New Stuff

I added a couple of features to RPIForecast.com:

1. Conference Pages: Each conference now has a page so that you can see how each team in the conference stacks up. Items on the pages include things like current and forecasted records for both conference and non conference games along with the usual RPI forecasts. To give you an example, here is the ACC's page.
2. As mentioned in #1, I have added summaries of conference W-L record forecasts and current conferce records to the main page and the individual team and conference pages. Additionally, on the team pages, you will also find the current OOC W-L record (along with the forecast). Ideally, I would include everything on the main page, but there just isn't room.

An Undefeated Conference Season?

Well, since I made the last post, Pittsburgh beat Duke, Memphis beat Georgetown, Mississippi beat Clemson, and Michigan St. beat Texas, so that now only 9 undefeated teams remain. Of course, 3 of those games were between unbeaten teams, so it was bound to happen.

An event that is not quite as rare is an undefeated conference regular season. With the exception of 2001 and 2002, at least two teams have gone undefeated in conference play in each of the last 10 years:

2007 - Memphis (CUSA) and Winthrop (Big South)
2006 - Bucknell (Patriot), George Washington (A10) and Gonzaga (WCC)
2005 - Davidson (Southern) and Pacific (WAC)
2004 - Austin Peay (OVC), Gonzaga (WCC) and St. Joseph's (A10)
2003 - Kentucky (SEC), Penn (Ivy), and Weber St. (Big Sky)
2002 - Kansas (Big 12)
2001 - none
2000 - Cincinnati (CUSA), Penn (Ivy), and Utah St. (Big West)
1999 - College of Charleston (Southern), Duke (ACC), and Utah (WAC)
1998 - Princeton (Ivy) and TCU (WAC)

Which teams are most likely to do so this coming year? Using Jeff Sagarin's PREDICTOR as of today, the 20 most likely candidates are listed below:

Rank Team Conf Prob. conf games
1 Winthrop, BSth 13.2% 14
2 Morgan St., MEAC 11.5% 17
3 Gonzaga, WCC 8.9% 14
4 Holy Cross, Pat 6.7% 14
5 Memphis, CUSA 6.3% 16
6 Butler, Horz 5.7% 18
7 Kansas, B12 5.5% 16
8 West Virginia, BE 5.5% 18
9 Duke, ACC 4.7% 16
10 Hampton, MEAC 3.6% 18
11 St. Mary's, WCC 3.4% 14
12 Davidson, SC 3.0% 20
13 Texas Arlington, Slnd 2.9% 16
14 Xavier, A10 2.7% 16
15 Brown, Ivy 2.0% 14
16 MD Baltimore County, AE 1.9% 16
17 Rhode Island, A10 1.5% 16
18 Alabama St., SWAC 1.4% 18
19 North Carolina, ACC 1.4% 16
20 South Alabama, SB 1.2% 18

Here are the conference-by-conference leaders:

A10 - Xavier 2.73%
ACC - Duke 4.73%
AE - MD Baltimore County 1.92%
ASun - East Tennessee St. 0.47%
B10 - Wisconsin 0.51%
B12 - Kansas 5.47%
BE - West Virginia 5.45%
BSky - Portland St. 0.12%
BSth - Winthrop 13.20%
BW - UC Santa Barbara 0.66%
CAA - Virginia Commonwealth 0.26%
CUSA - Memphis 6.34%
Horz - Butler 5.69%
Ivy - Brown 2.03%
MAAC - Siena 0.44%
MAC - Miami OH 0.70%
MEAC - Morgan St. 11.50%
MVC - Drake 0.82%
MWC - New Mexico 0.43%
NEC - Robert Morris 0.08%
OVC - Austin Peay 0.14%
P10 - UCLA 0.06%
Pat - Holy Cross 6.71%
SB - South Alabama 1.24%
SC - Davidson 2.99%
SEC - Tennessee 0.80%
SWAC - Alabama St. 1.44%
Slnd - Texas Arlington 2.85%
Sum - IUPUI 0.76%
WAC - Nevada 0.53%
WCC - Gonzaga 8.89%


