Sunday, October 31, 2010

2010 Week 9 College Football Projections; Oregon, TCU, and Boise State undefeated

Here are the projected records for the regular season using results through week 9 of the season.  Full projections on the site.

And then there were five.  With Missouri and Michigan State losing, there are just five undefeated teams remaining.  And that results in still three being projected to finish that way.

There is a change in the chances for those three with Oregon now with a 71% chance of no losses against a 26% chance of one, TCU at 56% and 44%, and Boise State at 55% and 38%.  All now have a better chance of no losses than any other result.

TCU has what will be their stiffest test next week visiting #7 Utah, but they have #44 San Diego State left too.  Boise State has #35 Hawaii next week then #22 Nevada in a month.  Oregon has the toughest schedule with #9 Arizona, #14 Oregon State, #24 California, and #45 Washington left, but they are rated so highly they are expected to breeze through.

What of the other 2?

Utah as noted above must play TCU, but they are still projected to finish with just that one loss and a 29% chance of none.

Auburn still hosts #36 Georgia, but then must go to #12 Alabama resulting in a 60% chance of the one loss projected record, but 34% chance of undefeated.

TeamRecord% chance% +1% -1
TeamRecord% chance% +1% -1
Boise St12-
Michigan St11-

2010 Week 9 College Football Ratings and Rankings; Oregon #1, TCU #2, Boise State #3; Auburn #10!

And Oregon is still #1.  Not only that, their rating margin over #2 TCU is now nearly 7 points.  According to my computer, they are the far and away clear #1, BCS be damned.  And so far my computer is picking games between BCS top-25 teams 2 games better than the BCS rankings.  Full rankings on the site, but top-30 below for quick reference.

There is naturally some movement behind with Missouri and Michigan State dropping and Nebraska and Iowa moving up, but OU suffers from the Missouri loss having just lost to them last week.  Stanford's whitewashing of UW also moves them up a few spots back to #4.  With their only loss on the road to far and away #1 Oregon, hard to argue with that.

What about Auburn you say?  They did move up some with their win, but are only #10.  That may not pass the smell test, but when you look at their performance chart you can begin to understand.

Auburn is winning, but especially early in the season was not doing so impressively.  Mississippi State at #30 isn't a bad team, but Auburn won by only 3, Clemson is only 4-4 and #43 and Auburn won by only 3 at home, the win over #17 and 6-2 South Carolina was a little better by 8, but then they win by only 3 against a #56 and 4-5 Kentucky?  Their most impressive win was against 6-2 and #25 Arkansas.  So they are winning, but not in dominating fashion against any of their quality opponents, and they have close wins against bad opponents, even at home.

Interested in charts for all the teams?  Subscribe to the newsletter!

The computer continued it prowess in picking games going 30-23-1 in all games and 2-1 in locks meaning 3 consecutive winning weeks.  After some of the strangeness early in the year, the computer seems dialed in.  See the prediction performance for more details.

1Oregon93.3328-065.628+0, +1.717
2TCU86.3919-063.364+0, -0.473
3Boise St84.3097-063.233+1, -0.447
4Stanford83.1777-170.011+2, +0.860
5Nebraska83.1647-166.546+4, +1.721
6Iowa82.7816-265.920+5, +1.661
7Utah81.7708-060.311-2, -1.901
8Missouri81.6537-168.662-5, -3.375
9Arizona81.2727-167.188+3, +0.190
10Auburn80.4739-067.665+3, -0.590
11Oklahoma80.4187-169.449-4, -1.538
12Alabama80.3557-167.036+3, -0.055
13Ohio State80.0108-162.308-5, -1.945
14Oregon St79.1484-376.978+10, +2.399
15Southern Cal78.9475-372.095-1, -1.745
16Wisconsin78.4827-163.538+1, -1.251
17South Carolina77.7666-267.216+4, -1.296
18Oklahoma St77.3417-165.455+8, +1.696
19Michigan St77.1568-165.267-9, -4.169
20LSU77.0497-167.195+2, -0.860
21Florida St77.0476-266.408-3, -2.570
22Nevada76.5207-160.502-2, -2.590
23Virginia Tech76.1746-263.504+0, -0.654
24California76.1214-472.312-8, -3.643
25Arkansas75.8246-266.408+2, +0.198
26Arizona St75.8214-470.783-1, -0.730
27Miami FL75.2265-367.838-8, -4.320
28Pittsburgh74.3675-366.598+1, -0.424
29Florida74.2975-368.793+4, -0.024
30Mississippi St74.2137-264.357-2, -0.763

Saturday, October 30, 2010

How the BCS does picking games compared to my computer part 2

Continuing what I promised as part of my analysis of the BCS, here is week 2 of tracking how well the BCS predicted games as compared to my computer.  Last week my computer picked 3 of 4 games right and the BCS picked 2 of 4.

