RPI Forecast Blog

Probabilities/Odds For NCAA Tournament Advancement

2011-03-14T17:15:00.000-07:00

We've added a page that lists the probabilities for reaching each of the rounds of the tournament - along with winning it all - for all 68 teams.

The page is here: http://www.rpiforecast.com/tourneyprobabilites.html

There are two types of probabilites listed: UNCONDITIONAL and CONDITIONAL

You are probably most familiar with UNCONDITIONAL. That tells you the overall probability of a team reaching each of the rounds. The CONDITIONAL probabilities, however, tell you the probability that a team reaches a round given that it has already reached the previous round. For example, a team like UT-San Antonio may have a very low unconditional probability of reaching the Final 4, however, conditional on it having already reached the Elite 8, it will have a much higher probability. We'll update the webpage as we reach each new round with new probabilities. These will change as teams (and potential future matchups) are eliminated.

How'd we do?

2011-03-14T17:02:00.001-07:00

The main goals (and the main strengths) of this site are (1) to provide accurate predictions of end-of-season RPI, and (2) to provide up to the minute current RPI. A secondary goal is to provide predictions of at-large bids and seeds. Up until this season, we were not predicting seeds, just bids. This season, we decided to use the "Dance Card" formulas for seeds in addition to the at-large formulas. In terms of at-large bids, we did a bit worse than seasons past - missing 3 (VCU, USC, and Florida State). However, only one major bracketologist got even 67 out of 68 (Fox Sports). As far as I have seen, nobody got 68. Based on the Bracket Project website, only a handful got 66 or better. Most were 65 or less, so 3 is nothing to be ashamed of - especially considering the fact that this is all done using a simple formula. Typically the Dance Card formula misses no less than 1 or 2, so this season was unusual.

As for seeds, there IS something to be ashamed of. Quite simply, the seed predictions stunk! In the name of objectivity, we wanted to avoid tweaking the seeds and let the formulas do all the work. However, something obviously needs to be done. The plan for next season is probably to abandon the formulas for seeding - or at least to modify them to allow for some "human analysis". We'll stick with the at-large formulas and figure out a better way to seed the teams. At this point, the seeding decisions seem to be more complicated than can be explained by a simple formula.

The big benefit from projecting bids/seeds relative to other bracketologists ought to come EARLIER in the season, rather than LATER. In fact, RPIFORECAST is one of the best at predicting seeds in December/January and February. We have archived all of the major bracketologists' predictions this season and will be determining who were the best and worst EARLY in the season.

Anyway, thanks for visiting, enjoy the tournament, and look for some more exciting things next season!

Bracket Predictions

2011-02-21T13:45:00.001-08:00

I've decided to switch up the at-large and seeding formulas a bit. I had been using an equation that includes conference affiliation and representation on the selection committee. However, just today, I switched to one of the models which does not take these into account. This was mainly because of the problem with the MWC teams seeds (BYU, SDSU, etc). Because the MWC has never been this good, it may not make sense to assume that the same "bias" would exist against them this season. Anyway, BYU immediately jumped to a 2 seed and SDSU to a 5 seed (the highest 5 seed).

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

Tourney Projections: A few new features

2010-11-30T13:11:00.000-08:00

I added the following features with regards to NCAA tournament bracket projections:

1. ON the main bracket page (www.rpiforecast.com/bracket.html), I added more stats: Current and Expected RPI, Expected OOC RPI, Current and Expected Records, and Current and Expected Conference Records

2. The daily projected tournament seed can now be found on the individual team pages (for example, see Duke: www.rpiforecast.com/teams/Duke.html)

3. I've added a graph on the individual team pages below the other two graphs which tracks the history of projected seeds for each team. You will be hard pressed to find others willing to archive the history of their projections.

Next on the to-do list is BracketBusters and Conference Tournaments.

New Updated Bracket predictions (Daily!)

2010-11-24T09:57:00.000-08:00

Based on the new updated "Dance Card" formulas from Jay Coleman, I have started making daily predictions of the field of 68 along with the seeds. A great thing about the methodology is that the factors that determine the at-large bids and the seeds are similar but there are subtle differences. The new model only missed one team from the field and did as well as pretty much anybody at picking the actual seeds. I will be keeping track of my predictions as well as those of the other major bracketologists through the course of the season. After the season is over, I'll present results. If the past is any indicator, my predictions should do especially well early in the season compared to the "human predictions".

Find the predictions here:

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

Live-RPI.com - Minute by Minute RPI Updates - Free!

2010-11-24T09:53:00.000-08:00

You may have noticed a brand new feature this season: live-rpi.com . This new website is for those who don't want to wait until the next morning (or even 5 minutes) to see what the current RPI is. As games finish, the RPI is automatically updated and the page should refresh. If it doesn't, simply click on the Live-RPI.com banner and it should refresh. From about 6 am EST to around 1pm, there are no updates unless games are being played. There is also a section on conference RPI which is updated at the same time.

As always, this is free!

Brackets!

2010-03-14T19:46:00.000-07:00

How'd we do in projecting the final bracket? Not quite as well as previous years. This is disappointing because in past years, the tournament projections have been more accurate than the vast majority of the other prognosticators - especially from November through February. The bracket forecasts are based on Jay Coleman's OLD "Dance Card Formula". This formula was estimated quite a few years ago to model the behavior of the selection committee. Inevitably, the selection committe changes its behavior over time, so maybe this is not a huge surprise.

The good news is that Jay Coleman has just published a new version of his model. In fact, this new version predicted all but one of the at large bids - and the team missed (Florida versus Miss. St.) were ranked right next to each other and had very similar scores. For next season, we will use the improved updated formula and will also start projecting seeds instead of simply ranking them in order of probability of an at-large bid. Jay Coleman and his co-authors also have a new model that predicts seeds which will be used on this site.

