Coaching Change not cheap

[COJacket]

Agree. If the only change was CPJ switched the offense to what he had in Hawaii, then we’d have an experiment. Or if the only change was the continued increased funding for recruiting that was already underway, then we’d have an experiment. With the offense change, the complete turnover in staff, a larger staff, additional funds being made available, and who knows what else, we can’t draw any real conclusions. Clearly, there were players who declined to come to GT because of the TO, but we also got some players who came because the TO gave them the opportunity they wanted (JT being the best example).

My post was a bit tongue in cheek. Perfect world it would be great to know. But in reality we will never know. I agree w your post here a 110%. Let’s just remember all the factors before saying any improvement was only because we dropped the TO.

 

[AE 87]

I'm guessing that there was at least 50% increase in coaching staff salary. Some of that probably comes from the end of Hewitt/Gregory coming soon.

 

[takethepoints]

TStan gets it and he's using his stick on those nerds on the hill finally.

This is a fantasy of the first water with, so near as I can tell, nothing to back it up.

My guess is that Coach will get the same exemption leash that Paul did and, if the athletes do well enough, he'll keep it. As to the rest, they'll leave it up to TStan to scrape up the money and run clean programs. If he does, all will be well. If he doesn't, the "nerds on the hill" will be heard from big time.

 

[takethepoints]

This increase in money applied to more recruiting staff and recruiting travel actually adds a variable to the TO recruiting experiment So when the new staff moves our recruiting from 40-60 level to 20-40 level how much is weighted to the TO effect, vs the new staff’s energy vs the increase in budget and resources. Can we do a regression analysis on these variables to determine the proper weighting?

Actually, you could do this. It would be an (ahem!) "difference of differences" design. What you would be testing is whether a change in offensive system led to better recruiting, controlling the relationship for "staff energy" and increases in resources. I say controlling here because a) we were planning an increase in resources anyway and b) staff energy would be a doubtful factor since I'm pretty sure Paul's staff gave their all in recruiting. Staff energy would be hard to measure, except, perhaps, by mentions in the press or something like that. Controlling for resources would be much easier.

You would have to wait a year or two to get trends; you could do it after next year, but trends are more reliable. Then you pick a couple of dependent variables (average stars and recruiting rankings would do it) and look to see if the change in offensive systems led to a significant change in the coefficients for years before and after the new system was adopted. You'd be controlling for resource changes and perceived "staff energy". That would give you a clear shot at seeing whether changing from the TO to a shotgun spread made any difference in recruiting.

As was pointed out above, chances are all this work - and don't look at me to do it - wouldn't change people's opinions, especially since the change we can expect will, imho, be too small to register as significant, given the small size of the datasets involved and the likely small shifts in average stars and rankings. (I think Coach will be able to get us on a par with VT in recruiting short term, but we'll see.) But it could be done.

 

[COJacket]

Actually, you could do this. It would be an (ahem!) "difference of differences" design. What you would be testing is whether a change in offensive system led to better recruiting, controlling the relationship for "staff energy" and increases in resources. I say controlling here because a) we were planning an increase in resources anyway and b) staff energy would be a doubtful factor since I'm pretty sure Paul's staff gave their all in recruiting. Staff energy would be hard to measure, except, perhaps, by mentions in the press or something like that. Controlling for resources would be much easier.

You would have to wait a year or two to get trends; you could do it after next year, but trends are more reliable. Then you pick a couple of dependent variables (average stars and recruiting rankings would do it) and look to see if the change in offensive systems led to a significant change in the coefficients for years before and after the new system was adopted. You'd be controlling for resource changes and perceived "staff energy". That would give you a clear shot at seeing whether changing from the TO to a shotgun spread made any difference in recruiting.

As was pointed out above, chances are all this work - and don't look at me to do it - wouldn't change people's opinions, especially since the change we can expect will, imho, be too small to register as significant, given the small size of the datasets involved and the likely small shifts in average stars and rankings. (I think Coach will be able to get us on a par with VT in recruiting short term, but we'll see.) But it could be done.