Notice that in some of the conferences (like the WCC, and MEAC) two teams have a nontrivial chance at going undefeated. With this in mind, here are the conferences ranked by the probability that any of the teams goes undefeated:

rank conf teams prob
1 MEAC 12 15.1%
2 BSth 8 13.2%
3 WCC 8 12.3%
4 Pat 8 6.7%
5 CUSA 12 6.4%
6 ACC 12 6.1%
7 BE 16 6.0%
8 Horz 10 5.9%
9 B12 12 5.6%
10 A10 14 5.1%
11 Slnd 12 3.3%
12 SC 11 3.1%
13 SB 13 2.3%
14 Ivy 8 2.1%
15 AE 9 2.0%
16 SWAC 10 1.4%
17 Sum 10 1.4%
18 SEC 12 1.1%
19 WAC 9 0.9%
20 MVC 10 0.9%
21 BW 9 0.8%
22 B10 11 0.8%
23 MAC 12 0.8%
24 MWC 9 0.7%
25 ASun 12 0.6%
26 MAAC 10 0.6%
27 CAA 12 0.5%
28 BSky 9 0.3%
29 OVC 11 0.3%
30 NEC 11 0.1%
31 P10 10 0.1%

From these probabilities, you can easily calculate the chance that at least one D1 team will go undefeated in conference play this season. In this case, the probability is 67.2%. This seems slightly low since 90% of the last 10 years it happened, but maybe we would have seen a similar number in 2001 or 2002.

An Undefeated Regular Season?

As of yesterday, there were 13 unbeaten teams in Division 1. Which one of these teams is most likely to finish the regular season undefeated? Well, to get an idea, I used Jeff Sagarin's PREDICTOR (what I use in my simulations) and Ken Pomeroy's Ratings to calculate the probabilties that each of the 13 teams will finish the regular season undefeated. Here are the results:

Using Sagarin's PREDICTOR:
rank Team Probability
1 Duke - one in 23 ( 4.2729% )
2 Kansas - one in 57 ( 1.7672% )
3 North Carolina - one in 154 ( 0.6509% )
4 Memphis - one in 334 ( 0.2991% )
5 Mississippi - one in 616 ( 0.1624% )
6 Texas - one in 1,797 ( 0.0556% )
7 Sam Houston St. - one in 3,820 ( 0.0262% )
8 Georgetown - one in 11,074 ( 0.0090% )
9 Washington St. - one in 26,243 ( 0.0038% )
10 Miami FL - one in 33,082 ( 0.0030% )
11 Pittsburgh - one in 33,801 ( 0.0030% )
12 Vanderbilt - one in 80,256 ( 0.0012% )
13 Clemson - one in 156,704 ( 0.0006% )


Using Pomeroy's Ratings:
rank Team Probability
1 Duke - one in 9 ( 11.7184% )
2 Kansas - one in 17 ( 5.7862% )
3 Memphis - one in 105 ( 0.9502% )
4 Mississippi - one in 1,440 ( 0.0694% )
5 Sam Houston St. - one in 3,087 ( 0.0324% )
6 Texas - one in 3,551 ( 0.0282% )
7 Georgetown - one in 4,830 ( 0.0207% )
8 Washington St. - one in 5,219 ( 0.0192% )
9 North Carolina - one in 6,088 ( 0.0164% )
10 Pittsburgh - one in 12,586 ( 0.0079% )
11 Clemson - one in 14,976 ( 0.0067% )
12 Miami FL - one in 17,175 ( 0.0058% )
13 Vanderbilt - one in 1,112,128 ( 0.0001% )


With the exception of North Carolina, the orderings are very similar. If you believe Ken Pomeroy's numbers, there is a decent chance that a team will run the table during the regular season. It is quite a bit less likely according to Jef Sagarin's numbers.

