Sunday, October 31, 2010

College Football Top 25 10/31

We lost a couple more undefeated teams this week, and there are now only 5 left. 3 of them sit atop the standings, but it's not the three you might expect. Here's this week's top 25:

Rank Team W L Rating SOS
1 TCU 9 0 606.24 30.91
2 Oregon 8 0 526.21 41.04
3 Boise St 7 0 483.11 30.94
4 Alabama 7 1 454.32 58.69
5 Stanford 7 1 395.61 55.18
6 Ohio State 8 1 382.07 30.42
7 Missouri 7 1 358.17 67.45
8 Nebraska 7 1 330.13 53.4
9 Iowa 6 2 326.21 46.13
10 Oklahoma 7 1 288.85 76.21
11 Utah 8 0 285.21 21.64
12 Arizona 7 1 254.08 46.79
13 Oregon St 4 3 231.62 175.86
14 Auburn 9 0 220.7 65.6
15 Arkansas 6 2 199.91 56.47
16 Florida St 6 2 193.64 44.51
17 Southern Cal 5 3 192.53 85.95
18 Oklahoma St 7 1 191.85 49.24
19 Virginia Tech 6 2 185.81 38.95
20 South Carolina 6 2 185.49 64.54
21 California 4 4 168.05 80.35
22 Florida 5 3 164.57 82.55
23 Illinois 5 3 163.45 62
24 Texas A&M 5 3 161.47 51.94
25 LSU 7 1 158.3 56.87

TCU still reigns atop (I've already talked a couple of times about how they actually are the best team in the nation), closely followed by Oregon and Boise St. Alabama is still the best one loss team in the nation, and Iowa is the best with two losses. Auburn still lags behind the other undefeateds, but will get at least another week's pass before their real challenge begins. Either Georgia, Alabama, or somebody in the SEC championship game will knock them off.
Other teams of interest include:
32 Michigan St 8 1 130.81 46.23
33 Wisconsin 7 1 119.38 39.88
41 Texas 4 4 82.15 65.03
59 Washington 3 5 48.61 151.14
73 Northwestern 6 2 37.44 19.04
426 Georgetown DC 3 6 0.52 0.79

Texas and Northwestern are now in serious bowl hunts. The Wildcats in particular may be 6-2, but will be underdogs in all four of their remaining games. It'll be a fight to keep Jimmy out of Detroit this year.

For a look ahead at the BCS standings, here's my no MoV top 10:
Rank Team W L Rating SOS
1 Auburn 9 0 32.26 8.06
2 Missouri 7 1 29.87 10.56
3 Oklahoma 7 1 27.66 9.78
4 TCU 9 0 25.7 6.42
5 LSU 7 1 24.71 8.73
6 Boise St 7 0 23.52 5.88
7 Nebraska 7 1 22.11 7.81
8 Oregon 8 0 21.64 5.41
9 Alabama 7 1 20.46 7.23
10 Utah 8 0 20.06 5.01

Auburn looks again like they'll top the BCs, with Oregon close behind still lacking the computer ratings (although they are 2nd in Sagarin's Elo rank this week). Missouri may have lost, but they're not dropping too far because they lost to a very good Nebraska team. While the Tigers may be out of the title hunt, they still play a roll. If they maintain their current record they could stay ahead of Boise St providing cushion for other 1 loss teams like Alabama or Oklahoma who may move up with more difficult wins. Once again, thanks to poor human polling and a week schedule, the best team in the nation, the TCU Horned Frogs, would be on the outside looking in if the championship happened today.

Saturday, October 30, 2010

Game of October 30th

Will this Halloween's undeadfeated have the blood sucked out of their title runs, or will they rise again this weekend? Predictions for a spooky Saturday:

Iowa 29 Michigan St 23
If the Hawkeyes really are a top level team, this Saturday is their time to show it.

Nebraska 27 Missouri 26
The Tigers big win last week was at home. This time they have to travel.

Stanford 40 Washington 27
It would be nice if the Cardinal's cornerbacks showed up this week, but either way the offense should get the job done.

Northwestern 28 Indiana 26
The Iniana Hoosiers got no D, but can the Wildcat O take full advantage?

