http://www.mrisports.com/ en Copyright 2012 Thu, 05 Apr 2012 13:58:14 -0600 http://www.sixapart.com/movabletype/ http://blogs.law.harvard.edu/tech/rss There will be no computer controversy this season; Kentucky is your MRI Tournament champion. Prior to the tip of the NCAA Tournament, North Carolina had snuck into the lead of a race between the Tar Heels, Kentucky and Ohio State. But an injury to Kendall Marshall derailed the boys is Carolina blue, leaving Ohio State and Kentucky to battle it out. Ohio State faltered down the stretch against the Jayhawks which kept the country from seeing a 1-2 matchup for the title game. It would have been close to a toss-up for the title had that occurred. As it was, Kentucky entered that final game as a prohibitive favorite, with a chance to win over 70 percent of the time. Unlike the Buckeyes, the Wildcats worked within the percentages and took home the title, thanks to probably the most impactful player this season, Anthony Davis. Davis didn't score much, but it was his all-around play that helped Kentucky. Those things are better captured in HOOPWAR, so it will have to wait for another post. As for Kentucky, they finished atop the computer also, with a score of 240.82, a full nine points ahead of Ohio State. The Wildcats would be favored in every matchup against every team in the country with that mark. That is domination. The computer showed some domination too. Despite an injury derailing its title pick and a run to another pool title, the MRI predicted 42 of 63 games prior to the start of the Round of 64. The best mark in one pool it entered was 45, and that person had Kentucky winning it all. The winner in its most competitive pool on ESPN only picked 40 games correctly, showing that scoring in a pool can matter a great deal to the finish. (If you ask me, getting more games correct, even though your final pick is wrong should be worth more points. Probably a scoring algorithm that makes that happen somewhere). Going round to round, the computer was correct in 48 of the tournament's 67 game, for a 71.64 percent mark. If you leave out the PIGs, the computer did a tad better -- it really did not do well on that Tuesday and Wednesday -- hitting 47 of the 63 games, for a 74.60 percent mark. For the season, once predictions began after Thanksgiving, the computer got 72.60 percent of all games correct. It is now correct on 71.64 percent of games all-time, a mark based on predictions of 28,529 games. Check out all the rankings in the Tournament Final of the 2011-12 Basketball MRI. http://www.mrisports.com/archives/2012/04/college_basketball_computer_rankings_2011_week_17.html http://www.mrisports.com/archives/2012/04/college_basketball_computer_rankings_2011_week_17.html Basketball Feature Rankings Thu, 05 Apr 2012 13:58:14 -0600 If the MRI were the sole arbiter of the NCAA Tournament, there would be six teams with wholly different reactions to their status today. But the MRI is not, and I would think probably shouldn't be. You can't just use one formula to rule them all.. although the committee does seem to do that with the RPI. There is a ting of bitterness this morning, but not much. Drexel did not make the field, despite a ranking of No. 39 in the final MRI, which I believe is the highest it has achieved since I started publishing this about 12 years ago. The key feature that half of the "snubs" have is that they come from smaller conferences. They didn't win their conference tournament, and therefore lost their "chance" to get into the dance. I can't begrudge teams that did make it like Colorado, when they won their playoff. That is one way in. The Buffs would have been NITing it or worse if they hadn't have beaten Arizona. Both teams were fighting for their tournament lives. But you can take exception with some of the committees choices. Here are the snubs and the list of replacements with the MRI rankings in parentheses. The cutoff in the MRI was No. 56, which was Purdue. Snubs: Drexel (39), Middle Tennessee State (40), Oral Roberts (45), Seton Hall (46), Wagner (54), Stanford (55). Questionable choices: San Diego State (59), Michigan (60), Notre Dame (64), Xavier (65), Colorado State (81), South Florida (105). Of course, the MRI has two questionable choices that probably aren't that way in most people's minds. And I will admit that Michigan and Notre Dame probably have good cases for being in. Both teams are within 9 points of the final spot, which if the game was a home game, would mean they would be favored over Purdue. I would imagine the same can hold true for San Diego State, since they were