Prep Time 2012

[beej67]

Again, the main problem is that strength of opponent isn't taken into account at all.

Unless you're saying that GT has the toughest schedule in the entire ACC over the last four years, including bowls, that should average out over time.

 

[Legal Jacket]

That's actually what I always thought and posted about.

Ha, I give myself 0 points for reading comprehension then. I for some reason thought most/all of you guys were complaining that the offense is too easy to figure out, which is why we supposedly do terrible after the opp team has time to prep.

 

[Legal Jacket]

Unless you're saying that GT has the toughest schedule in the entire ACC over the last four years, including bowls, that should average out over time.

That's exactly what I'm saying (aside - You'd expect it to even out after a lengthy period of time, not four years). I think we have tended to play tougher teams when they have come off the bye. I recall playing uga coming off a bye twice, Miami, VT x 2, UNC. Those are good teams. Duke, for example, played FSU, Miami x 2, Maryland, and Virginia the last 4 years after a bye week. Virginia played Duke, Clemson, UNC, VMI, and (last year when they were actually not bad) us. Those are the worst teams in our division over the last 4 years.

In the other division, we didn't get Wake or Maryland coming off a bye. We haven't played BC since they dropped to the cellar.

Our bowl opponents have been pretty tough too. We had defending champ LSU in the peach bowl, A stout Iowa defense with a great D coordinator in the Orange bowl, an underrated 10+ win Air Force team that, in part, runs an offense very similar to ours, and a sun bowl game against a very capable Utah. So in the last four years we've played in three of the top four ACC bowl games. We only played in the bottom 4 bowls once over that time. If that doesn't mean we've had tougher bowl opponents compared to the rest of the ACC I think you need to recheck your math.

 

[GoldZ]

It certainly has not been proven vs the TO over many years as shown in prior threads. As for CPJ's tenure to date, data is limited and results are mixed. Check ugag & Clemson. Try Army vs Navy with CPJ. A lot, and I mean a LOT of teams in the 70's/80's could have had THREE weeks to prepare for OU and Nebraska and still would have suffered a slaughter ( their bowl records weren't too shabby back then). Too many other variables involved that may influence results.

So, I guess if VT, with all Spring, Summer, and Fall camp, doesn't beat us by 50, then the flawed theory really is flawed, huh?

 

[cyptomcat]

So, I guess if VT, with all Spring, Summer, and Fall camp, doesn't beat us by 50, then the flawed theory really is flawed, huh?

No, we have nearly 50 games of evidence. One game doesn't matter, it's statistically insignificant.

 

[ThisIsAtlanta]

So, I guess if VT, with all Spring, Summer, and Fall camp, doesn't beat us by 50, then the flawed theory really is flawed, huh?

Not really. The theory only pertains to unequal prep time. Unless we somehow play them on the first week of the season and then they get to play us in week 2, gaining them a paradoxical extra week of prep time, then it doesn't apply here.

 

[beej67]

I don't even know that it's "the triple option" itself that causes it. I just know it happens. As I mentioned before, I think other coaches are more embarrassed to lose to CPJ than they are to lose to anyone else, because a loss to CPJ reflects more poorly on them than on their players. So they practice harder, drive their kids harder, etc, for the weeks before they play us. Then their kids are beat and tired the next week, so they have a higher chance of losing their next game.

Not really. The theory only pertains to unequal prep time.

Not my theory. My theory only pertains to their prep time. I'm counting all four bowl losses in data I'd reference. Cyptom does the same.

VT will count in our 2012 "opponent has extra time" games.

 

[ThisIsAtlanta]

Not my theory. My theory only pertains to their prep time. I'm counting all four bowl losses in data I'd reference. Cyptom does the same.

VT will count in our 2012 "opponent has extra time" games.

I stand corrected.

 

[gtzulu]

I have done a t-test on the set of the ratio numbers, and there is only about 10% chance there is no statistical meaning to these numbers.

p = ~0.1 means "not considered statistically significant". It does not mean "only about 10% chance there is no statistical meaning to these numbers". So the yardage statistics do not show a statistically significant difference based on days of preparation.

I think these stats you have compiled are great, so that we can talk about this question in an objective way, I just think your interpretation of the statistics is off.

 

[cyptomcat]

p = ~0.1 means "not considered statistically significant". It does not mean "only about 10% chance there is no statistical meaning to these numbers". So the yardage statistics do not show a statistically significant difference based on days of preparation.

I think these stats you have compiled are great, so that we can talk about this question in an objective way, I just think your interpretation of the statistics is off.

http://en.wikipedia.org/wiki/P-value

p-value is an actual probability even if I have described it a bit vague.

"p-value is the probability of obtaining a test statistic at least as extreme as the one that was actually observed, assuming that the null hypothesis is true"

 

[gtzulu]

http://en.wikipedia.org/wiki/P-value

p-value is an actual probability even if I have described it a bit vague.

