Pomeroy Predicts 17-11, 8-6 in WCC

I love KenPom... lots of great stats, easy to see opponents schedules/results, great rankings, etc.

One thing that he does is project out results for the season once there are enough games to get some data in there. As of right now, he predicts Portland to finish 17-11 (18-11 assuming a win over Seattle... he doesn't rank non-D1 teams) with an 8-6 record in WCC - losing to GU and SMC twice and at USF and at Santa Clara.

http://kenpom.com/sked.php?team=Portland&t=p

If that happens, I would be thrilled... a strong winning record and third in the WCC (I checked the other schools). But for me, I was kind of viewing that as the "if everything goes well", high-end prediction. But seeing KenPom say it, that - based on the stats so far - that is the most likely outcome. Which is AWESOME!

Anyway, something to smile about after a tough loss at BYU.

If we finish with this record, Reveno is WCC Coach of the Year. Absolutely no question about it.

Frankly, I think that's a high end projection. I'm still thinking we'll finish in the somewhere around 13 or 14 wins, but I'd love to be surprised.

He also left out whoever the Pilots play the second day of the Berkeley tournament--probably Cal if the Pilots win as he predicted.

For the Pilots of recent years 17 wins is a really big number, and it would be awesome to see. Just curious, how accurate have his projections generally been?

Also cool for the Pilots: He projects Washington to go 20-11 and ranks them 38th right now, making that a very good win for the Pilots.

It seems like a very optimistic prediction to me to. He has us having a 5 game winning streak before a 4 game losing streak. Did you notice that? Plus, after San Diego's recent win over Oregon, I just don't see us sweeping them home and away. Even though Brandon Johnson was huge for that team, they seem to have a system that works. Reveno has to get the team rolling on all cylinders for 17-11 to happen. Here's to hoping for a 17-11 season...

So, we're definitely one of the best shooting mid-majors in the country so far at about 46.7% thanks to Smoulders, Raivio, TJ and the boys (unless that's gone down with losses to BYU and EWU). Any reason we aren't listed at ESPNU in the stats? Are mid-major stats kept totally separate from the rest of the country? If so, that seems pretty lame.

http://sports.espn.go.com/ncb/leaders?cat=teamfg&groupId=&seasonType=2&seasonYear=2009

snap to, folk. I've been a Portland fan for almost 20 years, and we have NEVER had a 97-98% chance of beating anyone.

And some of the Gonzaga predictions look silly.

100% and 98% against San Diego, and 90% and 97% against St. Mary's and Santa Clara in their places just doesn't seem real.

Though they will probably beat Portland State, the odds aren't 99% and they won't do it by 29.

And if he's wrong on any of those 100% and 99% games, it throws the least squares tests on accuracy out the window.

Ha ha... oh man, what a cold shower by Geezer.

I realize he's not perfect and of course it's nothing more than a conversation starter, BUT... Pomeroy has been doing it a long time and is definitely one of the most respected guys out there doing stat research for college basketball. I can tell you with 100% confidence that coaching staffs use his site as an easy way to see lots of complex data about your own team and opponents. So... I put more stock in what he says than, say, predictions from bloggers and writers and stuff like that.

But yes... I would say that 18-11 and third in the WCC is reaching the ceiling of what this team is capable of. But at least seeing Pomeroy predict it means that those aren't just the starry-eyed thoughts of us Purple faithful. This team is improved and can (dare I say should?) start winning more than it looses.

I saw some of the other WCC teams in the ESPNU stats listed in the link above. But, we aren't there. Is it because ESPNU has forgotten about us? Have we not played enough games yet to be in the rankings? OR Did our losses to EWU and BYU really mess up our stats?

Thanks for your answer. I appreciate it.

The stat mentioned (FG made) is a total number, not a percentage.

We've made 174, well below the team listed at number 100 with 230 FG's made. (Gonzaga)

Purplegeezer wrote:snap to, folk. I've been a Portland fan for almost 20 years, and we have NEVER had a 97-98% chance of beating anyone.

And some of the Gonzaga predictions look silly.

100% and 98% against San Diego, and 90% and 97% against St. Mary's and Santa Clara in their places just doesn't seem real.

Though they will probably beat Portland State, the odds aren't 99% and they won't do it by 29.

And if he's wrong on any of those 100% and 99% games, it throws the least squares tests on accuracy out the window.

I'm sure the dude knows more about basketball predictions than I do, but, I swear he must have given his computers some really GOOD VODKA to come up with the Big East as only the third best conference in basketball with teams like UConn, Pitt, and Notre Dame this year. Seriously.