just a few spots off. On the "snub" side, You can maybe discount Wagner and Stanford, but anyone higher maybe deserved a longer look. As it was, if Iona hadn't gotten into the field, it would have been even a bigger snub, as they were higher ranked than any of these teams. The committee maybe got that one correct. The big issue should be with the Colorado State and South Florida inclusions. Just based on MRI numbers, they don't belong. In the Ken Pomeroy rankings, South Florida is No. 66 and Colorado State is No. 76. Drexel (41) and Middle Tennessee (61) are higher than both. Seton Hall is higher than both. Stanford is higher than both. I won't complain any more. It is a difficult job to make the choices for the last spots. It is made even harder now that the field has expanded and the choice to include three more teams of questionable character has to be made. Either way, we will have a great tournament. That much never changes. Enjoy the rankings in this week's MRI, along with the rankings of every tournament team after the normal update. All of the Top 25 made the tournament. http://www.mrisports.com/archives/2012/03/college_basketball_computer_rankings_2011_week_16.html http://www.mrisports.com/archives/2012/03/college_basketball_computer_rankings_2011_week_16.html Basketball Feature Rankings Mon, 12 Mar 2012 12:09:19 -0600 Hustle Belt, I published some data on player valuation. It is an overly complicated formula, designed to try and come up with a measure similar to Wins Above Replacement for baseball. I promised a breakdown of the formula so here goes. The basic formula is: (Points Saved + Points Earned) * Average Pace/Team Pace - Replacement level / 30. We will take each piece separately to try and explain the methodology ]]> http://www.mrisports.com/archives/2012/03/calculating_player_value_hoopwar.html http://www.mrisports.com/archives/2012/03/calculating_player_value_hoopwar.html Basketball Feature Thu, 08 Mar 2012 22:09:10 -0600 This week's rankings always look a little strange. Over the past seven days, many of the mid-major and low-major conferences have stacked the deck with their conference tournaments. Teams have played -- and in some cases won -- several more games than the high-major teams. The result is a surge by those teams in the rankings. The side effect is that many of them will experience a backslide this week as the top tournaments start. For example, Wichita State, which bowed out in the semifinals of the Missouri Valley tournament, is still ranked No. 6 in the MRI. But Kansas has yet to start its conference tournament, where most likely, it will play two more games this week. Win those and the Shockers are dropping. Other teams whose seasons are over will see the same thing happen to them. Teams like Belmont, Iona and Middle Tennessee State will likely be passed by "bubble" teams making a last push for tournament inclusion this week. As for the Gaels and Blue Raiders, they have a tough week ahead of them. Both teams availed themselves well in the regular season, each winning 25 games overall and their regular season conference crowns. Both teams took difficult losses in their conference tournaments though and now are looking at NIT bids. This is despite high rankings in the MRI that have both teams in the 26-35 range. Getting to that level in the rankings is nothing to sneeze at. You have to have a good season to get that far. The thing is that usually those teams win their conference tournaments, or have made such a regular season case for themselves, they are considered locks for the NCAA tournament. Neither is the case with these two. As much as the numbers are king at the MRI, the committee will not believe that wins by Middle Tennessee State over Belmont, Akron, UCLA and Mississippi will be enough. The same with Iona's scalps of Western Michigan, St. Joe's, Maryland, Nevada, Denver and Vermont. It is a tough sell. It is why tonight's game between Virginia Commonwealth and Drexel will means so much. The loser will likely be in the NIT, although current projections have VCU hanging on to a glimmer of hope should they lose. It is too bad that 25 wins, no matter who those 25 wins come against, means so little. That Murray State's 27 Division 1 wins mean more to the committee just because they were the final team to lose this season is sad. Both the Atlantic Sun and MAAC were better conferences than the OVC. The Racers didn't beat Goliaths. They handled St. Mary's, Memphis, Dayton and Southern Mississippi. Two of those four teams are very good. But the depth and breadth of the resume is no better than what