The following are the games between BCS top-25 teams:
  • #5 Michigan State vs #18 Iowa - The BCS clearly liked Michigan State but my computer liked Iowa.  Schmidt Computer 1, BCS 0.
  • #6 Mizzou vs #14 Nebraska - Again, the BCS liked Missouri quite a bit, and my computer did very narrowly, just short of picking the upset correct.  SC 1, BCS 0.
So, after two weeks, my computer extended its advantage, now picking 4 of 6 correct while the BCS is 2 of 6.

Stay tuned.

Sunday, October 24, 2010

How the BCS does picking games compared to my computer

When doing my analysis of the BCS last week, I mentioned I'd be tracking how well the BCS predicted games as compared to my computer, so here is the first edition of that comparison.

The following are the games between BCS top-25 teams:

  • #1 OU vs #11 Mizzou - The BCS clearly liked Oklahoma here, but my computer had Missouri at #5 vs #6 OU and picked the "upset".  Schmidt Computer 1, BCS 0.
  • #4 Auburn vs #6 LSU - I had Auburn #10 and LSU #18.  SC 2, BCS 1.
  • #13 Wisconsin vs #15 Iowa - My computer liked Iowa better with #9 vs #20.  SC 2, BCS 2.
  • #14 Oklahoma State vs #16 Nebraska - My computer saw it the other way with Nebraska #11 and OSU #22.  SC 3, BCS 2.
So, after one week, my computer has a slight advantage picking 3 of 4 games right versus the BCS 2 of 4.

Stay tuned.

2010 Week 8 College Football Projections; TCU, Boise State, and Oregon undefeated

Here are the projected records for the regular season using results through week 8 of the season.  Full projections on the site.

And then there were 7.  With Oklahoma, LSU, and Oklahoma State all losing, there are just 7 undefeated teams remaining.  And that results in 3 being projected to finish that way now.

Those three are TCU, new from last week and projected to have a 49.9% chance of undefeated against a 49.3% chance of one loss, Boise State at 45% and 43.9%, and Oregon at 44.1% vs 41.1%.  So no slam dunk for any of them, a one loss record being almost as likely as undefeated.  But there is a very good shot at least one of them does it.

What of the other 4?

Michigan State has a 33% chance of pulling it off, but with Iowa on the schedule for next week has a better shot of one loss at 62%.

Auburn improved their chances after their win against LSU, but must still play Alabama giving them a 34% chance of undefeated but 58% of one loss.

Missouri would be favored in the rest of their games this year and results in a 39% of an undefeated season, but with Nebraska on the schedule for next week and an Iowa State team that has given some teams some trouble several weeks later, one loss is more likely at 48%.

Utah must face TCU in a couple weeks and while they very well could win that game and finish undefeated, that has only a 25% chance and one loss is more likely at 45%.

TeamRecord% chance% +1% -1
TeamRecord% chance% +1% -1
Boise St12-
Michigan St11-

2010 Week 8 College Football Ratings and Rankings; Oregon #1, TCU #2, Missouri #3

And Oregon is still #1.  Full rankings on the site, but top-30 below for quick reference.

Oregon's margin over #2 TCU stays over four points, but is less than the nearly eight it was last week.  This is due to prior opponent ASU losing big to Cal and also Stanford's closer than expected win over WSU.

There is a new #3 in Missouri close behind TCU at #2.  Missouri replaces now #4 Boise State based on their win over Oklahoma.  The computer did pick the Missouri win and upset, one of the correct picks in a decent 28-22-1 week against the spread, but they still get moved up a bit based in part on Boise State dropping with some prior opponents doing more poorly than expected.

Utah makes a move up continuing their undefeated season at #5 and Ohio State bounces back too getting back into the top-10 at #8.

Cal makes the biggest move up in the top-30, 11 spots, and it is interesting to look at their performance chart as that is an effective way to see how schizophrenic they have been.

They've had no games that even overlap with their current rating.  They are either playing great (Colorado, UCLA, ASU), or poorly (Nevada, USC).  The Arizona loss was actually a pretty good result for them too.

Interested in charts for all the teams?  Subscribe to the newsletter!

As noted above, the computer had a pretty good week going 28-22-1 against the spread, this after going 30-23 last week.  Picks meeting the lock criteria also went 3-1 after going 5-2 last week.  See the prediction performance for more details.

