What a great way to finish the first week of college football season. Boise St. let the game slip away then snatched it back at the last minute. Entertaining stuff.
And with the win, Boise St. moves up to #3 in the computers rankings and they are right behind TCU and not that far behind Alabama. As noted earlier, trying to have accurate ratings after a single game is difficult at best but Boise St. certainly didn't hurt their chances with the win. They now need Virginia Tech to play well the rest of the season.
As far as picking games, the computer went 24-13-1 against the spread and 31-7 picking winners versus Vegas going 33-4. Considering the computer simply used last years end of season ratings, that isn't all bad.
With one game done, we can no update the regular season projections and see that both TCU and Boise St. are projected to finish undefeated. That shouldn't be a big surprise given their schedules, particularly Boise St's. They do have a 34% and 36% chance of one loss though so it isn't a done deal by any stretch. Two teams, Alabama and Ohio State are projected to have a single loss and each has a decent chance at going undefeated at 38% and 24% respectively.
Stay tuned for preview of this weekends games and perhaps the first newsletter if I get enough subscribers.
Monday, September 6, 2010
Sunday, September 5, 2010
Preliminary 2010 College Football Week 1 Prediction Performance; 23-11-1 against the spread!
I often times don't really pay attention to it since it is a bit of a game of luck early in the season, but this year I did check the computer's predictions against the spread in week 1 and was pleasantly surprised to see that it went 23-11-1 through yesterday's games.
These picks are done using last years end of season ratings as while not perfect, it is a much better starting point than treating all the teams the same. So going 23-11-1 simply means that those ratings aren't all bad as a starting point.
Full week by week results from last year can be seen here. The computer was at 57% against the spread and a stellar 78% on locks last year.
Stay tuned to this blog and the web-site for upcoming picks, and consider signing up for the newsletter if you want more info including best picks and locks.
These picks are done using last years end of season ratings as while not perfect, it is a much better starting point than treating all the teams the same. So going 23-11-1 simply means that those ratings aren't all bad as a starting point.
Full week by week results from last year can be seen here. The computer was at 57% against the spread and a stellar 78% on locks last year.
Stay tuned to this blog and the web-site for upcoming picks, and consider signing up for the newsletter if you want more info including best picks and locks.
Preliminary 2010 College Football Week 1 Ratings; Alabama staying #1, TCU up to #2
There are still a few more games that will be included in the week 1 ratings so these are not final, but I hate waiting for the last couple games so will discuss what we've seen thus far. Of course, with just a single game played the ratings, and thus rankings, are very volatile and you are going to see teams winning and dropping or losing and rising. At this point, a teams rating is 50% from their single game and 50% from last years rating.
See the full list of ratings on the web-site, but the top-25 are below.
Alabama stays #1 but drops a bunch of ratings points due to playing a very weak opponent. It is also due to them having such a dominating margin over #2 Florida at the end of last year and Florida also dropping. These drops are not uncommon and before fans of those schools panic, these games will carry very little weight in each teams rating as soon as they play a meaningful game so don't worry. However Florida really didn't play well and probably deserves the drop.
TCU takes advantage of Florida dropping to move to #2 and Va Tech and Boise St. are close behind and will decide a lot tomorrow night. Texas drops for similar reasons to Alabama and Florida leaving room for Ohio State, LSU, and Oregon to move up.
The big movers for the week are Michigan getting into the top-40 at #39, Virginia also doing the same at #40, South Carolina getting to #12, and Fresno State jumping to #32.
On the other side, UConn drops all the way to #72, Ole Miss to #57, Cincinnati to #45, and Richmond to #94.
Stay tuned for more!
See the full list of ratings on the web-site, but the top-25 are below.
Alabama stays #1 but drops a bunch of ratings points due to playing a very weak opponent. It is also due to them having such a dominating margin over #2 Florida at the end of last year and Florida also dropping. These drops are not uncommon and before fans of those schools panic, these games will carry very little weight in each teams rating as soon as they play a meaningful game so don't worry. However Florida really didn't play well and probably deserves the drop.
TCU takes advantage of Florida dropping to move to #2 and Va Tech and Boise St. are close behind and will decide a lot tomorrow night. Texas drops for similar reasons to Alabama and Florida leaving room for Ohio State, LSU, and Oregon to move up.
The big movers for the week are Michigan getting into the top-40 at #39, Virginia also doing the same at #40, South Carolina getting to #12, and Fresno State jumping to #32.
