone 11-3 last year, beaten Va. Tech, Miami, Clemson and Georgia and clobbered Miss State in the Orange Bowl, they'd probably be pre-season top four.

The truth is that right now nobody knows anything about who's going to do what this coming season. I'm nervous about it. But you take the good with the bad. We might go 0 - 12, or we might go 15 - 0.

All I know is that I'll be a Tech fan when the season starts, and I'll be a Tech fan when the season ends. To Hell With Georgia!

You seem like a nice person. Are you sure this board's for you?