A Few Additions

I added a couple of things to the conferences page:

1. Current (day-to-day) RPI
2. Current Non Conference Record

I also added the daily RPI to the conference graphs so you can see how the RPI forecast and the daily RPIs change by conference by clicking on the conference name.

In the future, I'll create individual conference pages with a list of the teams and their statistics...sort of like the individual team pages. That will have to wait though.

Teams Now Connected

If you noticed a very slight jump in the ratings of your favorite team between yesterday and today, but your team didn't play, it could be because Jeff Sagarin's PREDICTOR is now based solely on games this season. Even if your team did play, you might have seen an unexpected change.

Is West Virginia the best team in the Big East?

Over at the Bracket Project's Bracket Matrix, the consensus seems to be that West Virginia is a number 9 seed and the 6th best team in the Big East. My projections have West Virginia winning the Big East, going about 26-4 overall and 15-3 in the Big East which would be enough to win the conference title. They are also projected to win the Big East tournament and get a number one seed. So, why is the consensus that they will be a nine seed? Well, alot of the projections are based on the teams performances to-date. Mine are based on their projected records and RPIs in March.

If you look at the polls (Coaches and Press) as of last Monday, West Viriginia was 33rd in one poll and not getting votes in the other. Why the lack of respect for the Mountaineers? They have one loss, which came to Tennessee on a neutral court by two points. Aside from that one game, they have absolutely demolished their competition by an average of 36 points. I supect that many of the experts projecting brackets are using the polls as one of their tools. Because I use Jeff Sagarin's PREDICTOR in my simulations to predict the future RPI and the DANCE CARD methodology along with the forecasted RPIs to predict the tournament field, I don't have to rely only on what has happened to-date, but what will likely happen by the end of the season. One of the main reasons my projections put West Virginia so high is that Sagarin's PREDICTOR ranks them #1. This is due to the fact that they have been utterly humiliating the competition (aside from the Vol's of course). Will this spell success in the BE regular season? That is what Sagarin's numbers seem to suggest. The seeds themselves are somewhat subjective, but unless West Virginia completely falls apart a 9 seed certainly seens too low.

Finally, if you compare today's RPIs to the projected rpi's for the Big East, you get the following:

Current RPIs:
Current RPI Team
12 Providence
13 Marquette
36 Syracuse
39 Louisville
41 West Virginia
45 Pittsburgh
46 Villanova
51 Seton Hall
61 St. John's
63 Connecticut
69 South Florida
71 Notre Dame
84 Georgetown
146 Rutgers
220 DePaul
286 Cincinnati

Projected end-of-regular-season RPIs

ERPI Team
7.2 West Virginia
16.3 Pittsburgh
18.5 Georgetown
20.2 Marquette
39 Louisville
40.1 Providence
44 Notre Dame
47.4 Villanova
56.2 Connecticut
63.1 Syracuse
103.3 Seton Hall
110.1 South Florida
114.6 DePaul
120.5 St. John's
183 Rutgers
204.8 Cincinnati

So, judging by the current day-to-day RPI, West Virginia is sort of a good to middle-of-the-pack Big East team. However, if you look at the projected end-of-regular-season RPI, West Virginia is right at the top. The day-do-day RPI means nothing as far as the selection committee is concerned. The RPI that matters is the one in March, which is the purpose of my projections.

North Dakota St.

It was brought to my attention that North Dakota St. and South Dakota St. are both ineligible for post-season play this year. Not realizing this, I had projected North Dakota St. to get the automatic bid for the Summit League. I'll make the change for tomorrow's projections. At this point, Oral Roberts would likely get the automatic bid.

The following teams are ineligible for post-season play:

Cal St. Bakersfield
Central Arkansas
Kennesaw St.
NJ Inst of Technology
North Carolina Central
North Dakota St.
North Florida
Presbyterian
South Carolina Upstate
South Dakota St.
Winston Salem St.