Drexel, VCU, Middle Tennessee and Iona bring to the table. Murray State did the important thing. They removed doubt. They won the Ohio Valley tournament, even though Tennessee State provides them a tough matchup and almost pulled it out. This is the year for the committee to make a statement though. It can state that WINNING is important over all things. Leave Northwestern, or a few of the gagging 10 Big East teams that Joe Lunardi has dancing in his latest projection at home. Give a small school with a decent overall resume and a gaudy 25 wins a shot. You might be surprised how good a team like that actually is, and how well they perform on the big stage. And on to the rankings. http://www.mrisports.com/archives/2012/03/college_basketball_computer_rankings_2011_week_15.html http://www.mrisports.com/archives/2012/03/college_basketball_computer_rankings_2011_week_15.html Basketball Feature Rankings Mon, 05 Mar 2012 12:21:43 -0600 The MRI has had its ups and downs in terms of predicting which teams will make the NCAA basketball tournament. There is a reason for this: it was never designed to do such a thing. That said, it was designed to determine what teams are the best at playing basketball. That doesn't always mean they get selected to play in the big dance. There has been some criticism that the MRI relies too much on the team's record, with the understanding that wins equate to a better team (if you don't lose, you are doing something correct). What is lost by that argument is that the MRI does take into account who you played in order to rack up those wins. It isn't enough just to run through the SWAC. You might get a bulky win number that way, but you might not be ranked highly in the MRI. Instead, you have to do it convincingly because year after year, the SWAC is among the bottom three conferences ranked by this computer (and this year, clearly the worst). That is why Murray State of the one-loss and long unblemished record makes its first appearance at No. 25 in this week's rankings. It took an entire season for the computer to even give credit to the Racers, who play in what the MRI considers the 23rd best conference in the country. The team was always near the rankings, just not in them. They should have gotten an at-large bid to the tournament even if they somehow tripped in the conference tournament. They are good, just not among the top 10 best teams in the country as many would have you believe, namely the AP rankings. That the MRI considers who you play also leads to teams like Florida State, with a 19-9 record getting into the rankings, or a 21-8 Wisconsin team cracking the top 20. You have to play the best to be ranked highly. You have to win some of those games, and you have to be consistent and win convincingly to get high marks. This site has always been of the opinion that the tournament should be more open to the mid-major conferences. For many of these leagues, there are few chances to play up against the teams in the top 50 in the RPI, and therefore, the tournament should allow more bias towards teams in smaller conferences that performed exceptionally well, versus those in bigger conferences that were somewhere in the middle. In some ways, the RPI is an odd measure. There was talk back when the Missouri Valley landed so many teams in the tournament that they had gamed the system. The league did this by doing well in scheduling during the nonconference season (by playing teams that ENDED UP being good, not necessarily ones that were good at the time). They won some of those games, enough that when the league schedule started, every game was bump up in the RPI, as opposed to being dragged down by teams that were shutout early in the season. What people forget is that the big conferences manage to game the system every year. Very few of the teams in the best conferences are going to perform poorly in the nonconference season. There might be a couple of losses here and there, but for the most part, they win. When the league starts, they are constantly getting a boost in the RPI, because no team has such a dreadful record as to be a drag on the overall mark. Yes, those are usually the best teams anyway, but it doesn't mean that a team like Drexel, whose RPI is 67 isn't just as good. I bring up Drexel, not because I went there, and there might be a small amount of bias in how I feel about them. After all, they were the reason the MRI was created: to show that teams in smaller conferences were just as good as the big boys, if you evaluated teams based on actual game performance, not just schedule. Drexel has lost just a single