Most pundits project brackets using a "what if the season ended today" framework. The approach taken on this site is to project what the field will look like on Selection Sunday. The advantage of combining the simulations on RPIforecast.com and the Dance Card formula is twofold: (1) The actual forecasts of RPI and records which feed into the formula are much more accurate than using the current RPI and record and (2) The Dance Card Formula (especially the new version) is objective and unbiased and does a remarkable job at predicting the field.

Home Court Advantage

2010-02-27T17:05:00.000-08:00

One of the most important aspects of basketball - Home Court Advantage - is also one of the most difficult to measure. There are a few ways to do this. The most common is basically to estimate how many points playing on your home court is worth. Having teams play home-and-away in one season makes this possible. You could look at other variables such as Field Goal Percentage, but those may be affected by many factors besides the venue.

An alternative that came to mind is free throw shooting percentage. The idea is that because you can't really defend the free throw, there are only a few things should affect it: 1) who is shooting, 2) luck, and 3) the fans making noise and other distractions. To take care of number 1, we need to make sure to look not at the total free throw percentage of the road team, but instead look at who is shooting and how they do. That leaves us with luck and actual distractions from fans.

With this in mind, I wanted to figure out for each road team, what would you expect the free throw percentage to be based on who is shooting, and in doing so, you can measure the effect of the venue itself on performance. For those that have some knowledge of stats, I ran the following regression:

FTP_ik = a_i + venue_k + e_ik

where
FTP_ij is the free throw percentage of road player i in venue k
a_i is the unconditional average road free throw percentage of player i
venue_k is the effect of the home team on player i's free throw percentage
e_ik is the error term

venue_k is the thing I'm trying to measure. One more thing, I'm using Free throw attempts as a weight in the regression to give more weight in the estimation to those that shoot more often.

Below is what I came up with. The FT-Effect number is the (venue_k) estimate from above. For UME (Maine), is is -11.2%. That visiting players shot 11.2% lower Free Throw percentage against Maine than they would have in the average road venue. So, Maine it is! See where your team is:

Rank Home Team FT-Effect
1 UME -11.2%
2 DEST -10.6%
3 ORU -10.0%
4 NORF -9.9%
5 SVST -9.6%
6 MICH -9.2%
7 UVA -9.0%
8 SJSU -8.4%
9 FDU -8.4%
10 NEB -8.0%
11 GONZ -7.6%
12 BYU -7.5%
13 SUU -7.4%
14 USC -7.2%
15 CINC -6.9%
16 IONA -6.9%
17 DUQ -6.9%
18 MASS -6.6%
19 SYR -6.4%
20 LAF -6.1%
21 EMU -6.0%
22 TCU -6.0%
23 FGCU -5.5%
24 SOU -5.5%
25 GT -5.5%
26 COST -5.4%
27 UCR -5.4%
28 BETH -5.4%
29 RAD -5.3%
30 TROY -5.1%
31 WSSU -4.9%
32 PEPP -4.6%
33 UTA -4.5%
34 BOIS -4.4%
35 VAND -4.4%
36 USF -4.4%
37 LBSU -4.4%
38 GMU -4.4%
39 WVU -4.1%
40 VCU -4.1%
41 BING -3.9%
42 UNF -3.9%
43 IOWA -3.8%
44 UD -3.8%
45 CITA -3.7%
46 PITT -3.7%
47 MORE -3.7%
48 SIE -3.7%
49 PORT -3.6%
50 ELON -3.6%
51 WICH -3.6%
52 SPC -3.6%
53 NOVA -3.5%
54 WYO -3.4%
55 UNC -3.4%
56 SJU -3.4%
57 SC -3.4%
58 WASH -3.4%
59 ETSU -3.4%
60 ARMY -3.3%
61 NWSU -3.3%
62 AFA -3.2%
63 KENN -3.2%
64 TEX -3.2%
65 ID -3.2%
66 UCF -3.2%
67 MTSU -3.1%
68 TOWS -3.0%
69 ODU -3.0%
70 LAM -3.0%
71 UMES -3.0%
72 UNCW -2.9%
73 PSU -2.9%
74 LIP -2.9%
75 USU -2.9%
76 FSU -2.8%
77 SIUE -2.8%
78 SCL -2.8%
79 UNI -2.8%
80 SDAK -2.8%
81 BUFF -2.7%
82 AMER -2.7%
83 WKU -2.6%
84 OHST -2.6%
85 BSU -2.6%
86 CHAT -2.6%
87 NIAG -2.4%
88 VALP -2.4%
89 CLEM -2.4%
90 VMI -2.2%
91 USM -2.2%
92 CCU -2.2%
93 MON -2.1%
94 USD -2.0%
95 IND -2.0%
96 CAL -2.0%
97 LONG -2.0%
98 SCST -2.0%
99 WOFF -2.0%
100 SCU -1.9%
101 LTU -1.9%
102 QUIN -1.9%
103 MORG -1.8%
104 OU -1.8%
105 YSU -1.8%
106 GW -1.7%
107 UCSB -1.7%
108 SLU -1.6%
109 UCD -1.6%
110 UWM -1.6%
111 NMSU -1.6%
112 MOST -1.5%
113 PVAM -1.5%
114 NM -1.5%
115 SEMO -1.5%
116 UTAH -1.4%
117 IUPUI -1.4%
118 XU -1.4%
119 NIU -1.4%
120 INST -1.4%
121 STET -1.4%
122 NW -1.4%
123 POLY -1.3%
124 PRIN -1.2%
125 CAMP -1.2%
126 FURM -1.2%
127 BONA -1.1%
128 PUR -1.1%
129 ARIZ -1.1%
130 BC -1.1%
131 SAMF -1.0%
132 HAMP -0.9%
133 FAU -0.9%
134 FORD -0.9%
135 DUKE -0.9%
136 ASU -0.9%
137 NCAT -0.9%
138 FRES -0.8%
139 UIC -0.8%
140 UNCG -0.8%
141 LOYO -0.8%
142 SIU -0.7%
143 NAU -0.7%
144 NICH -0.7%
145 JSU -0.7%
146 SMC -0.7%
147 COFC -0.6%
148 ORE -0.5%
149 WRIG -0.5%
150 MSM -0.5%
151 PROV -0.5%
152 CANI -0.5%
153 MIO -0.5%
154 UMT -0.5%
155 UCLA -0.4%
156 HART -0.4%
157 EVAN -0.4%
158 MINN -0.4%
159 SDST -0.4%
160 KENT -0.3%
161 ALA -0.3%
162 MAN -0.3%
163 TUL -0.3%
164 COLO -0.3%
165 HAW -0.2%
166 LIUB -0.2%
167 JAX -0.2%
168 RID -0.1%
169 NCC -0.1%
170 WAG -0.1%
171 NCST -0.1%
172 ALC 0.0%
173 SDSU 0.0%
174 CORN 0.0%
175 UTEP 0.0%
176 UNH 0.0%
177 BELM 0.1%
178 MURR 0.1%
179 URI 0.2%
180 CONN 0.2%
181 SB 0.2%
182 NOCO 0.2%
183 WAKE 0.2%
184 FIU 0.2%
185 WISC 0.2%
186 TNT 0.2%
187 LEH 0.3%
188 FAMU 0.4%
189 DEP 0.4%
190 UTPA 0.4%
191 LSU 0.4%
192 WIU 0.4%
193 CHSO 0.4%
194 NAVY 0.4%
195 LIB 0.5%
196 SAMH 0.5%
197 WEB 0.5%
198 NJIT 0.6%
199 KU 0.6%
200 TXST 0.6%
201 MERC 0.7%
202 HOU 0.7%
203 KSU 0.7%
204 MTST 0.7%
205 ILST 0.7%
206 SFNY 0.8%
207 BAY 0.8%
208 TXSO 0.9%
209 BUTL 0.9%
210 DEN 1.0%
211 UVU 1.0%
212 UMKC 1.0%
213 ALAM 1.1%
214 HBU 1.1%
215 UMD 1.1%
216 UCA 1.1%
217 WSU 1.2%
218 UDM 1.2%
219 GWB 1.2%
220 OSU 1.2%
221 CCSU 1.2%
222 CLST 1.3%
223 TAM 1.3%
224 COLG 1.3%
225 RUTG 1.4%
226 PENN 1.4%
227 NTEX 1.4%
228 MEM 1.4%
229 TTU 1.4%
230 WMU 1.4%
231 STJ 1.5%
232 GU 1.5%
233 ILL 1.5%
234 AUB 1.5%
235 SMU 1.5%
236 UVM 1.6%
237 UAPB 1.7%
238 UWGB 1.7%
239 CST 1.8%
240 MAR 1.8%
241 COLU 1.8%
242 DART 1.9%
243 CENT 1.9%
244 FLA 1.9%
245 MARQ 1.9%
246 VT 1.9%
247 COPP 1.9%
248 IDST 2.0%
249 DREX 2.0%
250 APST 2.0%
251 UNCA 2.0%
252 ND 2.1%
253 EWU 2.1%
254 GS 2.2%
255 EIU 2.2%
256 CWM 2.2%
257 ECU 2.3%
258 CMU 2.3%
259 UALR 2.4%
260 UNLV 2.4%
261 GRAM 2.4%
262 OAK 2.5%
263 SF 2.5%
264 ARST 2.5%
265 NEV 2.6%
266 UTM 2.6%
267 UL 2.7%
268 MCN 2.7%
269 BUCK 2.7%
270 ULL 2.8%
271 APSU 2.8%
272 ALB 2.8%
273 HOFS 2.9%
274 ISU 2.9%
275 TOL 3.0%
276 UGA 3.1%
277 RICH 3.1%
278 NDST 3.2%
279 UTSA 3.2%
280 LAS 3.2%
281 STAN 3.2%
282 MVSU 3.2%
283 MSU 3.2%
284 IPFW 3.3%
285 GSU 3.3%
286 MIZZ 3.4%
287 MIA 3.4%
288 TAMC 3.6%
289 ULM 3.7%
290 DRKE 3.7%
291 BRY 3.7%
292 HP 3.8%
293 UNO 3.8%
294 PAC 3.9%
295 CREI 3.9%
296 MSST 3.9%
297 NU 3.9%
298 HC 4.0%
299 NDAK 4.0%
300 ORSU 4.0%
301 TU 4.1%
302 CSF 4.1%
303 PRES 4.1%
304 MISS 4.2%
305 RICE 4.4%
306 UAB 4.4%
307 FAIR 4.4%
308 HOW 4.5%
309 BRWN 4.5%
310 SH 4.5%
311 BGSU 4.6%
312 YALE 4.6%
313 TENN 4.8%
314 AKR 4.8%
315 JMU 4.9%
316 WINT 5.0%
317 JXST 5.2%
318 SFA 5.2%
319 SHU 5.3%
320 MRSH 5.3%
321 DEL 5.4%
322 OHIO 5.5%
323 UK 5.5%
324 LMU 5.5%
325 UCI 5.6%
326 SEA 5.6%
327 TSU 5.7%
328 DAV 6.0%
329 SFPA 6.1%
330 USA 6.3%
331 LMD 6.4%
332 CHAR 6.4%
333 RMU 6.6%
334 ARK 6.6%
335 EKU 6.7%
336 CSN 6.7%
337 SELU 6.7%
338 SAC 6.8%
339 ALST 7.7%
340 TULS 7.9%
341 BU 8.1%
342 WCU 8.1%
343 UMBC 8.2%
344 CSB 8.3%
345 POST 8.6%
346 BRAD 9.6%
347 HARV 10.3%

These numbers are all relative to the effect of the average venue, not relative to performance at home. I didn't include free throw percentage of any players at home in the estimation.