1Oregon91.6157-064.135+0, -2.961
2TCU86.8648-064.802+0, -0.006
3Missouri85.0287-066.943+2, +0.293
4Boise St84.7566-064.287-1, -1.685
5Utah83.6717-059.314+3, -0.151
6Stanford82.3176-170.572-2, -3.645
7Oklahoma81.9566-172.177-1, -2.371
8Ohio State81.9557-164.486+6, +0.982
9Nebraska81.4436-166.320+2, +0.112
10Michigan St81.3258-063.629-3, -2.574
11Iowa81.1205-265.363-2, -1.726
12Arizona81.0826-166.663+4, +0.687
13Auburn81.0638-067.955-3, -0.538
14Southern Cal80.6925-270.095-1, -0.458
15Alabama80.4107-167.345+0, -0.079
16California79.7644-372.676+11, +2.280
17Wisconsin79.7337-165.233+3, +0.244
18Florida St79.6176-165.444-1, -0.560
19Miami FL79.5465-269.038+6, +1.958
20Nevada79.1106-162.837+1, +0.087
21South Carolina79.0625-269.269-2, -0.864
22LSU77.9097-168.030-4, -2.027
23Virginia Tech76.8286-264.293+1, -1.214
24Oregon St76.7493-378.134-1, -1.401
25Arizona St76.5513-474.343-13, -4.732
26Oklahoma St75.6456-164.607-4, -2.703
27Arkansas75.6265-267.579+4, +0.136
28Mississippi St74.9766-264.644+1, -1.780
29Pittsburgh74.7914-368.488+6, +0.433
30Clemson74.5784-362.130+8, +0.794

Tuesday, October 19, 2010

2010 NFL Week 6 Projections

As the rankings become clearer after week 6, so too do the season ending projections.

The AFC playoff teams are projected to be the Jets, Pittsburgh, Indy, and KC as division winners with Baltimore and New England the wildcards.  Tennessee gets left out but after that, no one is close.

In the NFC, Atlanta, the Giants, Chicago, and Seattle win the divisions with New Orleans and Philly getting the wildcards.  There are a host of teams close behind the two 9-7 wildcards with four at 8-8 and another two at 7-9 so this race could be tight.

It is interesting how clearly this shows the AFC's dominance this year.  All the AFC playoff teams are 10-6 or better with one 10-6 team getting left out, while only one NFC team is better than 10-6.  Is that really the way it will end up?

Team Record % chance % +1 % -1
Team Record % chance % +1 % -1
NY Jets 13-3 36.1 24.5 23.8
Pittsburgh 13-3 32.2 22.6 24.4
Baltimore 12-4 32.8 31.5 17.5
New England 12-4 25.5 16.6 24.9
Atlanta 11-5 27.7 20.4 23.4
Indianapolis 11-5 26.9 17.4 25.4
Kansas City 10-6 26.2 24.2 18.6
Tennessee 10-6 25.2 24.5 17.4
NY Giants 10-6 24.4 17.0 23.7
New Orleans 9-7 26.5 18.4 24.4
Chicago 9-7 25.5 17.1 24.9
Philadelphia 9-7 24.6 24.0 17.0
Seattle 9-7 24.4 20.3 20.4
Green Bay 8-8 28.1 22.8 21.4
Miami 8-8 26.6 24.0 18.8
Houston 8-8 26.1 18.8 23.9
St Louis 8-8 25.6 21.8 20.2
Denver 8-8 25.3 18.5 23.1
Tampa Bay 8-8 25.0 18.4 23.0
Arizona 8-8 24.2 21.7 18.8
Jacksonville 7-9 26.3 22.3 20.3
Minnesota 7-9 26.1 22.9 19.4
San Diego 7-9 25.9 19.9 22.4
Washington 7-9 25.4 23.7 18.0
Cincinnati 6-10 31.0 25.3 20.7
Detroit 6-10 25.3 19.5 22.1
Dallas 6-10 24.8 20.3 20.9
Oakland 5-11 27.7 17.6 27.1
San Francisco 5-11 26.2 20.7 21.8
Cleveland 4-12 33.1 23.6 23.9
Carolina 2-14 31.6 17.5 31.1
Buffalo 1-15 36.1 29.2 18.3

2010 NFL Week 6 Ratings and Rankings; Pittsburgh #1, Jets #2, Baltimore #3

And there is finally a little stability at the top.  The top-4 teams remain at the top with Baltimore and the Jets just swapping spots at #2 and #3.  We are also seeing some clearer separation, particularly in the AFC, with 6 teams at 86 points or more well ahead of the rest at below 83.  The NFC is actually a bit of a logjam with #5 thru #12 all within 12 points.

The computer had a good week predicting games going 9-4-1 against the spread and 1-0 on best picks.  This after a 9-5 week against the spread, 3-0 on best picks, and 2-0 on locks.  Perhaps the computer has finally dialed in to this whacky year.

See full rankings and other info on the site.