game since the calendar changed to 2012. They have excelled in playing the game of basketball. They are now 25-5 against Division 1 teams, and they will be a tough sell come selection Sunday if they do not keep up the winning streak through the CAA tournament. They have no top 50 wins, and played just one team in that group, Virginia, early in the season, without one of the team's best scorers. But because schedules are set a couple of seasons in advance for the most part, Drexel had games against teams that could have been good, but weren't. They may have been better off, because at less than full strength, they may not have won those game anyway. Still the schedule could look better, if say a St. Joseph's weren't struggling, or one of the MAAC teams they played had turned it on. The Colonial did a bad job up and down the chart of scheduling this season. That is why many believe that between Virginia Commonwealth and Drexel, only the winner of the conference tournament will get in. And if it isn't one of those two teams, the league will not get a second bid. (Drexel at least gets an automatic entry into the NIT, for what that is worth. Barely missing the NCAA tournament has not been a recipe for success in the minor league bracket lately. Call it some sort of hangover effect). It is also why Drexel's margin for error to get to even the brink of consideration has been so small. The Dragons lost just two games in the conference season, but they couldn't have slipped up much more without it becoming a problem. Every game they played was a fight against a big drop in the RPI. A loss could have tanked whatever small chance they currently have. According to the MRI, Drexel is currently No. 44. That is by no means a lock for the tournament, even though this is an odd year. Currently, the cutoff point for an at-large berth would fall all the way down at No. 62, currently held by Wyoming. But a number of current small conference leaders fall within those 62 spots. If they lose and are near the bottom of the list (say Lehigh, or Texas Arlington), the bubble jumps up one more spot, leaving the Cowboys, who at one point were in the MRI top 25, in the dust. The year everyone assumes that Drexel was snubbed for the tournament (with an RPI of 43, again, not putting much stock in the stat), the Dragons were a lowly 71 in the MRI on Selection Sunday. Would I have been happy to see them in after beating Villanova and Syracuse that year? Yes. Did I think they were as badly snubbed as Dick Vitale did? No, not really, because they played well, just not well enough. This season they are playing very well, and yet even though major conference contenders for the last few spots are playing soft, and some scheduling quirks, the Dragons will likely be playing in the NIT should they lose in the next week. These mid-major teams live on the point of a needle, and some of them (not all) need to be given the benefit of the doubt when it comes to whether they are among the top teams in the country, and deserve to be in the bracket. Another school that seems to be on the verge of getting left out, despite its own RPI of 48 is BYU. The Cougars haven't excelled against the top part of their conference (1-3 against St. Mary's and Gonzaga, two likely locks), but they have performed well enough to be in the top 25 of the MRI for most of the year. The Cougars lost to Baylor and Wisconsin, and no one can fault them there. And the Cougars lost to Loyola Marymount, who also tripped up St. Mary's in case you were wondering. They haven't done anything wrong necessarily. And yet, despite what experts claim is a "great" RPI number, BYU is also sweating it out. That very few teams ranked at the end of the season in the MRI have missed the tournament probably doesn't help fans at this point (Missouri State and Akron are the two that come to mind off the top of my head). In the end, my point is that if you are an Oral Roberts, with a 26-5 record, or South Dakota State, with a 24-7 mark, you deserve some consideration, even though only one of those two teams will play in the dance. Both had excellent seasons. Can I argue that they are better than South Florida's season? You bet. The Bulls are 11-5 in the Big East so far, but the only significant win among those 11 is against Seton Hall, who is also fighting for its tournament life. Against the top four teams in the conference standings, South Florida has been shut out. Due to the fabulously bloated league, they don't have another chance against those teams. Should they be rewarded for playing the bottom of the Big East which is not that good