On the other side, UConn drops all the way to #72, Ole Miss to #57, Cincinnati to #45, and Richmond to #94.
Stay tuned for more!
| Rank | Team | Rating | Record | Schedule | Change |
|---|---|---|---|---|---|
| Rank | Team | Rating | Record | Schedule | Change |
| 1 | Alabama | 87.980 | 1-0 | 55.082 | +0, -5.267 |
| 2 | TCU | 84.965 | 1-0 | 75.769 | +1, -0.195 |
| 3 | Virginia Tech | 83.722 | 0-0 | 0.000 | +2, +0.000 |
| 4 | Boise St | 83.207 | 0-0 | 0.000 | +2, +0.000 |
| 5 | Ohio State | 82.505 | 1-0 | 60.031 | +3, +1.701 |
| 6 | Florida | 82.331 | 1-0 | 56.284 | -4, -6.188 |
| 7 | LSU | 80.923 | 1-0 | 75.315 | +2, +0.392 |
| 8 | Oregon | 80.496 | 1-0 | 50.214 | +2, +0.202 |
| 9 | Texas | 79.953 | 1-0 | 60.759 | -5, -4.908 |
| 10 | Georgia Tech | 78.546 | 1-0 | 55.041 | +4, -0.401 |
| 11 | Texas Tech | 78.405 | 0-0 | 0.000 | +6, -0.000 |
| 12 | South Carolina | 77.496 | 1-0 | 59.615 | +28, +3.830 |
| 13 | Arkansas | 77.439 | 1-0 | 47.983 | -6, -3.408 |
| 14 | Pittsburgh | 77.245 | 0-1 | 79.682 | +7, -0.563 |
| 15 | Wisconsin | 76.727 | 1-0 | 63.933 | +18, +2.280 |
| 16 | Utah | 76.682 | 1-0 | 74.245 | +11, +0.563 |
| 17 | Penn State | 76.638 | 1-0 | 51.530 | -2, -2.046 |
| 18 | Georgia | 76.617 | 1-0 | 51.507 | +11, +0.898 |
| 19 | Southern Cal | 76.388 | 1-0 | 64.499 | +7, +0.266 |
| 20 | Arizona | 76.378 | 1-0 | 58.592 | +12, +1.551 |
| 21 | Brigham Young | 76.354 | 1-0 | 69.162 | +1, -1.192 |
| 22 | Iowa | 75.917 | 1-0 | 50.730 | -2, -2.115 |
| 23 | Clemson | 75.818 | 1-0 | 52.352 | -5, -2.423 |
| 24 | Oregon St | 75.769 | 0-1 | 84.965 | +7, +0.195 |
| 25 | Miami FL | 75.694 | 1-0 | 46.542 | -6, -2.525 |
Saturday, September 4, 2010
2010 Week 1 College Football Thoughts; Big games for Utah, USC, LSU, TCU, Boise State, Virginia Tech
With a few games played but some interesting one still to be played, here are a few observations and thoughts. While I mention rankings from my computer below, keep in mind that they are from last year and so how teams do will tell us a bit about how they compare to that ranking and where they should be this year. Stay tuned for the updated rankings after Monday nights game.
See the weekly predictions for more info and subscribe to the newsletter for detailed pick information.
First, from Thursday, there were a few meaningful games.
The big one was Utah beating Pitt in overtime 27-24. They were rated fairly close to end last season (Pitt #21 and Utah #27) so the computer had Utah winning by 1.3 being at home, so it got that right, but it was big for Utah's hopes of somehow sneaking into BCS consideration. They have TCU, Notre Dame, and BYU on their schedule so should they win out they'd certainly be justified of being considered.
USC answered some questions about their offense, but their defense didn't look so good and they were unable to cover the 21 points winning by 13, but that is about what the computer predicted having them by 11.4. So they'll likely stay about where they were which was #26.
Today so far, there haven't been any too terribly interesting games but Florida and Texas perhaps didn't win as convincingly as one would have thought. The computer had Florida winning by 38 so their 22 point win will likely drop them a bit. And the computer had Texas by 29 so 17 will likely drop them some too.
The late games today see LSU and UNC playing, the computer picking LSU by 4.8. With LSU #9 coming in from last year, this could say a lot about their potential to contend this year.
Also playing tonight is Oregon State and TCU. This is TCU's big non-conference game so there is pressure on them to perform as this could determine whether or not they vie for BCS consideration. The computer likes TCU by about 10 so look for them to try to top that to impress voters.
