GW Posted March 17, 2013 Share Posted March 17, 2013 Wiz- could you provide the spreads for New Mexico State, Oregon, Ok state and Louisville? If it wouldn't be too much trouble, could you provide the post duquesne SLU A+ rated numbers as well? Many thanks. Quote Link to comment Share on other sites More sharing options...
bk18 Posted March 18, 2013 Share Posted March 18, 2013 I would like to see the Wiz's program's bracket Quote Link to comment Share on other sites More sharing options...
The Wiz Posted March 18, 2013 Share Posted March 18, 2013 Wiz- could you provide the spreads for New Mexico State, Oregon, Ok state and Louisville? If it wouldn't be too much trouble, could you provide the post duquesne SLU A+ rated numbers as well? Many thanks. I have this on the Bills prediction thread.....But here it is again Bills by 10 over NMSU Ok St by 4 over Ore Lville play-in game shows NC A&T beating Liberty. When I put in NC A&T vs Lville it wanted to know... ..how many points I wanted... and then said is 26 enough. Win probability=99.9% Quote Link to comment Share on other sites More sharing options...
GW Posted March 18, 2013 Author Share Posted March 18, 2013 I'm sorry that my request wasn't clear- I was actually interested in SLU's numbers against a potential future matchup of SLU vs Oregon, SLU vs Ok state, SLU vs Louisville. I'm especially interested in what the spreads would be if you calculated them using SLU's post duquesne A+ rating. I'm not trying to bash you or your methods, but post duquesne you have underestimated SLU's results in 14 games, while overestimating only the Xavier game. My belief is that SLU has become the post duquesne A+ rated team that they were meant to be all along had it not been for the early season Majerus/injury situations. Including those early season results into your full season numbers is causing you to consistently underestimate SLU's spread- not usually by a point or two, but by an average of about 9 points a game. Again, not criticizing- I hope you underestimate every SLU game going forward by 9 points as that would mean we are hoisting another trophy. Still, I think it would be valuable to see what the numbers are of SLU post duquesne as I believe this will provide a more accurate projection going forward. While it is understood that your system is self adjusting- as you miss, the automatic reajustment process will get you closer to the "true value"( and thus you have been missing by a smaller amount lately)- still with the end of the season upon us there are not enough future results left to make the early season results immaterial. Thanks. Quote Link to comment Share on other sites More sharing options...
The Wiz Posted March 19, 2013 Share Posted March 19, 2013 I'm sorry that my request wasn't clear- I was actually interested in SLU's numbers against a potential future matchup of SLU vs Oregon, SLU vs Ok state, SLU vs Louisville. I'm especially interested in what the spreads would be if you calculated them using SLU's post duquesne A+ rating. I'm not trying to bash you or your methods, but post duquesne you have underestimated SLU's results in 14 games, while overestimating only the Xavier game. My belief is that SLU has become the post duquesne A+ rated team that they were meant to be all along had it not been for the early season Majerus/injury situations. Including those early season results into your full season numbers is causing you to consistently underestimate SLU's spread- not usually by a point or two, but by an average of about 9 points a game. Again, not criticizing- I hope you underestimate every SLU game going forward by 9 points as that would mean we are hoisting another trophy. Still, I think it would be valuable to see what the numbers are of SLU post duquesne as I believe this will provide a more accurate projection going forward. While it is understood that your system is self adjusting- as you miss, the automatic reajustment process will get you closer to the "true value"( and thus you have been missing by a smaller amount lately)- still with the end of the season upon us there are not enough future results left to make the early season results immaterial. Thanks. Got it.... Bills over Ore by 5 1/2 Bills over Ok St by 1 1/2 Lville by 7 over Bills Of course these numbers will change a little as the tourney progresses. As far as spreads since Duq and the methodology used..... It is interesting that the Duq game is the turning point of the season. It was also the halfway point too. So a good place to compare data. Starting with the St. B game it was like we were starting from scratch...data wise ...as if it were Nov again...like a new year.....and maybe it was. First without using point spreads....just win/loss record... I finished 27-3-3 (3 games even) First part of the season,13-2-2 ...second part 14-1-1 With spreads....average miss was 9.3 pts....1st half 9.8....2nd half ...8.8pts A few games skewed the average...so I separated them out...those that were more than 10 pts away from prediction 1st half 11 close (6.36 pts ave) ...6 far....2nd half 11 close (5.18 pts) ...5 far Overall 22 close (5.7 pts)...11 far So what does that mean?....When I picked a winner I was correct 90% of the time....Also 2/3 of the time I missed the exact spread by 5-6 pts. Also 2nd half of season better than first ....more data. Not bad for free info and definitely better than chance. One last thing....you are correct about me lagging on the spreads once the Bills "exploded" after Duq. One of the things I have been tinkering with is adding a trending factor which weighs recent games more heavily. This will reflect not only when a team gels, but injuries, transfers....anything that suddenly changes the numbers. I started using it during the A-10 tourney. (Char game) Although a small sample of 3 games...the ave was 5 pts off. We will see how it does. And finally for those that would like to see what the whole season looked like........ Team.................actual spread...........predicted.........pts away USC...................17..............................12.....................5 San Cl.................12..............................18....................30......wrong outcome TAMU..................21...............................E.....................21 KU to win..............14...............................8.....................6 SIU.........................10..............................19...................9 Wash......................5.................................E...................5 Valpo......................13...............................5...................8 N Tex......................4................................14..................10 UT Mart...................22..............................25...................3 EIU..........................27..............................21...................6 Loy..........................21..............................12...................9 SIU-E.......................27.............................21...................6 New Mex.................14..............................3 1/2...............10 1/2 Sav..........................8.................................18..................10 UMass......................8................................12...................4 Temp to win..............10..............................2.....................8 RI...............................2...............................14...................16 wrong outcome Duq............................9................................9.....................0 St. B...........................10..............................1.....................9 But..............................17..............................2.....................15 Day..............................29.............................5......................24 Ford.............................17............................12......................5 Rich.............................10.............................2......................8 Char.............................18.............................11 1/2..............6 1/2 VCU..............................14.............................2......................12 But.................................4..............................E......................4 St. J................................17............................9 1/2...............7 1/2 GW.................................8..............................6.......................2 X.....................................11............................4......................15 wrong outcome LaS.................................24............................6.......................18 Char................................17...........................12......................5 But...................................11...........................4........................7 VCU.................................6.............................3.......................3 Quote Link to comment Share on other sites More sharing options...
68Alum Posted March 19, 2013 Share Posted March 19, 2013 It's late and I'm just thinking out loud here, but I've been looking ahead to Louisville. No secret here, but the Bills have been playing at a remarkably high level since late January. ESPN does a nice job of rating game performance in their BPI Ranking and the only team that has been consistently at our level over the recent period is, not surprisingly, Louisville. http://espn.go.com/mens-college-basketball/bpi/_/teamId/139 Louisville ranks #2 behind VCU in turnovers caused per game, and we're about 1.5 turnovers lower than that. However, we rank slightly better in points per shot taken and Louisville is a highly unpredictable 3 point shooting team as shown in the Syracuse game. The biggest gap is in rebounds per game at almost 5. Rick Pitino always has a bench loaded with leapers. On 5 days rest and if we can keep everyone's minutes to about 30 or less like Pitino does, we should be able to stand up to the pressure D with our 3 guards. The big challenge is to figure out how to stop their 5 from getting that second and third shot on the boards. Maybe we need to show them a box and one or a 1 2 2 at times to get them off balance. Quote Link to comment Share on other sites More sharing options...