1Pittsburgh90.0764-182.158+0, -0.935
2NY Jets89.6265-181.827+1, +0.120
3Baltimore88.5014-285.289-1, -1.025
4New England87.7384-181.554+0, +0.507
5Tennessee87.0514-281.412+2, +1.564
6Indianapolis86.3354-281.993-1, +0.378
7Atlanta83.5514-281.289-1, -2.308
8Philadelphia83.2364-279.713+7, +2.282
9NY Giants83.2214-279.580+2, +0.933
10Kansas City82.7863-280.497-2, -1.518
11Green Bay81.8713-379.460-2, -1.665
12Miami81.5873-282.841+5, +1.735
13Denver81.1662-484.857+3, +0.861
14Chicago80.6794-278.851-4, -1.959
15Minnesota80.5562-381.422+8, +1.206
16Houston80.4724-280.350+4, +0.781
17San Diego80.4632-479.678-3, -0.555
18Washington80.0313-380.740+3, +0.419
19Detroit79.8141-582.394+3, +0.242
20Dallas79.7301-482.358-1, -0.085
21Cincinnati79.6442-380.611-8, -1.858
22New Orleans79.3514-277.004+7, +3.211
23Seattle79.3463-279.767+3, +2.646
24St Louis78.7983-377.752+1, +1.003
25Jacksonville78.2823-380.611-1, -0.471
26Cleveland77.4861-583.627-8, -2.343
27Tampa Bay77.2043-278.537-15, -4.519
28Arizona76.8583-280.232+0, +0.421
29Oakland75.9982-479.728-2, -0.630
30San Francisco74.7291-580.711+1, +3.116
31Buffalo71.4130-583.221-1, -0.362
32Carolina69.1270-579.420+0, -1.279

Sunday, October 17, 2010

Comparing the BCS Computers to the Schmidt Computer

In my last entry, I pointed out a flaw in the BCS computers being that they aren't allowed to use the scores or margin of victory (MOV) in their algorithms.  I also pointed out that my algorithm does take it into account, but does so in a way that doesn't reward running up the score against strong opponents but allows it against weak opponents in order to not punish a team too much for not playing a strong opponent every game.

Since I've gone out and made that statement, lets compare my rankings to the BCS computer average.  The table below lists the two with the difference, red where the BCS is too high and blue where it is too low.

TeamBCS RankSchmidt ComputerDifference
Oklahoma 1 6 5
Oregon 8 1 7
Boise State 7 3 4
Auburn 3 10 7
TCU 5 2 3
LSU 2 18 16
Michigan State 4 7 3
Alabama 12 15 3
Utah 11 8 3
Ohio State 14 14 0
Missouri 6 5 1
Stanford 10 4 6
Wisconsin 16 20 4
Oklahoma State 9 22 13
Iowa 17 9 8
Nebraska 20 11 9
Florida State 13 17 4
Arizona 15 16 1
Texas 18 28 10
West Virginia 22 49 27
South Carolina 23 19 4
Kansas State 19 36 17
Arkansas 28 31 3
Mississippi State 21 29 8
Virginia Tech 2824 4

What do you think?  Which set of rankings is more realistic?

To see if we can determine which is more accurate, I'll keep track of which does better picking winners between BCS top-25 games the rest of the year.

Initial BCS Rankings Computer Analysis; Big-Televen Bias and Margin of Victory, Vote in a Poll!

The initial BCS rankings were released this evening and were remarkably close to what my earlier analysis predicted.  I had all of the top-10, nailed the top-3, #6, #9, and #10 and had #4 and #5 swapped and #7 and #8 swapped.  Brad Edwards, look out!

But now that they are out we can do a deeper analysis and see what makes sense and what doesn't.

First, the polls and computers are way out of whack on a few teams.  Comparing the computer after with the polls we see that:
  • Oregon is #1 in both polls but #8 in the computers.  Only Sagarin has them rated close to their poll ranking (#2) so the computers are pretty consistent.  This is likely because the computers cannot take margin of victory (MOV) into account against their (so far) weak schedule.  My ratings have their schedule #69 in FBS.
  • Boise State is #2 in the polls versus #7 in the computers.  Not quite as bad as Oregon but the same reason as their schedule is #59.
  • Alabama doesn't get respect from the computers being #7 in the polls and #12 in the computers.  The computers seem to ding the loss more than the pollsters.
  • On the other hand, Missouri is #16 in the polls and #6 in the computers.  But oddly, Billingsley doesn't have Missouri in the top-25!  Hard to explain that one as my computer has their schedule #55, but Billingsley seems to not use where a team is ranked when the game occurs and doesn't adjust if a team improves, and so Missouri doesn't get credit for Illinois and San Diego State being better than expected.  Seems to be a fatal flaw of the algorithm, just that's just IMHO.
  • The computers don't like Wisconsin having them #16 while the polls have them #11.  This seems to be pollster bias and not a MOV issue though as my computer has the #20 and their schedule is #76.
  • The computers do like Oklahoma State at #9 but the polls don't as much at #15.  I agree more with the polls having them #22 and their schedule strength is #77 so hard to figure why the computers have them so high.  Note that Sagarin agrees with me having them #19.
  • That pollster bias appears again for Iowa having them #12 while the the computer have them #17.  With a schedule rated #84 the bias seems the issue.  Note the two bias cases thus far are in the Big-Televen.
  • Nebraska at #13/14 in the polls is higher than the computer's #20.  My computer has them at #11 so this seems to be a case of the BCS computers not being able to take MOV into account.
  • Kansas State is #25 and NR in the polls but #19 in the computers.  I agree with the polls more as my computer has them #36.  They have a decent schedule at #43.  Here the computer's inability to take into account MOV in their big loss to Nebraska seems to be the issue.
A couple things stand out in the above.  First, there seems to be a bit of Big-Televen bias (Ohio State is 4 spots higher in polls than computers too).  Second, BCS handcuffing the computers not allowing MOV causes some problems.

