Pomeroy: Florida Will Probably Lose; It Will Probably Be Close

[UFreak]

Originally Posted by MaceoP

Florida Gators Football under Steve Spurrier 1996-2001

We had a record of 61-14, we were a dominant team.

Under SOS, in games decided by 7 points or less, we were 5-8 (38% win percentage) during that period

Anybody believe our problems during our golden years were SOS couldn't win a close game? It's a statistic that is meaningless when taken out of context of being a dominant team.

Contrast that with our football team last year, which won a majority of close games.

But that's also mitigated by an 81 percent winning percentage.

[madgator]

Originally Posted by SmootyGator

I'm sure that if Erving Walker hit all of those last second threes he took, Billy would be at 55%, right?

absolutely awesome!


rep for you sir!

[madgator]

Originally Posted by Matthanuf06

Style of play plays a role, and the coach decides that.

Coach Muschamp should do better than Steve (UF version) in close games, but he will do worse overall because he will have more close games.

the problem with that when it comes to Billy D. is that the blow out wins (25 points or more) are not happening at a higher rate than the majority of the coaches listed.

the rates of occurance are typical

[madgator]

Originally Posted by Jonas

Then the 5 point gamw stat is misleading. That's why I want to know where he is getting the numbers so I could look it up. I think he is getting it from that Brockway article from a while ago.

move the goals posts much?

or just 3 times per debate?

[madgator]

Originally Posted by InstiGATOR1

UF is a slightly worse FT shooting team this year compared to 2011, but a slightly better FT shooting team this year compared to 2010.

our offensive and defensive efficiency stats are also far better than either of those two years.


there is a reason why vegas has us as 3rd best odds to win.....AS A 3 SEED!

so go ahead akagator.....excuse me for coming across as confused. Seems like the stats sure are and the analysts interpretations of them.




this years team is seriously an analytical nightmare.

quality wins by huge margins against tournament qualifying out of conference teams.

offensive and defensive efficiency ratings through the roof

losses that don't seem to make sense


I've got them in the final losing against Louisville but wouldn't be shocked if they lose first or second round.....

if only we had Brad Beal

[Matthanuf06]

Originally Posted by madgator

the problem with that when it comes to Billy D. is that the blow out wins (25 points or more) are not happening at a higher rate than the majority of the coaches listed.

the rates of occurance are typical

Right. Billy's teams are worse in close games as compared to the elite coaches.

Of course the sample of elite coaches has quite a bit of survivorship bias in it.

[Jonas]

Originally Posted by madgator

move the goals posts much?

or just 3 times per debate?

misread much? I never moved the goal posts. That particular quote was a hypothetical. I'm wondering where you are getting your stats from because I want to see if it's consistent if you changed the criteria to a 10 point games instead of a 5 point game.

[Jonas]

mad, where are you getting your numbers from? They differ from what Brockway has.

http://brockway.blogs.gatorsports.co...ontemporaries/

[Matthanuf06]

Originally Posted by Jonas

misread much? I never moved the goal posts. That particular quote was a hypothetical. I'm wondering where you are getting your stats from because I want to see if it's consistent if you changed the criteria to a 10 point games instead of a 5 point game.

As the MOV builds our odds of victory increase.

If you segment the data Billy is going to look similar to other elite coaches in certain segments. I'm not sure how that would discredit the numbers in one-two possession games.

Coaches can impact outcomes in-game far more in 1-2 possession games than 3-4 possession games. My guess is your win probability for anything beyond say 7 points is really a function of how good your team is and not anything to do with in game strategy. Coaches influence that in various ways, scheme, recruiting, teaching, etc.

[Jonas]

well, I don't really want to say anything definitive till I see the numbers, but I'm just curious if this arbitrary 5 point MOV requirement is influencing the percentage. The KenPom article from the OP made the point that we had close games, but we played well down the stretch and turned them into 10+ point wins. Shouldn't those games show that Billy can be good in close games? But they aren't considered in that statistic at all.

I'm not sure if I follow this :Coaches can impact outcomes in-game far more in 1-2 possession games than 3-4 possession games.

I would think it would be the opposite because when the game comes down to the wire, all a coach can do is put his player in a position to get a good shot. So if the game is won is more dependent on that player's shooting percentage, hence why a lot of the stat guys consider close games to be a larger function of variance. When you have larger possession games, the overall in-game strategy will be less influenced by variance.

[madgator]

Originally Posted by Jonas

mad, where are you getting your numbers from? They differ from what Brockway has.

http://brockway.blogs.gatorsports.co...ontemporaries/

stat sheet

[Matthanuf06]

Because the larger the MOV the less the outcome is influenced by a single chess move but rather a function of the pieces you actually have on the board so to speak. Now the coaches clearly impact that, but it's a different skill than what we are talking about.

[MaceoP]

If you look at this year in particular. Excluding the last game against ole miss where we blew a big lead...

We played 9 games this year where the score was within 4 points at halftime.. Of those 9 games, we won 6 (all by double digits), and lost 3 by a total of 8 points!

[swampgas44]

Originally Posted by tebowharvin

This is not true. He weights more recent games.

Not true. His ratings are statistically driven for the whole season with no weight assigned to date played. Check his algorithm that is used.

[benheb]

Originally Posted by Jonas

well, I don't really want to say anything definitive till I see the numbers, but I'm just curious if this arbitrary 5 point MOV requirement is influencing the percentage. The KenPom article from the OP made the point that we had close games, but we played well down the stretch and turned them into 10+ point wins. Shouldn't those games show that Billy can be good in close games? But they aren't considered in that statistic at all.