GW Posted March 19, 2013 Author Share Posted March 19, 2013 Wiz- Thank you for the reply. You stated "As far as spreads since Duq and the methodology used..... It is interesting that the Duq game is the turning point of the season. It was also the halfway point too. So a good place to compare data. Starting with the St. B game it was like we were starting from scratch...data wise ...as if it were Nov again...like a new year.....and maybe it was." This was the point I was trying to make. That the post duq SLU team is a completely different entity than the first half team. That is the explanation on why you have consistenly underestimated post duq point spreads (while still correctly predicting the winner...)- by including first half numbers your projections are being weighted down. I am not singling you out- Vegas or anyone else with a predictive method probably didn't see this turnaround coming either. Why not just jettison that first half season anchor? Considering the first half season circumstances I am very comfortable with giving them an asterisk or "incomplete" on those results. Look at it this way- How would the SLU team as it currently stands fare against the november team if they were able to play each other? I'm sorry, but the November team gets soundly beaten and its no fault of theirs... As someone who values the input you provide, I was requesting an inclusion of the post duq point spreads as it appears that these would provide a more fair and accurate number. Providing both the full season and post duq spreads would be outstanding as this would give both full season and "playing to their potential" perspective. Thanks again for all that you do. Quote Link to comment Share on other sites More sharing options...
The Wiz Posted March 19, 2013 Share Posted March 19, 2013 It's late and I'm just thinking out loud here, but I've been looking ahead to Louisville. No secret here, but the Bills have been playing at a remarkably high level since late January. ESPN does a nice job of rating game performance in their BPI Ranking and the only team that has been consistently at our level over the recent period is, not surprisingly, Louisville. http://espn.go.com/mens-college-basketball/bpi/_/teamId/139 Louisville ranks #2 behind VCU in turnovers caused per game, and we're about 1.5 turnovers lower than that. However, we rank slightly better in points per shot taken and Louisville is a highly unpredictable 3 point shooting team as shown in the Syracuse game. The biggest gap is in rebounds per game at almost 5. Rick Pitino always has a bench loaded with leapers. On 5 days rest and if we can keep everyone's minutes to about 30 or less like Pitino does, we should be able to stand up to the pressure D with our 3 guards. The big challenge is to figure out how to stop their 5 from getting that second and third shot on the boards. Maybe we need to show them a box and one or a 1 2 2 at times to get them off balance. It's late and I'm just thinking out loud here, but I've been looking ahead to Louisville. No secret here, but the Bills have been playing at a remarkably high level since late January. ESPN does a nice job of rating game performance in their BPI Ranking and the only team that has been consistently at our level over the recent period is, not surprisingly, Louisville. http://espn.go.com/mens-college-basketball/bpi/_/teamId/139 Louisville ranks #2 behind VCU in turnovers caused per game, and we're about 1.5 turnovers lower than that. However, we rank slightly better in points per shot taken and Louisville is a highly unpredictable 3 point shooting team as shown in the Syracuse game. The biggest gap is in rebounds per game at almost 5. Rick Pitino always has a bench loaded with leapers. On 5 days rest and if we can keep everyone's minutes to about 30 or less like Pitino does, we should be able to stand up to the pressure D with our 3 guards. The big challenge is to figure out how to stop their 5 from getting that second and third shot on the boards. Maybe we need to show them a box and one or a 1 2 2 at times to get them off balance. I have Louisville as thr best team in the country. I have seen them play a number of times. There is something the numbers don't transmit . ..the teamwork and level of play combined with a high level of athleticism. In many ways they are like us but at a higher level. They, like us, have had a "gel moment"......a point where everything came together. The difference is ours happened like magic and was pleasantly surprising. The Cards moment was by design. Pitino uses the regular season like spring training. Feb is like the regular season and March is l the playoffs. Their goal is to get tuned up for the playoffs and they have done that and are ready to go. Does that mean they win the Nat Championship....No but they have the best chance. They have periods during games where they go cold or let up...but they have many ways to beat you. Being a 7 point underdog means if they have an off night and we play great we can beat them. They are much better than the Mich St team we played last year and the Kansas team we played this year. The fact that the point spread is only 7 now is a testament to how far and how fast we have come along. We are one of the best teams in the nation. Louisville is the best. Bad draw in the bracket. One thing I will miss is the RM factor. You know he would have had a plan. It truly would have been an epic coaching battle. But for now, to quote an earler post someone made.....we have a chance. I think the best plan right now is to fully focus on NMSU. We should be able to beat them but they are not Fordham...they are more like Richmond in the rankings. If we start looking ahead to Louisville or OK St we will be in trouble. One step at a time. Go Bills Quote Link to comment Share on other sites More sharing options...