The subject of MOV is controversial and the BCS got rid of the computers that used it after Nebraska snuck into the championship game a few years ago.  The issue is that without it you punish teams unnecessarily when they play their weaker opponents.  For example, Boise State would have been #1 in the BCS last week but because they played a weak opponent (whom they beat 48-0 after calling off the dogs) they dropped in the computers this week and are #3 in the BCS.

Now, you can't reward running up the score (RUTS) but an algorithm that gives diminishing returns against strong opponents so as to not reward it, but gives suitable credit against weak teams to not punish it what is needed.  And that is exactly what my algorithm does and (IMHO) results in rankings that pass the smell test much better than most of the BCS computers and also performs well picking games.

What do you think?  Vote in the poll on the right to let your voice be heard.

2010 BCS Rankings Prediction - First Release - Oklahoma #1, Oregon #2, Boise State #3

Brad Edwards on ESPN reported last week that if released last week, Boise State would have been #1.  I did some quick analysis and confirmed that, but with all the changes in the top-10 this week, I updated my analysis and it shows Oklahoma should be #1 when the BCS rankings are released later today.

My analysis is based on taking the computer's rankings that are available (Sagarin), those that publish their algorithm that I've reproduced (Colley Matrix), and guessing at the changes for the three others that have been releasing their rankings.  I've also included the latest Coaches poll and guessed at updates to the Harris poll.

Here is the rough result:

1 Oklahoma 0.9244
2 Oregon 0.9215
3 Boise State0.9189
4 TCU 0.8606
5 Auburn 0.8310
6 LSU 0.8113
7 Alabama 0.7327
8 Michigan State 0.7188
9 Utah 0.6516
10 Ohio State 0.5847

Oklahoma makes its move based primarily on the computer where they should be a near unanimous #1.  Oregon's #1 in both polls is offset by the computers having them around #7 and with Boise State #4 in the computers, the gap is there, narrow as it is, for OU to slip in.  It is very close though so any slight error in my extrapolation/guessing could change it so any of the 3 are #1.

There is a bit of a gap to #4 TCU with Auburn and LSU fairly close for #5 and #6 (they should be #2 and #3 in the computers).  It is interesting that it appears Alabama will be #7 ahead of undefeated Michigan State and Utah, although Michigan State is close so could eke by.

We'll see how close this is!

2010 Week 7 College Football Projections; Oregon, Boise State undefeated

Here are the projected records for the regular season using results through week 7 of the season.  Full projections on the site.

And then there were 10.  With Nevada, Ohio State, and Nebraska all losing, there is still a surprising number of undefeated teams at this point of the season.  Last year at this point there were just 7.

But unlike last week where 4 teams were projected to finish unblemished, that number is down to 2 this week, those being Oregon and Boise State.  Oregon has a reasonably tough schedule the rest of the year with USC, California, Arizona, and Oregon State, but has a 62% chance of no losses.  Boise State's chances were helped a bit by Nevada proving to be fallible and is at 52%.

Of the other 8, 6 are projected to have 1 loss, of those the teams with the best chance of finishing undefeated are TCU (48%), Michigan State (39%), and Auburn (26%).

2010 Week 7 College Football Ratings and Rankings; Oregon #1, TCU #2, Boise St #3

For the third week in a row, Oregon finds themselves at #1 and by a wide margin.  But there is a lot of change behind them with half of the top-10 being new members.  And with Oregon's fairly tough schedule the remainder of the year (3 are top-25 in the rankings and  all 6 are top-50) that gap could change or they could solidify their stranglehold.

The change behind them starts with TCU leapfrogging Boise State to move to #2 due primarily to BSU playing a far weaker opponent than TCU this week.  Stanford moves to #4 based on prior opponents USC and Notre Dame winning big and their only loss being to far and away #1 Oregon on the road.  Missouri leaps into the top-10 at #5 and Oklahoma moves up right behind them.  Michigan State also joins the top-10 as does Utah and Auburn with Iowa a hold over from last week between them.

With several new teams in the top-10, it can be helpful to look at performance charts.  First lets compare Missouri and Oklahoma, particularly since they are in a virtual ratings tie and play next week.

Missouri has been on a steady climb playing their two best games of the season the last two weeks.  OU on the other hand had their big win over FSU early, then struggled in a few games before having a good performance against a not so strong Iowa State this week.

Michigan State started off playing creampuffs so hard to judge them early, but has hit their stride the last 3 weeks.

Auburn has been pretty consistent, with the big win over Arkansas key to them moving up.