this season? Should they be rewarded over Drexel, despite losing to Old Dominion and Virginia Commonwealth in nonconference games, teams that the Dragons swept? (Heck, should they be rewarded over VCU for that matter?). Should they be rewarded just because the RPI puts them at No. 46, even though they couldn't beat Penn State, Auburn or Southern Mississippi? Decide for yourself. Just know that the MRI has the Bulls at No. 103. Here are all the rankings in week 14 of the basketball MRI. http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_14.html http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_14.html Basketball Feature Rankings Mon, 27 Feb 2012 13:30:22 -0600 Rankings through the games of February 19. http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_13.html http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_13.html Basketball Feature Rankings Sun, 26 Feb 2012 21:00:39 -0600 Rankings through Feb. 12. http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_12.html http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_12.html Basketball Feature Rankings Wed, 22 Feb 2012 16:56:17 -0600 Rankings through the games of Feb. 5. http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_11.html http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_11.html Basketball Feature Rankings Fri, 17 Feb 2012 13:22:46 -0600 No rants on Murray State this week, although it is likely the streak will end when the Racers meet St. Mary's in Bracket Busters. One quick note on the made-for-television event. If ESPN could pressure one of the big leagues into participating (and this is not the Mountain West, or Atlantic 10, this is one of the big boys), it would fulfill a much greater purpose. When the mid-majors face the big boys, it is always at the beginning of the season, before the smaller schools have had a chance for their teams to really come together. But if the schools played now, there is a greater chance we could see the "upsets", and see the teams earn that extra win that gets them into the tournament. All anyone is asking for is a shot, and this is the one shot that many of these high-performing teams are asking for. Beating another Missouri Valley team, or a Colonial team isn't as meaningful as beating one of the middle of the pack Big Ten or Big 12 teams, ones that might be the last one in the tournament. Instead, the teams who truly deserve that extra spot could get it, even if it is just for a single round. Enjoy the rankings this week. More in the weeks to come, including the return of the tournament projections. http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_10.html http://www.mrisports.com/archives/2012/02/college_basketball_computer_rankings_2011_week_10.html Basketball Feature Rankings Wed, 01 Feb 2012 15:43:26 -0600 Bracket Project , the Racers are on the six line. It is close to where the MRI would slot them (the rest of the Project does not agree so nicely). The inputs, such as a three-seed from several outlets, are a little nuts though. There is also one site that doesn't even mention Murray State (totally not right). Then again, this is the site that accused me of being nuts for rating George Mason and some other mid-majors high. Murray State is still a good team, and the Racers will likely be in the tournament should they not make it through the OVC tournament. They just aren't among the top 12 teams in the country.]]> http://www.mrisports.com/archives/2012/01/college_basketball_computer_rankings_2011_week_9.html http://www.mrisports.com/archives/2012/01/college_basketball_computer_rankings_2011_week_9.html Basketball Feature Rankings Tue, 24 Jan 2012 14:41:13 -0600 It is late, so no soapbox standing today. But let's just say this: Big Ten (Yes!) Georgetown (No!) Connecticut (No!) Pac-12 (Do they actually still play basketball?) Cue the rankings http://www.mrisports.com/archives/2012/01/college_basketball_computer_rankings_2011_week_8.html http://www.mrisports.com/archives/2012/01/college_basketball_computer_rankings_2011_week_8.html Basketball Feature Rankings Mon, 16 Jan 2012 14:03:55 -0600 finished in the top 10 percent of "America" finishing in the 90th percentile in the ESPN College Bowl Mania. Just goes to show that the computer continues to be smarter than most college football fans. So be careful what you criticize. Check out all the final rankings in the bowl final football MRI for 2011-12.]]