And the big game of the week is Boise State traveling back east to take on Virginia Tech. With both teams in the top-10 of the polls this year and #6 and #5 respectively from last years final computer ratings, this is a biggie. The computer likes Virginia Tech by a smidge, but that is based on last years ratings of course and this will certainly determine the fate of both teams BCS hopes.
See the weekly predictions for more info and subscribe to the newsletter for detailed pick information.
First, from Thursday, there were a few meaningful games.
The big one was Utah beating Pitt in overtime 27-24. They were rated fairly close to end last season (Pitt #21 and Utah #27) so the computer had Utah winning by 1.3 being at home, so it got that right, but it was big for Utah's hopes of somehow sneaking into BCS consideration. They have TCU, Notre Dame, and BYU on their schedule so should they win out they'd certainly be justified of being considered.
USC answered some questions about their offense, but their defense didn't look so good and they were unable to cover the 21 points winning by 13, but that is about what the computer predicted having them by 11.4. So they'll likely stay about where they were which was #26.
Today so far, there haven't been any too terribly interesting games but Florida and Texas perhaps didn't win as convincingly as one would have thought. The computer had Florida winning by 38 so their 22 point win will likely drop them a bit. And the computer had Texas by 29 so 17 will likely drop them some too.
The late games today see LSU and UNC playing, the computer picking LSU by 4.8. With LSU #9 coming in from last year, this could say a lot about their potential to contend this year.
Also playing tonight is Oregon State and TCU. This is TCU's big non-conference game so there is pressure on them to perform as this could determine whether or not they vie for BCS consideration. The computer likes TCU by about 10 so look for them to try to top that to impress voters.
And the big game of the week is Boise State traveling back east to take on Virginia Tech. With both teams in the top-10 of the polls this year and #6 and #5 respectively from last years final computer ratings, this is a biggie. The computer likes Virginia Tech by a smidge, but that is based on last years ratings of course and this will certainly determine the fate of both teams BCS hopes.
Friday, September 3, 2010
And the 2010 College Football Season Begins
The division I-A, ok FBS, season began last night with several games and gets into full swing this weekend. My computer simply uses last years end of season ratings to start the year which isn't perfect, but it is better than treating all teams the same to start the year. That means that Alabama starts the year a clear #1 over Florida.
Because I use last years ratings to start, I generally don't expect the computer to pick games very well the first few weeks and thus don't count those, but I've still posted predictions for this weekends games.
But what is perhaps more interesting is seeing what the computer projects for end of season records. This is basically saying what record each team should end up with if they play to last years performance against this years schedule. The full list can be found here, but two teams are projected to finish undefeated, TCU and Alabama, with Boise State, Texas, Ohio State, Virginia Tech, and Florida are projected to finish with 1 loss. However, the winner of the Boise State / Va Tech game this weekend could be projected to be undefeated too.
For comparison, after 1 week last year, the projections for the regular season were:
Because I use last years ratings to start, I generally don't expect the computer to pick games very well the first few weeks and thus don't count those, but I've still posted predictions for this weekends games.
But what is perhaps more interesting is seeing what the computer projects for end of season records. This is basically saying what record each team should end up with if they play to last years performance against this years schedule. The full list can be found here, but two teams are projected to finish undefeated, TCU and Alabama, with Boise State, Texas, Ohio State, Virginia Tech, and Florida are projected to finish with 1 loss. However, the winner of the Boise State / Va Tech game this weekend could be projected to be undefeated too.
For comparison, after 1 week last year, the projections for the regular season were:
- Boise State 13-0 (finished 13-0)
- Florida 12-0 (12-0)
- Penn State 11-1 (10-2)
- Alabama 11-1 (12-0)
- Cincinnati 11-1 (12-0)
- USC 11-1 (8-3)
Other than USC, nearly spot on.
In any case, stay tuned for more here and updates on the web-site.
Wednesday, April 21, 2010
Early 2010 NFL Season Projections
The NFL schedule came out yesterday which means I can now plug it in to my computer to see what it projects for the regular season ending records. The full projections are below, but first a little explanation and commentary.
First, these projections are based on the final ratings from the 2009/10 season. So they assume the teams performance will on average be unchanged, but clearly some teams will improve, others won't, and injuries will be a factor as well. But these projections are still useful as last years rating is a good starting point and the projections do take into account the specific schedule of each team.