Interested in charts for all the teams?  Subscribe to the newsletter! Full rankings on the site, but top-30 below for quick reference.

1Oregon94.5766-064.927+0, +1.423
2TCU86.8707-064.904+2, +0.011
3Boise St86.4416-065.717+0, -1.266
4Stanford85.9625-174.387+3, +1.089
5Missouri84.7356-065.924+10, +3.066
6Oklahoma84.3276-071.651+6, +2.004
7Michigan St83.8997-064.501+11, +3.847
8Utah83.8226-060.297+8, +2.321
9Iowa82.8465-162.968+1, +0.034
10Auburn81.6017-068.101+13, +3.007
12Arizona St81.2833-374.883+7, +1.480
13Southern Cal81.1505-270.801+14, +4.168
14Ohio State80.9736-165.440-9, -5.885
15Alabama80.4896-168.229-6, -2.561
16Arizona80.3955-166.866-3, -1.804
17Florida St80.1776-166.197-6, -2.276
18LSU79.9367-066.946-1, -0.755
19South Carolina79.9264-271.073-5, -2.085
20Wisconsin79.4896-164.034+15, +5.276
21Nevada79.0236-163.571-15, -6.215
22Oklahoma St78.3486-063.639+7, +2.270
23Oregon St78.1503-379.300-2, -1.281
24Virginia Tech78.0425-267.068-4, -1.626
25Miami FL77.5884-268.702+0, -0.531
26Air Force77.5105-265.133+0, -0.237
27California77.4843-373.084-19, -5.656
28Texas76.8544-271.689+11, +3.564
29Mississippi St76.7565-266.731+1, +0.904
30North Carolina St75.6605-266.497-6, -2.635

Tuesday, October 12, 2010

2010 NFL Week 5 Projections; Baltimore, Pittsburgh, Jets all 13-3, 10-6 Titans miss the playoffs

The computers projected records are up and they are interesting.

It is probably no surprise that the AFC is stronger, but with 7 of the best 10 records (10-6 and above), one of those 10-6 AFC teams will be left out of the playoffs.  Right now the computer says it will be Tennessee, but they are tied with Indianapolis at 10-6 and the nod goes to the Colts right now simply because they have a slightly better chance of finishing 10-6.  Those two games in weeks 14 and 17 will have an awful lot of meaning.

The other AFC teams aren't a surprise either being Baltimore and Pittsburgh also tied but both making it at 13-3, the Jets, Kansas City, and New England.  After the Titans at 10-6 things drop off to a group of 8-8 teams, the dangerous one there being San Diego.

In the NFC it could be a lot more competitive with 11 teams a 7-9 or better and an 8-8 team making it.  Right now the teams would be Atlanta, Chicago, the Giants, and Arizona, with Tampa Bay and Green Bay the wildcards.

Team Record % chance % +1 % -1
Team Record % chance % +1 % -1
Baltimore 13-3 33.5 23.0 24.7
Pittsburgh 13-3 31.8 22.8 24.0
NY Jets 13-3 29.6 15.5 28.8
Atlanta 12-4 27.5 20.8 22.8
Kansas City 12-4 26.3 18.7 23.7
New England 11-5 23.9 17.2 22.8
Tampa Bay 10-6 25.6 22.2 19.8
Indianapolis 10-6 25.4 22.1 19.8
Chicago 10-6 24.9 24.2 17.1
Tennessee 10-6 24.3 17.8 22.7
Green Bay 9-7 26.3 25.4 17.4
NY Giants 9-7 24.0 20.8 19.6
Arizona 8-8 24.3 22.2 18.4
Denver 8-8 24.1 17.1 23.2
Jacksonville 8-8 24.1 17.2 23.2
Philadelphia 8-8 24.0 21.7 18.6
Washington 8-8 23.9 17.1 23.2
San Diego 8-8 23.8 16.9 23.2
Cincinnati 7-9 32.3 20.5 25.9
Miami 7-9 25.3 18.6 23.1
Houston 7-9 25.0 21.8 19.5
St Louis 7-9 23.9 23.1 17.0
Seattle 7-9 23.8 21.8 18.2
Dallas 7-9 23.7 18.4 21.6
New Orleans 6-10 29.2 25.3 19.5
Oakland 6-10 26.0 17.4 25.1
Minnesota 6-10 24.7 22.2 18.7
Detroit 6-10 24.2 20.3 20.3
Cleveland 5-11 30.0 22.9 22.7
San Francisco 3-13 25.0 24.3 16.7
Carolina 2-14 34.4 20.4 28.5
Buffalo 1-15 37.1 28.0 20.2

2010 NFL Week 5 Ratings and Rankings; Pittsburgh #1, Baltimore #2, Jets #3

After four to five games each, we can start to get a clearer picture of teams, and the computer finally started predicting games better!