Oregon 37 USC 30
Some think USC can win, and the Trojans are good. But Oregon is really, really good.

Fordham 28 Georgetown 16
A win here would definitely improve the G-town football's chances of survival this off-season. However, unless the early season Hoyas show up it looks unlikely.

Last Week: 5-1 (the Hawks let me down)
Season: 19-12

How many for wins the Heat?

It's been covered, a lot. Everybody up through John Hollinger has projected the number of wins for the Miami Heat this season, and I have no secret methods, but I'm interested and there need to be some basketball articles for the NBA season. I looked at it two different ways. First, I added up last season's Win Shares of all the Heat players. Lebron had 18.5, Dwayne Wade had 13, and Chris Bosh had 9. Impressively, Udonis Haslem had 6 win shares last year (he's a solid defender), but that's where the help stops. For the season the Heat project to have a total of 67.1 win shares.
Some other historic teams:
08-09 Lakers: 60.8 win shares, 65 wins.
08-09 Cavaliers: 68.0 win shares, 66 wins.
07-08 Celtics: 70.2 win shares, 66 wins.
99-00 Lakers: 63.3 win shares, 67 wins.
96-97 Bulls: 71.3 win shares, 69 wins.
95-96 Bulls: 75.3 win shares, 72 wins.
It appears that win shares do a good job of predicting wi- Wait, THIS WON'T BE LEBRON'S BEST TEAM?!?! That's right, the 08-09 Cavaliers had 68 win shares, .9 more than this year's Heat project to have. The weak bench behind the Three Amigos will cost them this year.
The other way to analyze the Heat's upcoming performance is to look at their projected points. Last season they averaged 96.5 ppg, and gave up 94.2. Using Bill James's pythagorean method (winning percentage = ppg^14/(oppg^14 + ppg^14)), and updating projected ppg totals, it's possible to project win total. Adding Lebron and Bosh is a huge advantage here, bringing in 916 and 513 points each according to John Hollinger. The pythagorean method projects the Heat at 71.97 wins. One key difference between the two is Chris Bosh. He ranks the 8th best NBA player in Hollingers Value Added, but only the 19th best by win shares. Either way, high 60s is a good win projection for the Miami Heat this season. I like win shares better so I'll say the Miami Heat get 67 wins this season.
I'd also appreciate possible nicknames for this year's Heat. Andy, this is your time to shine.

Who Makes The BCS?

With an insight into non-MoV rankings, it's possible to predict what it would take to get top level teams Boise St or TCU into the BCS championship game, despite the system. Here's how a few teams would rank if they finished the season with the projected records and their opponents maintained roughly consistent ratings:
29- 2 loss Oklahoma
28- 1 loss LSU, undefeated Oregon
26- undefeated Boise St, 1 loss Alabama
25- undefeated TCU, 2 loss Missouri
Okay, so Oklahoma is a little overrated (the Big 12 as a whole. The same thing happened with the Pac 10 last year.) Voters would also never put a 2 loss team over undefeated teams (unless it was Utah). The predictions are supported though by inputting games into the Colley Matrix, a ranking used by the BCS that usually overrates undefeated teams. I would rate team's chance to get into the title game as follows:

1. Any of the undefeated BCS schools (Auburn, Oregon, Missouri, Michigan St. in order).
2. A one loss Big 12 champion (Oklahoma, Nebraska, Missouri, maybe even Oklahoma St.).
3. Undefeated Boise St.
4. A one loss SEC champion (Alabama, Auburn, LSU).
5. Undefeated TCU.
6. Anybody else.

The shame here is that while TCU may be the best team in the country, they likely won't get a shot at a national championship. One of Alabama, Auburn, or LSU will likely end the season with just one loss, and there will probably be another undefeated team. Too bad a playoff would take too much student athlete academic time. Over winter break.

Sunday, October 24, 2010

College Football Top 25 10/24

Many people say that the number 1 team lost this week, but here at Kobe, Tell Me How My Stats Taste we saw the Missouri "upset" coming. But there are a few changes up top, and other teams have solidified their ranking.