Not to mention that he gives BYU at 10-0 a 14% chance of going undefeated and Gonzaga at 7-1 a 78% chance.

Math don't work

Purplegeezer wrote:Not to mention that he gives BYU at 10-0 a 14% chance of going undefeated and Gonzaga at 7-1 a 78% chance.

Math don't work

My understanding is that it's a 76% chance of going undefeated in the WCC..which wouldn't be impossible for Gonzaga as much as I'd like to see them get some splits with teams...

"While the pythagorean winning percentage is calibrated to the likelihood of winning, the efficiencies are based purely on scoring per possession with no consideration of winning or losing."--Ken Pomeroy.

Huh. Crystal, Ken. Clear.

So you can calculate a Pythagorean record for basketball teams as well. Interesting that it works for a sport other than baseball, too.

The concept of Pythagorean record, for those who don't know, was first developed by baseball statisticians. Bill James discovered that if you take a team's runs scored (RS) and runs allowed (RA) and put them into the formula below, you get a number pretty close to their winning percentage:

(RS^2)/(RS^2 + RA^2)

It's called the Pythagorean record because of the three squares in the equation. Over the years others have played with the exponents to get an even better correlation with a team's winning percentage.

You can use this formula to predict how a team will perform. Example: Two years ago the Mariners were in the lead for the Wild Card in August. But a check of their Pythagorean record showed they were winning more games than you would expect and would probably not play as well as the season went on. That's just what happened: they faded down the stretch and missed out. It doesn't always work, but it's a good predictor for baseball. It sounds like you can adapt the formula for use in basketball using points scored and allowed, I would assume.

All right, so I looked around his site a little more.

He's calculated Portland's Pythagorean winning percentage to be .645. Portland's actual record versus DI teams right now is 5-3, for a .625 winning percentage. His projection means the stats show they will play a little better than they are right now.

A winning percentage of .645 in the 27 games he is counting translates to a record of 17-11. This record is based on how many points the Pilots are scoring and allowing per 100 offensive and 100 defensive possessions, respectively, to this point in the season. His projected record for the Pilots is not really a best or worst case scenario, but a prediction based on how the Pilots have played so far. Obviously this Pythagorean record will change based on how the Pilots play the rest of the season.

Believe it or not, he's actually making sense to me.

And also, just to piggyback on what Trumpet wrote... our strength of schedule has been pretty difficult early on. We've got tough games ahead with the likes of GU and SMC, but we've also got some "cupcakes" with Pep and LMU. I put us, USF, USD, and Santa Clara kind of in the same boat... what remains to be seen is which team(s) from that "middle four" steps up and holds serve at home and goes out and picks up some wins on the road. It'll definitely be a tough road ahead to reach that 17-18 win mark... Santa Clara has been playing a lot better, plus they add Decensae White (a transfer from Texas Tech) at the end of this semester. USF is looking good too - just picked up a real nice road win at Boise State. San Diego is the wild card... how much will they miss Brandon Johnson? But they still have enough to go and beat Oregon here in Portland.

But look at it this way... if we go 4-4 against GU/SMC/Pep/LMU like we should... really where we end up in the WCC is going to depend on our games against USF, SCU, and USD. Winning at home will be crucial. Sneaking one away on the road will be huge. And as of now Pomeroy is predicting us to win all three of those home games and to sneak away one of those road games. That doesn't seem all that unreasonable.

Anyway, I love all the stat stuff in baseball... I don't believe in it blindly and still think that in-person scouting is a crucial step (makeup does matter, know matter how much it irks the statheads), but I think it's so important to look past the box scores and really get into the numbers. Definitely a good way for teams that are otherwise disadvantaged to gain an edge.

I would also add T.J.'s GARP (Grit Above Replacement Player) is a WCC-leading 14.3665.

Ha ha... oh FJM, how I miss thee.

Grit, grinder, gamer, scrap, hustle, heart...

I'm not sure that I really buy that our schedule starts getting easier once we start play against the WCC. Are there examples of times when our early season record was actually worse than our WCC record in recent history? I can't think of any. It seems like it's always the opposite for us.

Plus, there are people like Bobby Knight who state that teams in the early part of the season are only 75% of what they are going to become by the end of the season. I'm not sure we beat UW when they were at their peak; we just gave them a nice early season wake up call. Not that there's anything wrong with that; it was still a great victory. But, beating them in January or February would mean more in my opinion...

I noticed that our projected record went down to 16-12.

athleticjames wrote:I noticed that our projected record went down to 16-12.

And our RPI dropped 30 slots with the win