And idle Pittsburgh moves up a spot to the top of the rankings with a slight margin over the Jets and Baltimore in a virtual tie.  The AFC dominance continues with four of the next five, Atlanta being the only NFC team to break things up at #6.

The big mover of the week is the Giants after easily handling what had been a good Houston team.  Houston in turn is a big loser for the week but New Orleans topped them and falls to #29 despite a 3-2 record.

Why is New Orleans so low?  Their performance chart is telling:

Their wins are over teams with a combined 1-13 record and they lost to the only good team they played (Atlanta) and a bad Arizona team with a rookie quarterback starting his first game.  And what may be more concerning for Saints fans is that the performance seems to be in a steady decline.

The computer finally had a good week predicting games going 9-5 against the spread and 7-7 picking winners versus Vegas going 5-9 picking winners.  The computer also went 3-0 on best picks and 2-0 on locks.

Stay tuned for an entry on projected regular season records.

Sunday, October 10, 2010

Checking up on the polls; anti Pac-10 bias, SEC ranked too high

As we get to the point that the BCS rankings will be coming out, let's take a look at how the polls stack up against my computer's rankings to see where they may be off.

Below is the AP top-25  with columns for the Coaches poll and my computer with a comment on why the computer ranks them where they do for each.  Red numbers indicate a team the polls have too high, green too low, both where difference is 3 or greater.

Ohio State 1 1 4 5 of 6 at home, all opps at least 2 losses
Oregon 2 2 1 Big win over 1 loss Stanford
Boise St. 3 3 3 Beat good Va Tech and Oregon St.
TCU 4 5 5 Beat good Oregon St. rolled others
Nebraska 5 4 2 Rolled over K-State and UW on road
Oklahoma 6 6 12 Beat FSU easily, others close, 6 too high
Auburn 7 7 23 Beat South Carolina but close, others very close, 7 too high
Alabama 8 8 9 Arkansas and Florida good wins, but the 1 loss
LSU 9 9 17 Lots of close games, haven't impressed even with no loss
South Carolina 10 12 14 Big win over Alabama, but close against not so good Georgia
Utah 11 10 16 No opponent thus far has a winning record
Arkansas 12 13 22 Close games haven't impressed
Michigan St. 13 11 18 Only 1 road game and that still in Michigan
Stanford 14 15 7 Only loss to undefeated Oregon, 14/15 too low
Iowa 15 14 10 Only loss to 1-loss Arizona on the road
Florida St. 16 17 11 Only loss to undefeated OU, easily won others
Arizona 17 20 13 Only loss to good Oregon St.
Wisconsin 18 16 35 Lost only game against opp with winning record
Nevada 19 21 6 Handily beat a pretty good Cal, handled others easily
Oklahoma St. 20 18 29 Only good opp beat by only 3 at home
Missouri 21 19 15 Only one opp as a losing record
Florida 22 22 28 Lost to only 2 good opps
Air Force 23 23 26 About right
Oregon St. 24 29 21 Only losses to 2 undefeateds
West Virginia 25 25 36 Lost to only good opp, Maryland's 4-1 a sham

There is a surprising amount of red, most of it being for SEC teams (5 of 11) which is common as everyone thinks the SEC is far and away the toughest conference.  But they play no one out of conference so this is mostly perception and it is hard to say for sure and the computer says they aren't as good as perceived.

Other conferences have teams going both ways, the Big-12 with Nebraska and Missouri too low, OU and OSU too high,  and the Big Televen has tOSU, MSU, and Wisconsin too high with Iowa too low.

The Pac-10 though is all too low with Stanford, Arizona, and Oregon St, typical bias against the Pac-10. It is also interesting to note how the coaches are more biased against the Pac-10 than the writers with the Coaches lower rankings for those teams highlighted in purple above.  Especially Oregon State, with 2 losses to undefeated opponents, not ranked by the coaches?  The beneficiary of the coaches bias is the Big-12 highlighted in blue.


2010 Week 6 College Football Projections; Nebraska, TCU, Oregon, tOSU undefeated

Here are the projected records for the regular season using results through week 6 of the season.  Full projections on the site.

And then there were 13.  We all know who the top undefeated teams are, and my computer has six of them in the top-6 of the rankings, but there are 7 more behind them going all the way back to #29 Oklahoma State.  But who will remain undefeated?

The current projections say only four teams will, Nebraska, TCU, Oregon, and Ohio State.  What happens to the other nine?  Here is the breakdown for all 13.

First the four projected to finish that way.

12 wins - 57%
11 wins - 34%
10 wins - 8%
9 wins - 1%

12 - 54%
11 - 43%
10 - 2%

12 - 48%
11 - 40%
10 - 11%
9 - 1%

Ohio State
12 - 47%
11 - 46%
10 - 6%

All have a good to very good chance of completing the season with no losses, but single loss is still possible and for Ohio State and Oregon, some number of losses is still more likely than undefeated.

What about the others?