"p-value is the probability of obtaining a test statistic at least as extreme as the one that was actually observed, assuming that the null hypothesis is true"

Sure, but my point was p = ~0.1 is not generally considered statistically significant. p < 0.05 is the usual cutoff for declaring something statistically significant. So to me you have shown either:
(1) We don't have enough data yet to say that there is a statistically significant connection between yardage per play and preparation days.
(2) There is not a connection between preparation days and yardage per play.

0.1 is in the realm of "interesting", though, I'd say.

 

[GoldZ]

No, we have nearly 50 games of evidence. One game doesn't matter, it's statistically insignificant.

No cpto, when including only those games with the bad guys having extra prep, it's not so insignificant, huh? Plus there's the premise (dumb theory) that extra prep time dooms us---well then, a LOT of extra prep time should more than doom us---i.e. 50 point loss---ok?

 

[GoldZ]

Not really. The theory only pertains to unequal prep time. Unless we somehow play them on the first week of the season and then they get to play us in week 2, gaining them a paradoxical extra week of prep time, then it doesn't apply here.

We always prep for traditional Os, yes? It's the uniqueness of the TO in play here, yes? What did we do different the past 9 months vs what they did? Beeg diff, TIA.

 

[AE 87]

GT was leading 24 10 after the 3rd quarter of the Sun Bowl. Sims was out. Orwin was hobbled. Julian Burnett gets injured. Philthy was suspended. Shaq Mason gets hurt. We end up losing by 3 in overtime.

cyptomcat: a team we were favored to beat beat us because they had more than a week to prepare.

We scored 26 on VPI who averaged giving up 17.6 points per game. They averaged 27.9 points /game but scored 37 on us.

cyptomcat: they beat us because they had more than a week to prepare for our offense

c'mon guys.

 

[ThisIsAtlanta]

GT was leading 24 10 after the 3rd quarter of the Sun Bowl. Sims was out. Orwin was hobbled. Julian Burnett gets injured. Philthy was suspended. Shaq Mason gets hurt. We end up losing by 3 in overtime.

cyptomcat: a team we were favored to beat beat us because they had more than a week to prepare.

We scored 26 on VPI who averaged giving up 17.6 points per game. They averaged 27.9 points /game but scored 37 on us.

cyptomcat: they beat us because they had more than a week to prepare for our offense

c'mon guys.

The error that you're making here, that I also made, is in assuming they are talking about prep time specifically for our offense. Beej and cyp are talking about prep time for the game, and aren't getting any more granular in their statements than that.

 

[beej67]

Sure, but my point was p = ~0.1 is not generally considered statistically significant. p < 0.05 is the usual cutoff for declaring something statistically significant. So to me you have shown either:
(1) We don't have enough data yet to say that there is a statistically significant connection between yardage per play and preparation days.
(2) There is not a connection between preparation days and yardage per play.

0.1 is in the realm of "interesting", though, I'd say.

If that's your cut off, then every single football statistic is meaningless.

Four years of win/loss record is enough to get anyone in the business fired. Therefore it's statistically significant for football.

 

[ThisIsAtlanta]

We always prep for traditional Os, yes? It's the uniqueness of the TO in play here, yes? What did we do different the past 9 months vs what they did? Beeg diff, TIA.

It would be remiss to say that the only way in which we differ from more traditional teams is our offense. Our defense is also not common at the college level, as well as a few other uniquenesses that we as a team currently possess.

 

[ThisIsAtlanta]

If that's your cut off, then every single football statistic is meaningless.

Four years of win/loss record is enough to get anyone in the business fired. Therefore it's statistically significant for football.

It's much less of a problem in the NFL, but it is true, you have to live on the edge with CFB stats. One of the things that makes the game so great, anything can happen.

 

[beej67]

The error that you're making here, that I also made, is in assuming they are talking about prep time specifically for our offense. Beej and cyp are talking about prep time for the game, and aren't getting any more granular in their statements than that.

^this

In fact, if we can show that prep time definitely helps other teams beat us more than it helps most other teams beat other teams, and we can also simultaneously show that prep time doesn't help teams contain our offense, then it tells us other, similarly interesting things.

As I say again, for the third time, my new operating theory is that coaches just take prepping vs GT more seriously than other opponents, because they don't want their own local media to paint them as being dumber than Paul Johnson. Comments leading into the UVA game cement that, in my mind. Go dig those up.

 

[stinger78]

GT was leading 24 10 after the 3rd quarter of the Sun Bowl. Sims was out. Orwin was hobbled. Julian Burnett gets injured. Philthy was suspended. Shaq Mason gets hurt. We end up losing by 3 in overtime.

cyptomcat: a team we were favored to beat beat us because they had more than a week to prepare.

We scored 26 on VPI who averaged giving up 17.6 points per game. They averaged 27.9 points /game but scored 37 on us.

cyptomcat: they beat us because they had more than a week to prepare for our offense

c'mon guys.

+1

Extra prep helps every team in some way.