Rank Team W L Rating SOS
1 TCU 8 0 602.51 34.45
2 Oregon 7 0 529.32 35.69
3 Boise St 6 0 491.91 29.78
4 Missouri 7 0 468.61 58.05
5 Alabama 7 1 449.01 58
6 Ohio State 7 1 384.28 36.27
7 Utah 7 0 347.57 19.41
8 Stanford 6 1 335.04 61.5
9 Nebraska 6 1 329.63 49.22
10 Arizona 6 1 319.09 50.98
11 California 4 3 288.72 82.32
12 Oklahoma 6 1 260.65 89.48
13 Iowa 5 2 242.98 41.21
14 Southern Cal 5 2 231.64 74.04
15 Auburn 8 0 220.96 67.89
16 South Carolina 5 2 216.92 76.42
17 Florida St 6 1 213.72 37.92
18 Oklahoma St 6 1 212.41 51.55
19 Virginia Tech 6 2 194.03 40.67
20 Miami FL 5 2 189.32 59.49
21 Michigan St 8 0 181.61 37.64
22 Texas A&M 4 3 174.16 58.16
23 LSU 7 1 173.94 62.49
24 Arkansas 5 2 168.4 62.26
25 Florida 4 3 166.39 78.21

Up top again is TCU which extended it's lead this week. If you're not convinced by the Horned Frogs yet, check out their 38-7 win over Air Force. Air Force has been ranked by the AP off and on this season, and only lost to Oklahoma by 2 points, yet TCU destroyed them. Sounds like something the best team in the nation might do. Unfortunately for TCU, voters don't realize that, and are okay letting a now over rated Auburn team jump them in the polls.
There are a couple of other surprises in these rankings. One is the California Golden Bears, who somehow show up at number 11. Honestly though, nobody knows what to make of them, they can't decide whether to kick butt or not show up. Apparently my computer thinks they're good.
Also surprising are the Auburn Tigers ranked down at number 15, even though they're likely to be either 1 or 2 in the BCS this week. Why are they so low? Citing 3 point wins over Kentucky, Mississippi, and Clemson (in overtime!) I'd ask why they're so highly ranked (LSU's not that good).
A few other teams of interest:
30 Wisconsin 7 1 131.38 43.89
42 Texas 4 3 89.49 62.1
53 Iowa St 4 4 60.35 98.5
54 Washington 3 4 60.23 126.3
64 Northwestern 5 2 46.72 22.48
438 Georgetown DC 3 5 0.43 0.65

Why is Wisconsin ranked so low? A fairly weak schedule, and close wins (two by 1 point) are to blame. I'd expect another loss or two from them by the end of the year, even with a mediocre schedule.
Looking into the BCS this week, we can estimate the computer rankings based off the no MoV rankings. Here's the Top 10:

Rank Team W L Rating SOS
1 Missouri 7 0 38.24 9.56
2 Auburn 8 0 32.27 8.06
3 Oklahoma 6 1 30.27 11.24
4 TCU 8 0 28.23 7.05
5 LSU 7 1 23.96 8.47
6 Boise St 6 0 23.65 5.91
7 Michigan St 8 0 22.08 5.52
8 Oregon 7 0 19.55 4.88
9 Nebraska 6 1 18.75 6.96
10 Alabama 7 1 18.62 6.58

Missouri will get rewarded for their win over Oklahoma, and Auburn also sneaks up near the top. Interesting to note is Oregon's low ranking, which I expect to be true in most computer polls, possibly costing them the BCS number 1. Also important is TCU's 4th place ranking. It looks possible that the BCS won't even give the best team a chance this year.

Saturday, October 23, 2010

Weekend's Games

Again, it's late, but still, it's unbiased. Here's this week's predictions, and I'm feeling like they're better than last week.

Iowa 29 Wisconsin 20
John Clay may have his hands full trying to deal with the Hawkeye defense.

Stanford 48 Washington St 13
Really? The Cougars are a Pac-10 team? Perfect game for this week's homecoming.

Michigan St 35 Northwestern 27
Not really. Actually I predicted Michigan St 28 Northwestern 15, which isn't too far off.