Schedules, Resutls and Probabilities

I added Schedules, Results and Probabilities to the individual team pages for each team. These include the date, opponent, opponent's RPI Forecast, Location, scores, probabilities of winning and the predicted spread. These probabilities (and spreads) are all derived from Jeff Sagarin's PREDICTOR and are used to generate the random wins and losses used in the 10,000 simulations each day. Basically, you can come up with probabilities of winning if you have a prediction of the spread and some sort of a standard deviation. Click on any team's name to see this new feature. Here is what Florida's looks like:

http://www.rpiforecast.com/teams/Florida.html

As with everything else (except for the tournament projections), these are updated daily.

One more thing: The expected record can be easily calculated from the projections. If you make the assumption that the probabilities are independent across games, then you simply add up the probabilities which will give you the expected wins. The expected losses will be the total number of games minus the expected wins.

Thus, even if a team is favored win in each individual game, like UCLA, for example, when you look at them all as a whole, on average, UCLA will lose 5 out of the 17 remaining. Naturally, this all depends on the probabilities being correct and again, this is an expectation. UCLA could do better or worse in any given simulation (and in reality). But, what I am presenting is the best estimate. If you think your team will do better (or worse? What kind of a fan are you?) then simply look at the RPI Forecast broken down by end-of-season record and pick your favorite.

Blog Archive

If you are new to the site and want to learn a little more about what all of this means, then the blog archive is here:

http://www.rpiforecast.com/blog.html

RPI Forecast Tournament Projections

I have started making the tournament field projections for next March. I know, I know, it is very early in the season to be able to say much about March, but as long as others are doing so, I figured I may as well too. Because my projections are purely objective, they should do quite a bit better than others early in the season, and maybe not quite as well as some very late when there is more likely to be some inside information about which teams will make the cut.


Here is the basic procedure:

1. Make end of season RPI Forecasts which entails:
   a. take completed games as given
   b. Use Jeff Sagarin's updated PREDICTOR to generate probabilities over future games
   c. Simulate 10,000 seasons based on random draws using the probabilities
   d. For each simulated season, calculate RPIs by conference, team, vs. 1-25, etc.
   e. Take averages to get an estimate of Expectations of the above



2. Using the "Dance Card Methodology" and the end-of-season projections from step 1, project the field (including automatic bids).

To be included in the Bracket Project Matrix, you need to take some sort of stance on seeding which is really quite subjective. Basically, I am not spending too much time on that aspect.

Last season, I kept track of the number of correct picks (for the field, not for the seeds) for many of the top sites including:

Basketball Predictions
Beat the Experts
Bracket Express
Bracket Project
Bracket Racket
Bracket WAG
Bracket Watch
Bracketography
Bracketology 101
Bracketology 3
Breaking Down the Bracket
Bryce's Bracket Predictions
Build a Bracket
CBS Sportsline
College Hoops Net
College RPI
Colton Index (JCI)
Crashing the Dance
DhankLily
ESPN Bracketology
FOX Sports
JCI
Jerry Palm
MAG
March Madness 07
Mr. Bracket
MRI Sports
NCAA Bracket Predictions
NCAA Hoops Digest
NetWire
PHSports

I calculated the number of teams that were correctly picked from the field over time, and compared it to the RPIForecast projections. Here are the results:

As you can see, the RPIForecast tournament projections did reasonably well compared to others. One thing to note is that I didn't start tracking others until early January, so there is really no telling what happend before then with the others.

New Season of Forecasts

NCAA Basketball season is here again. Forecasts are now up and running at rpiforecast.com. Because the probabilities used in the simulations are based on Jeff Sagarin's PREDICTOR, for the first few weeks of the season, those are tied to the starting ratings. Keep that in mind. Eventually, everything will be based only on games played this season.

One thing that I am implementing from the get-go is the RPI forecasts broken down by different end-of-season records. That way, if you disagree with what the projected W-L reccord is for your team, you can insert what you may think to be a better prediction and see what the resulting RPI forecast is. Another way of looking at it is: How many games does the team have to win to have an RPI above a certain number.

Click here for an archive of blog entries from last season