Boise State
12 - 37%
11 - 50%
10 - 12%
9 - 1%

13 - 38%
12 - 48%
11 - 13%
10 - 1%

Boise State and Nevada have to play each other which accounts for a lot of why each team may end up with one loss.  But whoever wins that game will most likely finish a perfect season.

12 - 9%
11 - 30%
10 - 35%
9 - 19%
8 - 5%

Oklahoma still has Missouri, Texas A&M, and Oklahoma State plus other games that aren't gimmies.

12 - 2%
11 - 26%
10 - 40%
9 - 24%
8 - 6%

Missouri has A&M, Oklahoma, and Nebraska still to play.

12 - 9%
11 - 29%
10 - 36%
9 - 20%
8 - 5%

Utah must play TCU, Notre Dame, and Air Force.

12 - 14%
11 - 39%
10 - 35%
9 - 11%

LSU still has Auburn, Alabama, and Arkansas.

Michigan State
12 - 19%
11 - 63%
10 - 16%
9 - 1%

MSU has Iowa as the main competition with no tOSU on the schedule, but also closes with PSU.

12 - 6%
11 - 32%
10 - 40%
9 - 18%
8 - 3%

Auburn has Arkansas LSU, and Alabama.

Oklahoma State
12 - 1%
11 - 6%
10 - 20%
9 - 33%
8 - 27%
7 - 12%
6 - 2%

OSU must still play Texas Tech, Nebraska, K-State, Texas, and Oklahoma.  There's a tough closing schedule.

What do you think?  Make a comment.

2010 Week 6 College Football Ratings and Rankings; Oregon #1, Nebraska #2, Boise St #3

And Oregon stays #1.  But there is a new #2 and #3 with Nebraska and Boise State making big moves up.  Ohio State rolls and stays #4 and TCU and Nevada make small moves up to #5 and #6.

Nebraska makes the move based on their stomping of previously undefeated K-State and Boise State does it based partially on their convincing win, but more based on previous opponents VaTech still improving and Oregon State knocking off previously undefeated Arizona.

To understand this, it can be helpful to look at the performance charts for the teams.

Oregon's #1 is based on them playing their best in their two most meaningful games (larger dots) against ASU and Stanford.

Nebraska has also played well in their meaningful games.

Boise State has been very consistent and like noted above is being helped by Va Tech and Oregon State improving since they played.

And Ohio State has been very consistent, but doesn't have a really big result to pull them higher.

Interested in charts for all the teams?  Subscribe to the newsletter!

Full rankings on the site, but top-30 below for quick reference.

1Oregon92.9196-065.027+0, -3.170
2Nebraska92.1395-063.664+3, +4.996
3Boise St87.4835-067.561+7, +3.017
4Ohio State86.6726-063.035+0, -1.407
5TCU86.4986-065.066+2, +0.847
6Nevada85.0386-062.957+5, +1.730
7Stanford84.6455-173.588-4, -4.327
8California82.9213-272.875+7, +3.076
9Alabama82.7715-170.165-7, -9.841
10Iowa82.6064-161.547-4, -4.471
11Florida St82.2585-167.847+16, +6.234
12Oklahoma82.1135-071.379+4, +2.504
13Arizona81.9954-168.439-5, -3.241
14South Carolina81.7274-169.233+15, +6.262
15Missouri81.4755-063.382+2, +2.128
16Utah81.2835-059.345+25, +8.481
17LSU80.4106-070.254+3, +2.141
18Michigan St79.8916-062.342+20, +6.807
19Arizona St79.5973-373.086+11, +4.532
20Virginia Tech79.3824-269.765-6, -1.109
21Oregon St79.1943-280.247+15, +5.552
22Arkansas78.7444-165.913-9, -2.208
23Auburn78.2926-066.150+0, +1.705
24Miami FL77.9303-270.518-15, -6.589
25North Carolina St77.7915-165.256-1, +1.339
26Air Force77.5525-162.609-1, +1.258
27Southern Cal76.7404-269.896+12, +3.776
28Florida76.3874-268.507-16, -6.473
29Oklahoma St75.6535-061.592-3, -0.480
30Mississippi St75.5484-264.945+15, +4.223

Sunday, October 3, 2010

2010 Week 5 College Football Projections; Alabama, TCU, and Oregon Undefeated

Now things get interesting part 2!  With the shake-up in the rankings, the regular season projected records are a bit of a surprise.  Here is what my computer projects and below are the 1-loss and undefeated teams.

And there are still 3 projected to finish undefeated but they are now Alabama, TCU, and Oregon, all with a very good chance of it happening (91%, 75%, and 69%).  Nevada (41%), Ohio State (36%), Boise State (30%), and Nebraska (26%) all have a good shot at finishing there too. 

TeamRecord% chance% +1% -1
TeamRecord% chance% +1% -1
Miami FL11-
Ohio State11-
Boise St11-150.829.817.0
San Diego St10-243.93.438.6
North Carolina St10-239.616.232.0
Kansas St10-233.713.333.6