NextYearBill Posted March 19, 2013 Share Posted March 19, 2013 in the back of my mind i can see CSU or Mizzou doing us a huge favor Quote Link to comment Share on other sites More sharing options...
The Wiz Posted March 19, 2013 Share Posted March 19, 2013 I'm sorry that my request wasn't clear- I was actually interested in SLU's numbers against a potential future matchup of SLU vs Oregon, SLU vs Ok state, SLU vs Louisville. I'm especially interested in what the spreads would be if you calculated them using SLU's post duquesne A+ rating. I'm not trying to bash you or your methods, but post duquesne you have underestimated SLU's results in 14 games, while overestimating only the Xavier game. My belief is that SLU has become the post duquesne A+ rated team that they were meant to be all along had it not been for the early season Majerus/injury situations. Including those early season results into your full season numbers is causing you to consistently underestimate SLU's spread- not usually by a point or two, but by an average of about 9 points a game. Again, not criticizing- I hope you underestimate every SLU game going forward by 9 points as that would mean we are hoisting another trophy. Still, I think it would be valuable to see what the numbers are of SLU post duquesne as I believe this will provide a more accurate projection going forward. While it is understood that your system is self adjusting- as you miss, the automatic reajustment process will get you closer to the "true value"( and thus you have been missing by a smaller amount lately)- still with the end of the season upon us there are not enough future results left to make the early season results immaterial. Thanks. It would be nice to jettison the first part of the season. Have you ever tried to tell a professor to get rid of your early semester grades...you just didn't get the course in the beginning but have since gelled. What he might do is to weight your recent test scores and help your grade that way. Just like the pollsters have done. They have weighted our recent performance and made us 13th. If we had started 15-1 we would have been top 5 or 10 team. But alas, I can't throw the early season numbers out any more than the professor can drop the early test scores. As I pointed out earllier, the trending factor I have added during the A-10 tourney should help speed up the self adjusting numbers.....but if we continue to improve...continue to take it to a new level quickly...then you will have to live with under reported winning spreads. I will take that. Quote Link to comment Share on other sites More sharing options...
The Wiz Posted March 19, 2013 Share Posted March 19, 2013 in the back of my mind i can see CSU or Mizzou doing us a huge favor Mizzou has already been crushed once by the Ville in the Bahamas. The Zoo fans will say they are better. So are the Cards. Better root for CSU if you are looking for help. Quote Link to comment Share on other sites More sharing options...
brianstl Posted March 19, 2013 Share Posted March 19, 2013 Mizzou has already been crushed once by the Ville in the Bahamas. The Zoo fans will say they are better. So are the Cards. Better root for CSU if you are looking for help. Mizzou added a weapon since then. Still don't like their chances, but they have the talent to beat them. Quote Link to comment Share on other sites More sharing options...
The Wiz Posted March 19, 2013 Share Posted March 19, 2013 Mizzou added a weapon since then. Still don't like their chances, but they have the talent to beat them. Lville has their best player back....Mizzou took him out early in that game (and for the next 4 weeks) and still lost by 23pts. Lville remembers Mizzou. Better root for CSU if you want help. Of course their is a reason they are 8 and 9 seeds...about equal so I am not sure CSU is much more help but at least a different look. Quote Link to comment Share on other sites More sharing options...