Here are the same estimates using equal weights (ignoring how many free throw attempts a player had):

Rank Venue FT-Effect
1 SVST -14.3%
2 UME -12.0%
3 NORF -11.8%
4 UNF -11.0%
5 MASS -10.9%
6 FDU -10.8%
7 BYU -10.2%
8 SUU -10.0%
9 MICH -10.0%
10 NEB -9.9%
11 GT -9.8%
12 CINC -9.7%
13 UVA -9.6%
14 AFA -9.0%
15 COST -8.7%
16 UCR -8.6%
17 PORT -8.5%
18 WYO -8.1%
19 LAF -7.9%
20 MORE -7.0%
21 SYR -7.0%
22 SOU -7.0%
23 IONA -7.0%
24 USC -6.8%
25 UIC -6.8%
26 DEST -6.8%
27 UVU -6.6%
28 IOWA -6.6%
29 PRIN -6.3%
30 BETH -6.2%
31 EMU -6.0%
32 SIUE -6.0%
33 GONZ -5.9%
34 SPC -5.9%
35 VAND -5.9%
36 PEPP -5.8%
37 UMT -5.8%
38 QUIN -5.7%
39 WICH -5.7%
40 WASH -5.5%
41 ODU -5.5%
42 SMC -5.3%
43 ARMY -5.3%
44 SCL -5.2%
45 ID -5.2%
46 BOIS -5.1%
47 TROY -4.9%
48 DUKE -4.9%
49 ASU -4.9%
50 PSU -4.9%
51 TCU -4.8%
52 UNI -4.8%
53 FGCU -4.8%
54 SJU -4.7%
55 ORU -4.7%
56 ALA -4.7%
57 WVU -4.7%
58 UD -4.7%
59 MSM -4.5%
60 VALP -4.5%
61 ETSU -4.4%
62 BSU -4.4%
63 GMU -4.3%
64 SDSU -4.3%
65 IUPUI -4.1%
66 UNC -4.1%
67 NIU -4.1%
68 SCST -4.0%
69 RID -4.0%
70 SDAK -4.0%
71 SJSU -3.9%
72 HART -3.9%
73 HBU -3.8%
74 KENN -3.7%
75 ELON -3.7%
76 UTA -3.6%
77 ORE -3.6%
78 NOCO -3.4%
79 NOVA -3.3%
80 VMI -3.3%
81 XU -3.3%
82 FSU -3.2%
83 UCSB -3.2%
84 BC -3.1%
85 OAK -3.1%
86 GW -3.1%
87 NJIT -3.1%
88 YSU -3.0%
89 LBSU -3.0%
90 SIU -2.9%
91 UCF -2.9%
92 NCAT -2.9%
93 RAD -2.8%
94 BING -2.8%
95 SEMO -2.8%
96 SDST -2.8%
97 VCU -2.8%
98 MTSU -2.8%
99 AMER -2.8%
100 MTST -2.7%
101 DUQ -2.6%
102 WKU -2.6%
103 LAM -2.6%
104 STJ -2.6%
105 SLU -2.4%
106 SB -2.3%
107 NIAG -2.3%
108 LONG -2.3%
109 NMSU -2.2%
110 CHAT -2.2%
111 UNCG -2.2%
112 MORG -2.2%
113 KENT -2.2%
114 EVAN -2.2%
115 CHSO -2.1%
116 CAL -2.1%
117 TEX -2.1%
118 USF -2.1%
119 MON -2.0%
120 LIB -2.0%
121 UTPA -2.0%
122 URI -2.0%
123 CAMP -1.9%
124 NAU -1.9%
125 BUFF -1.9%
126 CANI -1.8%
127 UMES -1.8%
128 NAVY -1.8%
129 NM -1.7%
130 PITT -1.7%
131 CLEM -1.7%
132 LAS -1.6%
133 WOFF -1.6%
134 STET -1.6%
135 UVM -1.6%
136 DEN -1.5%
137 MARQ -1.5%
138 UCD -1.5%
139 SIE -1.4%
140 INST -1.3%
141 WSSU -1.3%
142 UWM -1.3%
143 LEH -1.2%
144 WAKE -1.1%
145 MOST -1.1%
146 USU -1.1%
147 MERC -1.1%
148 FAMU -1.1%
149 SC -1.0%
150 CITA -1.0%
151 FAIR -1.0%
152 LIP -1.0%
153 CST -0.9%
154 VT -0.9%
155 COFC -0.9%
156 UWGB -0.9%
157 UDM -0.7%
158 NICH -0.7%
159 USD -0.7%
160 UTAH -0.7%
161 SCU -0.7%
162 CORN -0.6%
163 LOYO -0.6%
164 FORD -0.6%
165 LTU -0.5%
166 SFNY -0.5%
167 UCA -0.5%
168 CCU -0.5%
169 PUR -0.5%
170 BONA -0.4%
171 BUTL -0.3%
172 CLST -0.3%
173 MURR -0.3%
174 ALC -0.3%
175 IND -0.2%
176 PENN -0.2%
177 ARIZ -0.1%
178 WRIG 0.0%
179 OHST 0.0%
180 NCST 0.2%
181 UMKC 0.2%
182 ILL 0.3%
183 JAX 0.3%
184 WSU 0.3%
185 OU 0.4%
186 MIO 0.4%
187 WIU 0.4%
188 TAM 0.4%
189 FIU 0.5%
190 OSU 0.5%
191 MEM 0.5%
192 PRES 0.6%
193 KSU 0.7%
194 UTM 0.7%
195 KU 0.8%
196 ILST 0.9%
197 PAC 0.9%
198 MAN 1.0%
199 TNT 1.0%
200 UNCW 1.0%
201 GU 1.0%
202 HAMP 1.1%
203 CWM 1.1%
204 BELM 1.1%
205 LIUB 1.1%
206 UCLA 1.1%
207 SMU 1.2%
208 WISC 1.2%
209 FRES 1.2%
210 NW 1.4%
211 HOU 1.4%
212 CENT 1.5%
213 MVSU 1.5%
214 UNH 1.6%
215 CONN 1.6%
216 TUL 1.7%
217 UNCA 1.7%
218 TOWS 1.7%
219 USM 1.8%
220 DART 1.9%
221 SF 1.9%
222 FLA 1.9%
223 COLU 1.9%
224 MAR 1.9%
225 NEV 1.9%
226 ISU 2.0%
227 PROV 2.1%
228 NWSU 2.2%
229 BAY 2.2%
230 COLG 2.2%
231 DRKE 2.2%
232 WMU 2.2%
233 GSU 2.3%
234 GWB 2.3%
235 CMU 2.4%
236 COPP 2.4%
237 SFPA 2.4%
238 UALR 2.5%
239 CSF 2.5%
240 MINN 2.5%
241 WEB 2.5%
242 FAU 2.6%
243 IDST 2.7%
244 RUTG 2.7%
245 UNLV 2.8%
246 TTU 2.9%
247 NCC 2.9%
248 WAG 2.9%
249 HAW 2.9%
250 UK 3.0%
251 DREX 3.0%
252 TOL 3.1%
253 TU 3.1%
254 COLO 3.1%
255 UL 3.1%
256 UMD 3.2%
257 BRY 3.2%
258 POLY 3.3%
259 BRWN 3.3%
260 ULM 3.4%
261 GS 3.4%
262 MISS 3.5%
263 TXST 3.5%
264 UNO 3.5%
265 NU 3.6%
266 STAN 3.7%
267 UAPB 3.7%
268 CHAR 3.7%
269 JMU 3.7%
270 SEA 3.8%
271 SAMF 3.8%
272 PVAM 3.9%
273 TAMC 3.9%
274 FURM 3.9%
275 UAB 4.0%
276 TXSO 4.1%
277 DEL 4.1%
278 SHU 4.2%
279 RICE 4.3%
280 JSU 4.3%
281 EIU 4.4%
282 GRAM 4.4%
283 DEP 4.4%
284 TENN 4.4%
285 MSU 4.5%
286 ARST 4.5%
287 RICH 4.5%
288 ECU 4.6%
289 HOFS 4.6%
290 ND 4.7%
291 SAMH 4.8%
292 UTEP 4.9%
293 EWU 5.1%
294 EKU 5.1%
295 ALB 5.2%
296 NDST 5.2%
297 NTEX 5.2%
298 AKR 5.3%
299 LSU 5.3%
300 MSST 5.4%
301 SAC 5.5%
302 MIZZ 5.5%
303 ORSU 5.5%
304 IPFW 5.5%
305 AUB 5.5%
306 BUCK 5.5%
307 CCSU 5.6%
308 NDAK 5.7%
309 CREI 5.7%
310 HP 5.7%
311 APSU 5.8%
312 UTSA 5.8%
313 UGA 5.8%
314 HC 5.8%
315 SELU 5.9%
316 BU 5.9%
317 LMD 6.0%
318 ULL 6.4%
319 SH 6.6%
320 APST 6.8%
321 USA 6.8%
322 POST 6.8%
323 OHIO 6.8%
324 MCN 6.9%
325 LMU 7.0%
326 TSU 7.0%
327 MRSH 7.1%
328 RMU 7.3%
329 DAV 7.3%
330 ALAM 7.3%
331 JXST 7.7%
332 ALST 7.8%
333 BGSU 8.1%
334 TULS 8.5%
335 WINT 8.8%
336 ARK 8.9%
337 MIA 8.9%
338 CSB 9.0%
339 UCI 9.0%
340 CSN 9.3%
341 BRAD 9.6%
342 WCU 9.9%
343 SFA 10.0%
344 UMBC 10.3%
345 YALE 10.5%
346 HOW 11.4%
347 HARV 13.2%