I'm not sure if I follow this :Coaches can impact outcomes in-game far more in 1-2 possession games than 3-4 possession games.

I would think it would be the opposite because when the game comes down to the wire, all a coach can do is put his player in a position to get a good shot. So if the game is won is more dependent on that player's shooting percentage, hence why a lot of the stat guys consider close games to be a larger function of variance. When you have larger possession games, the overall in-game strategy will be less influenced by variance.

This makes sense. I don't buy numbers created by arbitrary definitions of "close" or number of set plays. Also, the losses we have this year reflect what you are saying. Billy D has gotten guys in position with a good look or at the foul line with the game on the line and the shot haven't fallen.

[corpgator]

Originally Posted by swampgas44

Not true. His ratings are statistically driven for the whole season with no weight assigned to date played. Check his algorithm that is used.

Yes true. A tweet from the man himself:

@mitchgoldich More recent games do get more weight. Might overrate Lehigh slightly, but not too much.

https://twitter.com/kenpomeroy/statu...24015031341057

[benheb]

Originally Posted by madgator

Go ahead and read what I was writing about Nate Silver. Never once "smeared his model" or what he was saying. What I was doing was saying that there were several historical factors (some of which are consistent over 100 years) that his model was running head long into and the Obama supporters were focusing on Silver and ignoring the historical trends. There were very credible statisticians not named Nate Silver who came to different conclusions. So I'm not sure what your point is there.

however, I do understand what a probability model is and did quite well in all of my statistics classes at UF.

obviously you don't understand what I am saying here either. When you compare donovan against the long list of coaches above, there is a trend that is fairly consistent. 55%. In fact, amongst the top coaches the % is more between 55-65%.

and coaching/playing basketball has a lot more variables than those involved in a coin flip. So a pure 50/50 win/loss outcome analysis with "luck" being the only determining factor is simply ridiculous.

I've been trying to substantiate my point with some details as to why other than "luck."

I think his point was that you were demonstrably wrong. Nate Silver's model was nearly flawless and you characterizing those advocating his approach as "Obama supporters" shows that you don't understand statistics. I don't know how your "historical factors" or "historical trends" argument could possibly be more exposed as completely invalid.

Silver became significant because he's almost always been right with his statistical analysis of baseball or elections. Read Nate's analysis of the Gators in his tournament article for a spot on interpretation of the meaning of our scoring differential this year.

The Billy D close game stats don't mean anything significant.

[madgator]

Originally Posted by benheb

I think his point was that you were demonstrably wrong. Nate Silver's model was nearly flawless and you characterizing those advocating his approach as "Obama supporters" shows that you don't understand statistics. I don't know how your "historical factors" or "historical trends" argument could possibly be more exposed as completely invalid.

Silver became significant because he's almost always been right with his statistical analysis of baseball or elections. Read Nate's analysis of the Gators in his tournament article for a spot on interpretation of the meaning of our scoring differential this year.

The Billy D close game stats don't mean anything significant.

Again, I wasn't advocating Silvers model as "obama supporters" I was saying that obama supporters were holding silver up as the holy oracle pre-election when there were other reputable models coming up with different results.

Wasn't questioning Silver and I certainly know what a probability model is. Again, it was in the context of conflicting models and who gets held up as most credible. And in the spirit of politics, in order to get under the skin of the obama supporters focusing solely on silver, I made a few fantasy baseball wise cracks.

Silver won......so be it.

[tommyuf21]

Originally Posted by madgator

Again, I wasn't advocating Silvers model as "obama supporters" I was saying that obama supporters were holding silver up as the holy oracle pre-election when there were other reputable models coming up with different results.

Wasn't questioning Silver and I certainly know what a probability model is. Again, it was in the context of conflicting models and who gets held up as most credible. And in the spirit of politics, in order to get under the skin of the obama supporters focusing solely on silver, I made a few fantasy baseball wise cracks.

Silver won......so be it.

Not to get into politics too much here, but Silver's projections were even more optimistic than most Obama supporters were. From someone who worked on the ground, I often have issues with polling since so many potential voters are not accessible anymore. Silver's model pretty much ignores political organizing or journalism and that's where the criticism comes in. I was confident that Obama would win, mainly because of the Ohio screw up (with Chrysler) and Romney's terrible ground game. However, I thought that several of the swing states like Florida, Virginia would go red and keep Mitt a lot closer in the final tally. Obviously, Nate knew something we didn't or he was just lucky. I guess there's no way to ever really know.

Here's a quote from Joe Scarborough, who I actually admire, in regards to his dispute with Nate.

"I won’t apologize to Mr. Silver for predicting an outcome that I had also been predicting for a year. But I do need to tell Nate I’m sorry for leaning in too hard and lumping him with pollsters whose methodology is as rigorous as the Simpsons’ strip mall physician, Dr. Nick. For those sins (and a multitude of others that I’m sure I don’t even know about), I am sorry.

Politics is a messy sport. And just as ball players who drink beer and eat fried chicken in dugouts across America can screw up the smartest sabermatrician’s forecast, Nate Silver’s formula is sure to let his fervent admirers down from time to time. But judging from what I saw of him this morning, Nate is a grounded guy who admits as much in his book. I was too tough on him and there’s a 84.398264% chance I will be less dismissive of his good work in the future"

[BobK89]

Suck it, Pomeroy. Gatas up 20 in the second half!

Sent from my iPhone using GatorCountry