Second, how do these projections work? For each game, the computer predicts the chance of a team winning that game based on last years rating, the opponent, and where the game is played, and then looks at every permutation of how a team could achieve every record from 0-16 to 16-0 and the chance of that occurring, then it identifies the record that has the highest probability of occurring. This is what appears in the Record column along with the chance of that record in the % chance column. Since it does compute the probability of every record, I also include the probability of a team winning and losing one more game but have the full range of probabilities if anyone is interested.
With that out of the way, what does it project? It says that New Orleans and San Diego will lead the league at 11-5 with 6 teams close behind at 10-6. Even more (7) teams will be just behind at 9-7 making it 15 teams vying for 12 playoff spots. At the bottom, St. Louis and Detroit will still be the cellar dwellers, but again, this doesn't take into account them presumably improving with their top picks in the draft.
You might ask, why is New Orleans only projected to go 11-5 after going 13-3 last year? And why a 4 game drop-off for Indy after they went 14-2 last year? Part of it is tougher schedules as these teams get the (in theory) toughest being the Super Bowl teams from last year, and part of it, particularly in Indy's case, is that they eked out a bunch of their wins last year and thus overachieved a bit and so they could easily have been a few games worse last year. Look for another blog entry on schedule strengths soon.
But also, by looking at the table below, there is quite a good chance that New Orleans will go 12-4 (20.8%) and even a 13 win season has a 14% chance so still very well within the realm of possibility. An undefeated season you ask? That comes in at only 0.23%.
Will the records below happen as is? Highly unlikely, but it is still fun to look at them as a baseline. If you think your team will improve more than the rest of the league (or their schedule), then add a game or two. If not, perhaps subtract a game or two.
And while the records are interesting, in the end what matters is who makes the playoffs. So I've put the AFC teams in red and NFC teams in blue with the division winners also in bold. Where the record is the same, I break the tie by giving it to the team with the higher chance of finishing one game better.
If things play out this way, the week 17 games that could be especially meaningful are:
What do you think? Will your team do better? Worse? Leave a comment!
First, these projections are based on the final ratings from the 2009/10 season. So they assume the teams performance will on average be unchanged, but clearly some teams will improve, others won't, and injuries will be a factor as well. But these projections are still useful as last years rating is a good starting point and the projections do take into account the specific schedule of each team.
Second, how do these projections work? For each game, the computer predicts the chance of a team winning that game based on last years rating, the opponent, and where the game is played, and then looks at every permutation of how a team could achieve every record from 0-16 to 16-0 and the chance of that occurring, then it identifies the record that has the highest probability of occurring. This is what appears in the Record column along with the chance of that record in the % chance column. Since it does compute the probability of every record, I also include the probability of a team winning and losing one more game but have the full range of probabilities if anyone is interested.
With that out of the way, what does it project? It says that New Orleans and San Diego will lead the league at 11-5 with 6 teams close behind at 10-6. Even more (7) teams will be just behind at 9-7 making it 15 teams vying for 12 playoff spots. At the bottom, St. Louis and Detroit will still be the cellar dwellers, but again, this doesn't take into account them presumably improving with their top picks in the draft.
You might ask, why is New Orleans only projected to go 11-5 after going 13-3 last year? And why a 4 game drop-off for Indy after they went 14-2 last year? Part of it is tougher schedules as these teams get the (in theory) toughest being the Super Bowl teams from last year, and part of it, particularly in Indy's case, is that they eked out a bunch of their wins last year and thus overachieved a bit and so they could easily have been a few games worse last year. Look for another blog entry on schedule strengths soon.
But also, by looking at the table below, there is quite a good chance that New Orleans will go 12-4 (20.8%) and even a 13 win season has a 14% chance so still very well within the realm of possibility. An undefeated season you ask? That comes in at only 0.23%.
Will the records below happen as is? Highly unlikely, but it is still fun to look at them as a baseline. If you think your team will improve more than the rest of the league (or their schedule), then add a game or two. If not, perhaps subtract a game or two.
And while the records are interesting, in the end what matters is who makes the playoffs. So I've put the AFC teams in red and NFC teams in blue with the division winners also in bold. Where the record is the same, I break the tie by giving it to the team with the higher chance of finishing one game better.