BracketBuster matchups added

2010-02-01T19:11:00.001-08:00

I've added BracketBusters matchups to the schedules. In the next few days look for an analysis of which teams are most likely to benefit.

Conference Tournament Simulations

2010-01-25T09:46:00.000-08:00

As promised, I've started doing the conference tournament simluations in the separate pages. Basically I seed all of the teams at the end of each simulation and run a simulated tournament for each appropriate conference and then calculate RPI. For a more detailed description of what I do see here:

http://rpiforecast.blogspot.com/2008/12/conference-tournaments.html

To see your team's webpage, there are links near the tops of the overall page and individual team pages to those with conference tournaments included in the simulations.

Here's the overall page which includes conference tournament simulations in the RPI: http://www.rpiforecast.com/ct/index.html

And as an example, here's Duke's page with the tournament included: http://www.rpiforecast.com/teams/ct/Duke.html

SO the difference is the ct folders in the address. You should be able to find your team, however, by going to their page and clicking on the link.

Server Issues Fixed/Conference Tournaments

2010-01-16T13:51:00.000-08:00

Well, the server issues seem to have been resolved, so hopefully no more problems going forward. As soon as I get a chance, I'm going to start adding (as I have done in past seasons) simulated conference tournaments for each of the 10,000 simulated seasons. The big hurdle in implementing that is figuring out all of the different formats and seeding schemes.

Server Issues

2010-01-14T16:20:00.000-08:00

I'm having some serious issues with the hosting service I'm using to the point that It's virtually impossible to update the site. Hopefully that will be resolved soon. I had all of the data ready to update this morning, but was not able to do so. I did end up getting the main page and the conference pages up, but that took quite some doing. Some of the pages appear to be a few days old. That's because they somehow defaulted to a recent backup.

RPI broken down by contributions by individual team

2009-12-29T14:02:00.000-08:00

I've added another feature to the individual team pages on the site. I've broken down a team's RPI by the contribution made by each team (including the team itself, its opponents, and its opponents' opponents). Remember that the RPI is defined as follows:

0.25*(Adjusted Winning Percentage) + 0.5* Opponents' winning percentages + 0.25 * Opponents' Opponents' winning percentages

This is then sorted and a rank given.

This can be broken down by team if the appropriate weights are used and you can get an idea of how big of an influence an individual team has on any given team's RPI. I've included current weights and end-of-season weights along with current winning percentages and predicted end-of-season winning percentages and an index to gauge the overall influence of a team on the rpi. The winning percentage of the team itself is adjusted according to home/road/neutral games.