I love this takethepoints! Only on Gtswarm would we have someone respond like this. In general I have loved your posts over the years! Thanks for all your great posts.

 

[Jim Prather]

Actually, you could do this. It would be an (ahem!) "difference of differences" design. What you would be testing is whether a change in offensive system led to better recruiting, controlling the relationship for "staff energy" and increases in resources. I say controlling here because a) we were planning an increase in resources anyway and b) staff energy would be a doubtful factor since I'm pretty sure Paul's staff gave their all in recruiting. Staff energy would be hard to measure, except, perhaps, by mentions in the press or something like that. Controlling for resources would be much easier.

You would have to wait a year or two to get trends; you could do it after next year, but trends are more reliable. Then you pick a couple of dependent variables (average stars and recruiting rankings would do it) and look to see if the change in offensive systems led to a significant change in the coefficients for years before and after the new system was adopted. You'd be controlling for resource changes and perceived "staff energy". That would give you a clear shot at seeing whether changing from the TO to a shotgun spread made any difference in recruiting.

As was pointed out above, chances are all this work - and don't look at me to do it - wouldn't change people's opinions, especially since the change we can expect will, imho, be too small to register as significant, given the small size of the datasets involved and the likely small shifts in average stars and rankings. (I think Coach will be able to get us on a par with VT in recruiting short term, but we'll see.) But it could be done.

Possible, but how do you quantify “staff energy”?

 

[bravejason]

Possible, but how do you quantify “staff energy”?

Easy! It’s the Difference in the average daily calories consumed at Waffle House between the prior staff and the current staff.

 

[gtpi]

Opens the door to massive recruiting violations......that is why
https://www.ncaa.org/enforcement/role-boosters

big boys have 30 +/- grunts on their staffs. volunteers. that gather data . write letter, etc. delegated by a paid staffer. no recruit.contact. i dont see a problem?

 

[takethepoints]

I love this takethepoints! Only on Gtswarm would we have someone respond like this. In general I have loved your posts over the years! Thanks for all your great posts.

Now you're embarrassing me! I'm glad you like my posts and I've always liked yours too. At least you call for using data to solve some of the disagreements here. Even if (ahem, again) you can't find anybody to take you up on actually doing the work.

 

[takethepoints]

Easy! It’s the Difference in the average daily calories consumed at Waffle House between the prior staff and the current staff.

Now that's funny!

 

[takethepoints]

Possible, but how do you quantify “staff energy”?

Like you say, it would be hard, maybe even impossible, to actually quantify staff energy. It would be less difficult, though still no walk in the park, to quantify the perception of staff energy. Text analysis of the local papers or of Twitter feeds might work. This, btw, is one of the reasons why I wouldn't touch an actual project to get the analysis done with a fork. Way too much work.

 

[Animal02]

big boys have 30 +/- grunts on their staffs. volunteers. that gather data . write letter, etc. delegated by a paid staffer. no recruit.contact. i dont see a problem?

That is far different than local groups/ fans / satellite groups doing it. It is a recipe for prohibitive contact etc.

 

[coldbeer]

I'm guessing that there was at least 50% increase in coaching staff salary. Some of that probably comes from the end of Hewitt/Gregory coming soon.

This.... when you are paying 3 hoops coaches and 2 fall coaches, it gets expensive. Especially for a middle of the road program like GT (financially).

Sent from my SM-N950U using Tapatalk

 

[bravejason]

I'm guessing that there was at least 50% increase in coaching staff salary. Some of that probably comes from the end of Hewitt/Gregory coming soon.

Unless something changed, Gregory rolled off the books last March and Hewitt comes off this March. Between the two of them the annual payment was about $1M. It’s good to have them off the books, but it’s hardly a game changer.

