Is there a way on Pomeroy to recall his predictions for games after the fact? Trying to figure out what the BE should have done to this point in the tourney according to him. Hear everyone saying the BE should be at least something like 12-6 so far instead of 8-7, and obviously the 1st three days have not been good for the league, but the perception seems to be that they should have won every game in which Pomeroy would have had them in the 50/60 percentiles to win. Being as I believe Pomeroy has the best predictive game model, would like to do some research on this. Thanks in advance.
So, do it yourself?
Quote from: Eye on March 20, 2011, 01:10:30 PM
Is there a way on Pomeroy to recall his predictions for games after the fact? Trying to figure out what the BE should have done to this point in the tourney according to him. Hear everyone saying the BE should be at least something like 12-6 so far instead of 8-7, and obviously the 1st three days have not been good for the league, but the perception seems to be that they should have won every game in which Pomeroy would have had them in the 50/60 percentiles to win. Being as I believe Pomeroy has the best predictive game model, would like to do some research on this. Thanks in advance.
Ken Pom picked the Big East to go 10-0 in the first round with one 50-50 game (Nova and George Mason).
Big East went 7-4
In the next round, they Ken Pom predicted UCONN, Pitt, Syracuse, Notre Dame wins with Cincy, WVU, MU losses...or a prediction of 4-3. Right now the conference is a minimum of 2-4...the best we can do is 3-4 if ND wins.
Basically, we haven't performed as a conference up to what Ken Pom predicted. Pretty simple.
Quote from: ChicosBailBonds on March 20, 2011, 01:25:55 PM
Basically, we haven't performed as a conference up to what Ken Pom predicted. Pretty simple.
Or, perhaps, Pom over-predicted. Your choice.
For KenPom predictions, if there are 10 games and he says that BE teams are 51% favorites in each games, he is not predicting 10-0. He is predicting 5-5, maybe 6-5.
Quote from: Benny B on March 20, 2011, 01:28:47 PM
Or, perhaps, Pom over-predicted. Your choice.
It's based on statistics and his prediction algorithm...he doesn't hand pick them.
Quote from: CrazyEcho on March 20, 2011, 01:39:57 PM
For KenPom predictions, if there are 10 games and he says that BE teams are 51% favorites in each games, he is not predicting 10-0. He is predicting 5-5, maybe 6-5.
They weren't 51% favorites, all were significant favorites with the exception of MU over XU (he had us at 51.5%) and the Mason-Nova game.
Today, for example, he predicts ND over Florida State with FSU at a 78% chance of winning....Syracuse with a 68% chance of beating Marquette.
By the way, would it surprise anyone to see Florida State beat Notre Dame today?
Quote from: ChicosBailBonds on March 20, 2011, 01:44:54 PM
They weren't 51% favorites, all were significant favorites with the exception of MU over XU (he had us at 51.5%) and the Mason-Nova game.
Today, for example, he predicts ND over Florida State with FSU at a 78% chance of winning....Syracuse with a 68% chance of beating Marquette.
By the way, would it surprise anyone to see Florida State beat Notre Dame today?
But, the point stands that he would not say he predicted the Big East to go 10-0. You'd have to add up the percentages (MU was 53% I thought) with all the others and that is what it would be.
Quote from: CrazyEcho on March 20, 2011, 01:46:49 PM
But, the point stands that he would not say he predicted the Big East to go 10-0. You'd have to add up the percentages (MU was 53% I thought) with all the others and that is what it would be.
Sigh.
OK, let's do it this way. In 10 of the Big East games did Ken Pomeroy's rankings have the Big East teams favored to win using his system in all ten of those games? In the 11th game, was it a 50-50 split?
Quote from: ChicosBailBonds on March 20, 2011, 01:44:54 PM
They weren't 51% favorites, all were significant favorites with the exception of MU over XU (he had us at 51.5%) and the Mason-Nova game.
Today, for example, he predicts ND over Florida State with FSU at a 78% chance of winning....Syracuse with a 68% chance of beating Marquette.
By the way, would it surprise anyone to see Florida State beat Notre Dame today?
I wouldn't be surprised at all.
FSU looked great the other night.
Quote from: ChicosBailBonds on March 20, 2011, 01:49:41 PM
Sigh.
OK, let's do it this way. In 10 of the Big East games did Ken Pomeroy's rankings have the Big East teams favored to win using his system in all ten of those games? In the 11th game, was it a 50-50 split?