For example, I'll choose Utah State and pick the 12 teams with the biggest (current) weights in its RPI:

Team	Curr Wght	fut wght	curr w/l	exp w/l
Utah St.	27.73	26.45	68.63	73.09
Weber St.	8.59	3.67	55.56	58.85
Cal St. Fullerton	4.56	2.01	37.5	38.33
Morehead St.	4.41	1.82	44.44	54.54
Brigham Young	4.37	2.06	100	89.66
Utah	4.37	1.94	36.36	34.61
Cal St. Bakersfield	4.3	2.07	10	17.07
Southern Utah	4.2	1.85	22.22	28.56
Idaho St.	4.14	2	10	30.88
St. Mary's	4.04	1.94	81.82	83.61
Utah Valley	4.02	1.85	28.57	51.66
Northeastern	4.01	1.75	22.22	45.55

Not surprisingly, the biggest influence on a team's RPI comes from the (adjusted) winning percentage of the team itself. Obviously opponents with higher winning percentages help increase an RPI, but this may be diminished if the weight on a team is very small. To get an idea of which teams help or hurt Utah State's RPI the most, we can look at how far above (or below) 50% a team's winning percentage is and multiply that by the weight*. Doing this for Utah State gives us a "net help index". The idea is that anything less than 100% in terms of winning percentage hurts the rpi and on the other hand, things could be very bad so anything greater than 0% winning percentage helps. taking the difference (or net) between this two is like finding the differnce between winning percentage and 50%

Calculating this for Utah State gives the following

*we'll actually multiply by 2* the weight so that this could in theory range between -100% and 100%. In other words, a team with zero winning percentage but a weight of 100% (I know, not possible) would have a net help index of -100%. In other words, it brings what could potentially be an RPI of 100% down to zero.

Sorting by Help Index:

rank	Team	curr wght	fut wgt	curr W/L	exp W/L	curr help index	exp help index
1	Utah St.	27.73	26.45	68.63	73.09	10.33	12.21
2	Brigham Young	4.37	2.06	100	89.66	4.37	1.63
3	St. Mary's	4.04	1.94	81.82	83.61	2.57	1.30
4	Weber St.	8.59	3.67	55.56	58.85	0.96	0.65
5	Kentucky	0.38	0.06	100	83.76	0.38	0.04
6	Nevada Las Vegas	0.54	0.33	84.62	72.04	0.37	0.15
7	Notre Dame	0.37	0.06	84.62	60.27	0.26	0.01
8	Texas	0.19	0.03	100	91.66	0.19	0.02
9	West Virginia	0.19	0.03	100	82.52	0.19	0.02
10	San Diego St.	0.37	0.23	72.73	70.28	0.17	0.09

What's surprising about this sort is that many of the teams that seem to be helping USU's current RPI are not opponents, but opponents of opponents with high winning percentages. However, these are all expected to drop by the end of the season.

By the end of the season, the top ten should look something like this:

Sorting by Expected (end-of-season) Help Index:

rank	Team	curr wght	fut wgt	curr W/L	exp W/L	curr help index	exp help index
1	Utah St.	27.73	26.45	68.63	73.09	10.33	12.21
2	Louisiana Tech	0	4.3	84.62	71.76	0.00	1.87
3	Brigham Young	4.37	2.06	100	89.66	4.37	1.63
4	St. Mary's	4.04	1.94	81.82	83.61	2.57	1.30
5	Nevada	0.14	4.32	58.33	64.41	0.02	1.25
6	Weber St.	8.59	3.67	55.56	58.85	0.96	0.65
7	Boise St.	0.17	4.36	66.67	55.3	0.06	0.46
8	Fresno St.	0.14	4.36	46.15	54.95	-0.01	0.43
9	Long Beach St.	3.85	1.79	33.33	58.91	-1.28	0.32
10	New Mexico	0	0.36	92.31	82.14	0.00	0.23

Here's a look at the ten teams that hurt USU's RPI the most in terms of current help index:

rank	Team	curr wght	fut wgt	curr W/L	exp W/L	curr help index	exp help index
1	Cal St. Bakersfield	4.3	2.07	10	17.07	-3.44	-1.36
2	Idaho St.	4.14	2	10	30.88	-3.31	-0.76
3	Southern Utah	4.2	1.85	22.22	28.56	-2.33	-0.79
4	Northeastern	4.01	1.75	22.22	45.55	-2.23	-0.16
5	Utah Valley	4.02	1.85	28.57	51.66	-1.72	0.06
6	Long Beach St.	3.85	1.79	33.33	58.91	-1.28	0.32
7	Utah	4.37	1.94	36.36	34.61	-1.19	-0.60
8	Cal St. Fullerton	4.56	2.01	37.5	38.33	-1.14	-0.47
9	Morehead St.	4.41	1.82	44.44	54.54	-0.49	0.17
10	Southern	0.3	0.06	10	24.86	-0.24	-0.03

No big surprises. However, some of these ought to reverse by the end of the season.
Sorted by lowest expected (end-of-season) help index:

rank	Team	curr wght	fut wgt	curr W/L	exp W/L	curr help index	exp help index
1	Cal St. Bakersfield	4.3	2.07	10	17.07	-3.44	-1.36
2	Southern Utah	4.2	1.85	22.22	28.56	-2.33	-0.79
3	New Mexico St.	0.37	4.36	50	41.08	0.00	-0.78
4	Idaho St.	4.14	2	10	30.88	-3.31	-0.76
5	Utah	4.37	1.94	36.36	34.61	-1.19	-0.60
6	Hawaii	0.49	4.38	50	43.35	0.00	-0.58
7	Cal St. Fullerton	4.56	2.01	37.5	38.33	-1.14	-0.47
8	Eastern Washington	0.14	0.41	16.67	17.16	-0.09	-0.27
9	Northeastern	4.01	1.75	22.22	45.55	-2.23	-0.16
10	Pepperdine	0.35	0.24	25	25.42	-0.18	-0.12

To see these stats, go to any team's page and scroll down past the graphs.