If things play out this way, the week 17 games that could be especially meaningful are:
- Carolina at Atlanta - Both projected at 9-7 but only one likely to make the playoffs
- Dallas at Philly - Dallas projected at 10-6 against Philly's 9-7, and Philly might not get in at 9-7 so could be a swing game
- Arizona at San Francisco - Both projected at 9-7 and one likely the division winner, but that team also possibly so only one to get in the playoffs
What do you think? Will your team do better? Worse? Leave a comment!
| Team | Record | % chance | % +1 | % -1 |
|---|---|---|---|---|
| Team | Record | % chance | % +1 | % -1 |
| New Orleans | 11-5 | 22.1 | 20.8 | 17.4 |
| San Diego | 11-5 | 21.1 | 16.4 | 20.2 |
| Minnesota | 10-6 | 21.6 | 19.9 | 17.6 |
| Indianapolis | 10-6 | 20.7 | 19.9 | 16.6 |
| Dallas | 10-6 | 21.5 | 18.9 | 18.5 |
| NY Jets | 10-6 | 20.7 | 18.1 | 18.2 |
| Green Bay | 10-6 | 21.3 | 17.5 | 19.6 |
| New England | 10-6 | 20.3 | 16.5 | 19.2 |
| Baltimore | 9-7 | 20.4 | 20.1 | 15.9 |
| Carolina | 9-7 | 20.0 | 20.0 | 15.6 |
| Philadelphia | 9-7 | 20.0 | 19.5 | 15.9 |
| Atlanta | 9-7 | 20.5 | 18.8 | 17.2 |
| Arizona | 9-7 | 21.1 | 17.8 | 18.9 |
| San Francisco | 9-7 | 20.9 | 16.9 | 19.7 |
| Houston | 9-7 | 20.1 | 16.1 | 19.4 |
| Pittsburgh | 8-8 | 20.5 | 19.3 | 16.8 |
| Denver | 8-8 | 20.5 | 18.5 | 17.5 |
| Miami | 8-8 | 21.3 | 17.5 | 19.6 |
| Tennessee | 8-8 | 20.1 | 16.6 | 18.9 |
| Buffalo | 8-8 | 20.6 | 16.5 | 19.7 |
| NY Giants | 8-8 | 20.4 | 16.3 | 19.6 |
| Cincinnati | 7-9 | 20.4 | 19.5 | 16.4 |
| Chicago | 7-9 | 21.3 | 17.2 | 19.9 |
| Tampa Bay | 7-9 | 21.4 | 16.8 | 20.5 |
| Jacksonville | 6-10 | 20.5 | 19.9 | 16.1 |
| Kansas City | 6-10 | 20.5 | 19.5 | 16.6 |
| Washington | 6-10 | 20.6 | 19.4 | 16.8 |
| Seattle | 6-10 | 21.1 | 17.6 | 19.2 |
| Oakland | 6-10 | 21.1 | 16.6 | 20.3 |
| Cleveland | 5-11 | 21.8 | 18.6 | 19.1 |
| St Louis | 4-12 | 22.7 | 20.5 | 18.2 |
| Detroit | 3-13 | 26.4 | 17.8 | 25.7 |
Tuesday, April 6, 2010
NCAA Tournament Final System and Poll Comparison
And Duke wins it all. And with no poll or system picking them, there aren't any big scores for any system/poll, but I am pleased to see that my computer eked out the best score beating Sagarin by 2 points.
The difference between my computer and Sagarin ultimately was my computer having Butler getting to the sweet 16 and Sagarin having them losing in the second round. The big difference against the rest was the my computer and Sagarin both had Baylor getting to the Elite Eight (AP, Coaches, Seeds, Vegas had Villanova) and Sagarin and my computer had Duke getting to the final.
I'd like to think this proves my computer did a better job ranking and ratings teams than the polls or Vegas, but it is just one year. But I'll do this again next year and share it all again.
| System/Poll | Points |
|---|---|
| Schmidt Computer | 89 |
| Sagarin | 87 |
| Coaches | 73 |
| RPI | 69 |
| AP | 68 |
| Seeds | 62 |
| Vegas | 61 |
The difference between my computer and Sagarin ultimately was my computer having Butler getting to the sweet 16 and Sagarin having them losing in the second round. The big difference against the rest was the my computer and Sagarin both had Baylor getting to the Elite Eight (AP, Coaches, Seeds, Vegas had Villanova) and Sagarin and my computer had Duke getting to the final.
I'd like to think this proves my computer did a better job ranking and ratings teams than the polls or Vegas, but it is just one year. But I'll do this again next year and share it all again.
Subscribe to:
Posts (Atom)