 

[vamosjackets]

Unless something changed, Gregory rolled off the books last March and Hewitt comes off this March. Between the two of them the annual payment was about $1M. It’s good to have them off the books, but it’s hardly a game changer.

$1M/year could hire 10-20 new staff people. That seems like it could move the needle. Or, it could've bought CPJ the elite DC he always needed, which would've had an immeasurable impact (like winning one of those fancy NC's or something).

 

[bravejason]

$1M/year could hire 10-20 new staff people. That seems like it could move the needle. Or, it could've bought CPJ the elite DC he always needed, which would've had an immeasurable impact (like winning one of those fancy NC's or something).

When you are expecting to run a deficit of $9M in 2019, does $1M really change anything? The Athletics Initiative is $125M over three years, a million a year is about 2.4% of that.

Don’t misunderstand me, we’re better off with the money than without it, but I just don’t think that $1M/yr is the difference between GT FB today and the GT FB that will meet fans expectations. Maybe it is, but I just don’t see how the GT FB program could be that close to it from a financial perspective given what I see from other highly successful programs.

 

[vamosjackets]

When you are expecting to run a deficit of $9M in 2019, does $1M really change anything? The Athletics Initiative is $125M over three years, a million a year is about 2.4% of that.

Don’t misunderstand me, we’re better off with the money than without it, but I just don’t think that $1M/yr is the difference between GT FB today and the GT FB that will meet fans expectations. Maybe it is, but I just don’t see how the GT FB program could be that close to it from a financial perspective given what I see from other highly successful programs.

It depends on how the $1M is spent. $1M could be spent in a way that makes very little impact or it could be spent in a way that changes everything. For example, Clemson spending the money to get Venables took them from a really good program to the top of the mountain. If they had spent the same money redoing their weight room or building another miniature water-park, it would've had much less impact. Give $1M more per year to the program under CPJ to hire a Venables, and the entire program changes overnight (including the revenue aspect). I said the same thing in a post a few months ago. Of course, now all of the pieces are different, so this singular move/variable no longer applies.

 

[vamosjackets]

Our current regime seems to be banking on recruiting success and CGC, himself, being a top-notch DC. So, for our current regime, and extra $1M/year would best be spent on 20 extra recruiting staff or hiring an elite OC. It seems to me that either of those could have a major impact on the program. Or, if other factors preclude success, then it might not make any difference at all. I guess the point is that, an extra $1M/year, well-spent, COULD drastically change the program.

 

[Augusta_Jacket]

Our current regime seems to be banking on recruiting success and CGC, himself, being a top-notch DC. So, for our current regime, and extra $1M/year would best be spent on 20 extra recruiting staff or hiring an elite OC. It seems to me that either of those could have a major impact on the program. Or, if other factors preclude success, then it might not make any difference at all. I guess the point is that, an extra $1M/year, well-spent, COULD drastically change the program.

Split the difference. Assuming the Current OC makes at least $500K and most likely closer to $700k, then bump that to $1.2m and divert the remainder to recruiting staff.

 

[IEEEWreck]

Like you say, it would be hard, maybe even impossible, to actually quantify staff energy. It would be less difficult, though still no walk in the park, to quantify the perception of staff energy. Text analysis of the local papers or of Twitter feeds might work. This, btw, is one of the reasons why I wouldn't touch an actual project to get the analysis done with a fork. Way too much work.

This is an interesting problem. There's substantial published work that is probably quite similar in politics on questions like approval or likeability or in marketing on brand perception. These tend to use word vector based machine learning models. In many ways the blueprint is there to take up, and tuning the model for 'energy' would be a worthwhile exercise if you want to get into machine learning on language datasets. Of course, you'd need a data source, like Lexis Nexis, and some machine learning compute resources. The first is more expensive. Would be fun to hack around on.

I think the larger model after making a hype metric would be problematic though. I suspect trying to control for resources along with a pretty limited data set would be tough.