> http://www.mrisports.com/archives/2012/01/college_football_computer_rankings_2011_bowl_final.html http://www.mrisports.com/archives/2012/01/college_football_computer_rankings_2011_bowl_final.html Feature Football Top 25 Tue, 10 Jan 2012 12:45:55 -0600 one of these. Wisconsin is very much still a contender. Remember that North Carolina took a few weeks to even appear in the MRI rankings. And any team that reaches the magical 100 point mark in the rankings is almost guaranteed a tournament spot. Many teams go through these funks. Most recover. Wisconsin is one of those teams. There are other teams that people should be questioning right now. Both Connecticut and Georgetown took hard losses this weekend. Both were ranked highly in the AP poll. Neither had been higher than No. 20 in the MRI. The schedules were not overwhelming. They didn't dominate any games. They squeaked their way to their position in the computer, and it was a shaky one at that. So why did they get any love, other than for the name on their jersey, in the human polls? The same question can be asked about Harvard. If not for a win over Florida State, the record would be built on the backs of a number of questionable teams. Add in the loss to Connecticut and they look even shakier. Before that loss they were not even sniffing the top 25 in the computer despite an undefeated record (the same could be said for Murray State right now, who at least makes the ranking bubble list). If there is no Tommy Amaker in Boston, there is no Harvard with a number next to it. It is important to keep in mind as we move deep into the season that the human polls are very flawed. (Note: The computers are flawed too. The MRI is historically only 75 percent correct.) But just because a team is highly ranked, or has a big gaudy record, that doesn't mean that they are excellent/unstoppable/elite eight material. Remember that Connecticut came out of seemingly nowhere last season to win the title. Remember that teams consistently "surprise" come March, because expectations are so skewed by the human polls. You have to use all the data available to make judgements. Wisconsin could still make a big splash come the tournament. Whether that is a second weekend appearance, or a Final Four run remains to be seen and a lot depends on what happens over the next 6 weeks. Don't overreact to numbers you see on January 9. Teams will change. Teams will evolve. And only then will we see who the real contenders for the title are. And maybe it will be Connecticut, Georgetown, Harvard and the Badgers after all. Check out all the rankings in week 7 of the basketball MRI. ]]> http://www.mrisports.com/archives/2012/01/college_basketball_computer_rankings_2011_week_7.html http://www.mrisports.com/archives/2012/01/college_basketball_computer_rankings_2011_week_7.html Basketball Feature Rankings Mon, 09 Jan 2012 20:31:19 -0600 No blabbering this week. Just the rankings. All rankings are based on Division I game results through January 1, 2012. Happy New Year. Happy rankings. Unless you went to Grambling. http://www.mrisports.com/archives/2012/01/college_basketball_computer_rankings_2011_week_6.html http://www.mrisports.com/archives/2012/01/college_basketball_computer_rankings_2011_week_6.html Basketball Feature Rankings Wed, 04 Jan 2012 17:16:52 -0600 For some schools, it just takes time. Everyone knew that North Carolina was good. The Tar Heels showed that in flashes during the early season. But it took until Christmas before they truly found themselves. North Carolina destroyed Nicholls State (which is to be expected) and then dismantled Texas to the tune of 20 points. That is why the Tar Heels have vaulted to No. 4 in the MRI rankings. The things that had been holding North Carolina back -- rebounding and turnovers -- are suddenly a thing of the past. The two losses to UNLV and Kentucky almost feel like a different team was playing. North Carolina is now averaging almost 10 more rebounds a game than its opponents. And it is now just about average in the takeaway department. North Carolina has arrived, and is poised to challenge Syracuse, Ohio State and Wisconsin for the top of the heap. There are still six teams that have not lost a game this season, to go with 11 teams that have not beaten another Division 1 team. Keep an eye on Murray State to see if they can survive as the lowest ranked of the undefeated. And watch Alabama A&M for when they might break through. Although being ranked 309 isn't exactly encouraging. Check out all the rankings in Week 5 of the MRI. http://www.mrisports.com/archives/2011/12/college_basketball_computer_rankings_2011_week_5.html http://www.mrisports.com/archives/2011/12/college_basketball_computer_rankings_2011_week_5.html Basketball Feature Rankings Tue, 27 Dec 2011 12:56:32 -0600