I don't fault you for not understanding statistics. I do fault you for clearly not even looking at the numbers before posting what you think they said. Here are his actual predictions:
WVU was 49% to win
Cinci was 58% to win
St. Johns was 48% to win
UCONN was 85% to win
Louisville was 88% to win
Pitt was 95% to win
http://www.kenpom.com/fanmatch.php?d=2011-03-17
Nova was 50% to win
Marquette was 53% to win
Georgetown was 74% to win
Cuse was 90% to win
ND was 94% to win
http://www.kenpom.com/fanmatch.php?d=2011-03-18
So, even taking your understanding, you are wrong b/c he predicted WVU and St. Johns to lose with one true toss-up.
Doing it correctly, you add them up and get 7.84 wins. He would have predicted the BE to get 7.84 wins in the first round.
I have the printouts from Sunday night on my desk when I made my bracket picks and Ken Pom had Big East teams favored in ten of the 11 matchups. He also puts a W or L next to it. It's right in front of me and he has 10 W's in the first games by the Big East teams. Just as today he has a "L" next to MU http://www.kenpom.com/team.php?team=Marquette
I appreciate the link...what I have printed out is directly from Ken Pom. I most certainly LOOKED at them and even printed them out. Hell, I'm looking at them again right now.
Quote from: ChicosBailBonds on March 20, 2011, 02:21:28 PM
I have the printouts from Sunday night on my desk when I made my bracket picks and Ken Pom had Big East teams favored in ten of the 11 matchups. He also puts a W or L next to it. It's right in front of me and he has 10 W's in the first games by the Big East teams. Just as today he has a "L" next to MU http://www.kenpom.com/team.php?team=Marquette
I appreciate the link...what I have printed out is directly from Ken Pom. I most certainly LOOKED at them and even printed them out. Hell, I'm looking at them again right now.
Do you understand that even if he had the Big East favored in 10/11 matchups, that he is not predicting that the big east would go 10-1?
Quote from: CrazyEcho on March 20, 2011, 02:23:00 PM
Do you understand that even if he had the Big East favored in 10/11 matchups, that he is not predicting that the big east would go 10-1?
I understand that he is saying that using his formula and algorithm, 10 of the 11 are FAVORED to win those games. Do you disagree?
Quote from: ChicosBailBonds on March 20, 2011, 02:24:59 PM
I understand that he is saying that using his formula and algorithm, 10 of the 11 are FAVORED to win those games. Do you disagree?
Well, I would agree that KenPom (according to the links I provided) said that Big East Teams were favored in 8 of the games, underdogs in two, and toss-up in one.
This thread (and the question you purported to answer and I actually answered) is about how many victories he would have predicted the Big East to have in the first round of the tournament. The answer is between 7 and 8.
Quote from: ChicosBailBonds on March 20, 2011, 02:24:59 PM
I understand that he is saying that using his formula and algorithm, 10 of the 11 are FAVORED to win those games. Do you disagree?
You are wrong on this one Chicos. If you e-mail KenPom we will tell you straight up that he would NOT have predicted 10 wins for the Big East in the second round. That would be the equivalent of saying his number one team in the country should go undefated this year, but we will all know that his website NEVER EVER predicts that.
Having teams favored to win games by numbers like 60% is barely favored in his book. That means in ten games the favored team only wins six of them.
He would have predicted the Big East to go 8-3 in the second round not 10-1.
Quote from: ChicosBailBonds on March 20, 2011, 02:24:59 PM
I understand that he is saying that using his formula and algorithm, 10 of the 11 are FAVORED to win those games. Do you disagree?
It's simple statistics. Say a team has a 73% confidence of winning. That means that 7.3 days out of 10, that team will win. 2.7 days out of ten, that team will lose.
So if you extrapolate that across the Big East, it would have been a statistical anomaly (something like .4%) if the Big East went 11-0 or 10-1.
Got my answer for the first round, he would have said 8-3 adding up the percentages given by Crazy Echo. So 7-4 wasn't good, below average, but not the disaster it's been portrayed as in the national media, and by nearly everyone I'm forced to deal with here in Rodentland. Problem is two of the individual losses were bad (UL and margin for Gtown).
Also got my answer on how to figure it out for yesterday's and today's ND - FSU game, too.
Thanks CE.
Edit: Would have been 2-2 yesterday, so now predicted mark would be 10-5. Once again 8-7 below average, but not the disaster as portrayed nationally and locally. Problem once again is one of the losses (Pitt) was bad.
2nd Edit: Time to make up for it with UConn beating SD State, ND making the Final 4 and MU beating UNC next weekend.
Quote from: Eye on March 20, 2011, 03:34:13 PM
Got my answer for the first round, he would have said 8-3 adding up the percentages given by Crazy Echo. So 7-4 wasn't good, below average, but not the disaster it's been portrayed as in the national media, and by nearly everyone I'm forced to deal with here in Rodentland.
Also got my answer on how to figure it out for yesterday's and today's ND - FSU game, too.
Thanks CE.
No problem, man. Sorry about the snarky first response!