GW Posted March 19, 2013 Author Share Posted March 19, 2013 It would be nice to jettison the first part of the season. Have you ever tried to tell a professor to get rid of your early semester grades...you just didn't get the course in the beginning but have since gelled. What he might do is to weight your recent test scores and help your grade that way. Just like the pollsters have done. They have weighted our recent performance and made us 13th. If we had started 15-1 we would have been top 5 or 10 team. But alas, I can't throw the early season numbers out any more than the professor can drop the early test scores. As I pointed out earllier, the trending factor I have added during the A-10 tourney should help speed up the self adjusting numbers.....but if we continue to improve...continue to take it to a new level quickly...then you will have to live with under reported winning spreads. I will take that. Wiz- It doesn't have to be an either/or proposition- both full season and post duq numbers could be provided. Now if this is too much to ask or requires too much effort, fair enough do not do so. Citing your professor example- if I told my professor that in the fall semester, the father figure of my life battled for his life in a hospital before succumbing in December when I served as his pallbearer, while I also suffered a broken ankle(KM), a seperated shoulder(GG) and a groin pull(JJ)- uh, then ya, I would expect him to have a heart and give me some slack up to and including disregarding my fall semester grades. Now perhaps a professor could not grant that request due to education regulations, however there are no regulations preventing you from doing so.... Quote Link to comment Share on other sites More sharing options...
Quality Is Job 1 Posted March 19, 2013 Share Posted March 19, 2013 One thing I will miss is the RM factor. You know he would have had a plan. It truly would have been an epic coaching battle. Go Bills Yes, but Rick Pitino (then at Kentucky) was always Cockstrong's "kryptonite." Quote Link to comment Share on other sites More sharing options...
majerus mojo Posted March 19, 2013 Share Posted March 19, 2013 in the back of my mind i can see CSU or Mizzou doing us a huge favor I believe those are called "dreams", and I've had a fair share of them myself since Sunday evening. MIZ! Quote Link to comment Share on other sites More sharing options...
The Wiz Posted March 19, 2013 Share Posted March 19, 2013 Wiz- It doesn't have to be an either/or proposition- both full season and post duq numbers could be provided. Now if this is too much to ask or requires too much effort, fair enough do not do so. Citing your professor example- if I told my professor that in the fall semester, the father figure of my life battled for his life in a hospital before succumbing in December when I served as his pallbearer, while I also suffered a broken ankle(KM), a seperated shoulder(GG) and a groin pull(JJ)- uh, then ya, I would expect him to have a heart and give me some slack up to and including disregarding my fall semester grades. Now perhaps a professor could not grant that request due to education regulations, however there are no regulations preventing you from doing so.... Don't think I will be able to do it. I can't just pull out SLU's numbers and start on Jan. 23 (Duq game) . That is because all of the numbers from all the games are interdependent on one another. To pull out just the Bills numbers without changing all the other teams will mess up the data. Now if this was my full time job ...maybe...but it isn't. You would almost have to start the season for all 347 teams on Jan 23 to make it work. It is kind of like Newton's Law of Universal Gravitation....everything pulls or effects everything else in the Universe. So that is why I implemented the trending factor which weighs recent games more heavily when there is a discrepancy in the spread numbers. Superimposing the data isn't a perfect solution but it may speed up the self adjustment factor. Btw...if I was your prof, I would give you the A. Quote Link to comment Share on other sites More sharing options...
GW Posted March 20, 2013 Author Share Posted March 20, 2013 Don't think I will be able to do it. I can't just pull out SLU's numbers and start on Jan. 23 (Duq game) . That is because all of the numbers from all the games are interdependent on one another. To pull out just the Bills numbers without changing all the other teams will mess up the data. Now if this was my full time job ...maybe...but it isn't. You would almost have to start the season for all 347 teams on Jan 23 to make it work. It is kind of like Newton's Law of Universal Gravitation....everything pulls or effects everything else in the Universe. So that is why I implemented the trending factor which weighs recent games more heavily when there is a discrepancy in the spread numbers. Superimposing the data isn't a perfect solution but it may speed up the self adjustment factor. Btw...if I was your prof, I would give you the A. Thank you for the kind words. Not to beat a dead horse, but you had said SLU had an A+ rating post duq. Would it not be possible to base that post duq spread on whatever team that rating was then closest too? In other words, if you have SLU ranked #8 post duq, could you not just give the numbers of whatever team you currently have ranked #8 as an approximation? If that's not viable, no need for you to reply back. Btw... if I didn't find what you do to be valuable, I wouldn't be going through the effort to get these numbers out of you... Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.