ESPN BracketBusters projections

2009-12-12T14:43:00.000-08:00

As I have done in previous seasons, I've started making projections of what the records and RPIs for all 98 bracket buster teams will be on February 1, which is the day that they are set to announce the matchups. I've broken down by road and home teams. I'll probably update them once or twice a week - maybe more frequently. They can be found here: http://www.rpiforecast.com/bb.html

Bracket Projections

2009-12-03T13:53:00.000-08:00

I've started making projections of the field of 65 for the tournament. This is purely a statistical exercise and involves no "gut feeling", however, it is usually more accurate early on in the season than those that DO rely solely on gut feeling (Joe Lunardi, I'm looking at you ;-) ). Anyway, have a look. I use the "Dance Card" methodology along with the simulations to predict who will be in the field. Look through the blog archives to see how I've done in the past. I do not make any projections about seeds, just about who makes the cut.

If you see a team that is in the field that isn't eligible for post-season play, let me know. Here are the teams that I'm giving home court advantage to in the conference tournaments: Mercer (ASun), Siena (Metro Atlantic), UNLV (MWC), and Nevada (WAC). The same is true for those conferences that are held at the higher seed.

Here's the bracket:
http://www.rpiforecast.com/bracket.html

Here's the "Dance Card" site for an explanation of the dance card methodology:
http://www.unf.edu/~jcoleman/dance.htm

New Season!

2009-11-10T20:59:00.000-08:00

With 4 new teams and 1 new conference, the 2009-2010 NCAA Basketball conference is underway and I have started making predictions. At some point, I'll add conference tournaments into the forecasts like last season and will probably update them daily.

Daily Updates This Week

2009-02-01T14:21:00.001-08:00

I'm going to be traveling much of this week and will not be able to update the site very early in the day starting on Wednesday, but will try to get the updates in when I can.

Podcast of interview

2009-01-28T10:08:00.000-08:00

You can hear a podcast of my interview on the radio at this website:

http://usuaggies.com/?p=1008

Media Alert

2009-01-27T08:33:00.000-08:00

I will be appearing on KVNU's radio program "Full Court Press" at 610 on the AM dial tonight at 6:00pm Mountain time zone. Those of you that aren't in Northern Utah/Southern Idaho can listen online at http://www.610kvnu.com . Fair warning: I will be focusing on Utah State's RPI/BracketBuster outlook, etc. I think they will be taking phone calls.

Today's update

2009-01-23T08:23:00.001-08:00

Today's update will be a little later in the day than usual.

Update on Friday

2009-01-15T07:22:00.001-08:00

FYI, Friday's numbers may be posted a little late in the day.

New Features

2009-01-08T14:39:00.000-08:00

I have added two features in the past few days:

1. Bracket Busters RPI Projections:

For all 102 Bracket Busters teams, I've started projecting what their RPIs will be on February 1st. That is the day before the first matchups will be announced. This should give you some idea of where teams will likely stand at that point. The address is: http://www.rpiforecast.com/bb.html (bb = bracket busters)

2. Protrade

For all of the NCAA basketball teams on Protrade.com, I have started projecting their end of season fantasy earnings. This involves doing something similar to what I do in the conference tournament simulations, but I include NCAA tournament and NIT games. There is an explanation of the methodology at the website which is here: http://www.rpiforecast.com/protrade.html

Each of these two features will be updated at least weekly, and probably more frequently than that.

Conference tournaments

2008-12-21T22:42:00.000-08:00

I've started making (at least) weekly forecasts which include simulated conference tournaments. It is always a pain to go through all 30 conference tournament formats each year and figure out how they have changed since last year. If you see any problems, let me know. Here's what I do:

After each of the 10,000 simulated seasons:

1. Figure out seeds based on records
2. Run a simulated tournament
3. Calculate RPI for that one simulation

I repeat this for all 10,000 simulations. For each team, I report RPI broken down by different end-of-season records which will include every scenario that happened in the 10,000 simulations. To give you an idea of what I'm talking about, here is the page for Utah State of the WAC:

http://www.rpiforecast.com/teams/ct/Utah%20St..html

Through Games of Dec 20, the WAC conference tournament looks like it will help USU on average by a few spots. Without taking into account the tournament, their Expected RPI is about 46. Once the simulated tournament is added, it becomes about 43 - a slight improvement. For some teams, the conference tourneys will be a huge boost. This is especially true for teams which have byes until later rounds. One of the problems with conference tournaments (at least in terms of RPIs) is that teams with high seeds are paired up with low seeded teams until the final rounds of the tournament. This will tend to hurt their RPIs even if they win. If I were on the chairman of a mid-major conference, given the choice, I would choose a format with byes for the top seeds.

Thanks Ken.

2008-12-08T09:25:00.000-08:00

As some of you know, Ken Pomeroy has stopped publishing the day-to-day RPI. I just came across this newspaper article about the topic and noticed this bit:

Q: Where should fans go now?
A: There's a Web site called
RPIforecast.com; what the guy does there is he projects the end-of-season RPI.
That's probably more useful.

Thanks for the plug, Ken. Ken has always been more than helpful. I think we both agree that the RPI is not the best measure of a team's quality. My philosophy is that as long as the NCAA is going to use it, why not try to predict what it will be come tourney time.

New Season

2008-11-11T12:41:00.000-08:00

With the new season underway, I've started making projections again. Keep in mind that because it is very early in the season, the probabilities (which are from Sagarin's PREDICTOR) are partly based on past seasons. Within a few weeks this will no longer be the case. This season, updates might be more sporadic than last year, especially during January and February, but I should still be able to update every day.

Conference Tourney Upsets

2008-03-10T13:23:00.000-07:00

Bubble teams (and their fans) all over the country will be pulling for favorites to win the conference tournaments this week, especially when the favorite would otherwise likely get an at-large bid. We've already seen VCU lose in the CAA tournament. They are now definitely on the bubble with about a 40% chance of an at-large bid. They probably didn't burst anybody's bubble but their own.

The tournament that may very well burst a bubble is the MAC tourney. Kent State has a very strong case for an at-large bid even with a loss in the MAC tournament. I'd say they have about a 75% chance of an a bid if they don't get the automatic bid. No other MAC team has much of a case. However, Kent State only has a 25% chance of winning the tournament. In fact, Akron is slightly more likely to win based on Sagarin's PREDICTOR:

Team (Seed) Prob(Auto Bid)
Kent St. (1) 25.8%
Western Michigan (2) 18.0%
Akron (3) 26.7%
Ohio (4) 11.7%
Miami OH (5) 15.3%
Central Michigan (6) 1.3%
Eastern Michigan (7) 0.6%
Toledo (8) 0.3%
Bowling Green (9) 0.2%
Ball St. (10) 0.1%
Northern Illinois (11) 0.0%
Buffalo (12) 0.0%

So, there is basically a 75% chance that the MAC tournament will burst somebody's bubble (other than Kent State's). In no other conference tournament is it more likely than not that a team with no at-large bid chance will steal the automatic bid from a team that's a near lock for an at-large.

Updates

2008-03-04T11:26:00.000-08:00

Today marks the first day of conference tournament play. The basic RPI Forecast doesn't include conference tournaments. For that reason, I added the forecasts that do include them which are updated a couple of times per week: http://www.rpiforecast.com/ct/index.html

Now that conference tournament season is upon us, I'll start updating that page more frequently (probably every day). The main page will only include conference tournament games that have already been played. If you want to see all of the possible outcomes along with the expected RPI for each scenario including conference touranments, go to the above link and click on your team. Next year, the forecasts which include conference tournaments will become the standard forecasts.

As explained in detail below, the seedings for conference tournaments are determined for each of the 10,000 simulated regular seasons, and a simulated tournament is run for each of the 10,000 simulations. By now, many of the conference tournament seeds are already set in stone. If a team does as expected in the conference tournament, their end of regular season RPI will typically be very close to their end of conference tournament RPI.

Updates

2008-02-26T12:16:00.001-08:00

Updates were slow for a few days because I'm on vacation in Miami and am without regular internet access. That should only last a few more days.

Conference Tournaments

2008-02-12T12:17:00.000-08:00

I have been updating the pages that include confernce tournaments a couple of times per week (not necessarily on Mondays). One thing to remember for seeding purposes, is that when two or more teams are tied, I'm randomly assigning the higher seed to one of them. To incorporate all of the tiebreaker rules for all of the conferences would be mind-numbing and will probably not affect the results much at all.

Team Pages added for Conf Tourney Forecasts

2008-01-21T18:45:00.000-08:00

I decided to go ahead and add the team pages that include the conference tournament forecasts in the simulations (as described in the previous post). I may add more info for next week. Just click on a team's name on the page:

http://www.rpiforecast.com/ct/index.html

Projections INCLUDING Conference Tournaments

2008-01-21T15:53:00.000-08:00

There have been many requests to incorporate the conference tournaments into the simulations. I have started doing that on a weekly basis on a separate page:

http://www.rpiforecast.com/ct/index.html

I will update this on Mondays to start. We'll see how that goes.

Here's the procedure:
1. Run the simulations as usual.
2. At the end of each simulated regular season, sort teams in each conference (or division) based on its record.
3. From the conference records in #2, seed the teams for the conference tournaments
4. If two or more teams are tied, the seeds are randomly picked between the tied teams (Learning and incorporating all of the complex tiebreakers for every conference is just too time consuming)
5. For all 30 conferences that hold tournaments, run a simulated tournament based on the specific structure and home court advantages of each tournament.
6. Calculate RPIs, SOS, and the other usual stats for the entire season including the tournaments.
7. Run the simulations 10,000 times

Anyway, eventually I'll add individual team pages (maybe next week) that include the conf tournament forecasts. Items included will be things like probability of winning tournament, expected number of wins in the tournament, etc.

BracketBuster update

2008-01-16T15:32:00.000-08:00

I updated the Bracketbuster RPI forecasts. Remember that these are forecasts of what the RPIs, records, etc. will be on Sunday, Feb. 3, which is the day that the big wigs get together to decide the initial matchups in the Bracket Buster games. As usual, the W-L forecasts are rounded to the nearest digit since they are expectations. If you want to figure out exactly what they are, simply add up the probabilities of winning over the remaining games to get the expected number of wins. The expected numer of losses is then easy to infer. Here's the page:

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

BracketBusters

2008-01-02T18:27:00.001-08:00

For those of you who are interested in the ESPN BracketBuster matchups, I performed a little exercise. The beauty of the RPI Forecast methodology is that not only can you forecast the RPI at the end of the season, you can predict it for any future date.

With that in mind, I decided to provide a forecast of what it will look like on February 3, 2008. That is the date that the powers that be will get together to decide the initial parings for the Bracket Buster matchups. I'm fairly certain that the RPI is only a very small factor in those decisions, with money (TV ad revenues) being the largest. Nevertheless, records and RPIs are likely to have some influence on the decision.

Anyhow, here is the page with the teams sorted by RPI Forecast (for February 3) and separated into the 50 home teams and 50 road teams so that you can make your own predictions. Enjoy:

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

I will NOT be updating that particular page daily. If I make any futher updates, I'll make it known.

EDIT: The page temporarily reverted to one that I had from last season. It has now been fixed.

New Stuff

2007-12-30T18:03:00.001-08:00

I added a couple of features to RPIForecast.com:

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.
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?

2007-12-25T11:51:00.000-08:00

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?

2007-12-19T12:40:00.000-08:00

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

2007-12-19T09:20:00.000-08:00

I added a couple of things to the conferences page:

Current (day-to-day) RPI
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

2007-12-16T09:52:00.000-08:00

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?

2007-12-10T08:44:00.001-08:00

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.

2007-12-04T10:44:00.000-08:00

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

2007-11-30T07:48:00.000-08:00

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

2007-11-21T19:19:00.001-08:00

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

2007-11-11T10:43:00.000-08:00

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:

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

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

2007-11-11T07:26:00.000-08:00

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