Missouri 30 Oklahoma 24
Upset Alert! This week's "number 1" is also in danger of going down.

Auburn 36 LSU 28
More SEC battles? When will it stop! There's no way both overrated teams can lose here, but LSU still can finally.

Sacred Heart 22 Georgetown 21
Well, I got this one right already. Which is good news for me, bad news for Allison (my sister).

Last Week: 2-4, a disappointment
Season: 14-11

Friday, October 22, 2010

Physics of Sports- Jump Balls


As a physics major, I thought it might be interesting to introduce some physics concepts. Also, I have the sweet picture above which needed to make it in here. That fool in red in the middle is me, clearly getting skied on by second team All-American and player of the year candidate Nneka Ogwumike.
The weird thing about this photo though, is that I actually won the tip. It should be a pretty even match, my 6'4" height and 24 inch vertical against her 6'2" and 30. This particular tip went my way though because I timed my jump better, she went up to early, and is already going down in this picture as I'm going up. Fortunately, thanks to physics, I knew when to time my jump:
Beware, this makes heavy use of one equation: 4.9*t^2 = x where t is time from stop and x is distance object has fallen. Hopefully I have derived it right, or else I've made myself a fool for all the internet to see.

Say the ball is thrown about 13 feet in the air (a reasonable estimate). Being my height, with my ups (which aren't quite as high when there's somebody jumping against you) I would probably best tip the ball at around 10 feet plus a few inches off the ground so I can get a solid touch on it. To jump two feet off the ground would take .35 seconds:
4.9*t^2 = .6 meters, t = .35 seconds
For the ball to drop those 3 feet from it's apex takes .41 seconds:
4.9*t^2 = .82 meters, t = .41 seconds
That means to time the jump, I have to leave .06 seconds after the ball reaches the top, or when it's 1.8 centimeters below the top.
4.9*.06^2 = x, x = .018 meters = 1.8 cm
Given that when already in position, it takes about .2 seconds to go from deciding to jump to actually leaving the ground, I actually want to try to jump .14 seconds before the ball peaks. That's 9.6 cm below the top (you know the equation by now). Any before that and I'll fall before I get to the ball. Any after that, and I won't get the ball at my peak.
So with a little help from physics, I won the tip. Then I returned to my usual duties of taking elbows in the gut for the sake of Stanford Women's Basketball. While wearing my new free basketball shoes of course.

Sunday, October 17, 2010

Random Chance in the NFL

Normally over the course of an NFL season power rankings regress towards the mean and extreme ratings get closer to 1. This season however, the best and worse ratings are already closer to 1 than last season's final ratings, and we haven't had any undefeated teams for a couple of weeks. Has NFL parity really increased?
One way to judge parity might be to look at the percentage of time the better team wins. With no parity, the best team in a given game would win 100% of the time, being far better than it's opponent. But with perfect parity the better team would win only 50% of the time, and NFL game would be random chance. It's obvious that chance plays some role because otherwise the Rams would never win a game, but how much of a role it plays can be uncertain. To look at that, I'm going to look at what percentage of the time the best team in a game wins.
The key to my search was to look at variance in win numbers across the NFL. If all teams won 50% of the time, then winning percentages would be clustered around .500, giving a small variance. On the other hand, if the best team always one there would be undefeated and winless teams along with other extreme records, leading to a high variance.
Through 5 weeks in the NFL, variance per team was 1.24 wins per game. That matches up with the better team winning 62% of the time. To confirm, I looked back at the previous games to see how often the higher ranked team won. Turns out, the higher ranked team won 63% of the time, so it's pretty clear that in the NFL this year the better team wins somewhere around 62% of the time.
For comparison, I also looked at the last three years by variance. In 2009 the better team won 78% of the time, and in both 2008 and 2007 the better team won 80% of the time. If the Chiefs and Bears hadn't already made it clear, statistics show that this year parity and random chance have reigned over the NFL, so much so that picking winners is basically a crapshoot. At least that explains why my rankings systems can't pick NFL games. Also, it may give the Bears a chance at the playoffs. I kinda like this random chance thing.

College Football Top 25 10/17

A tumultuous Saturday! For the second week in a row, the number one AP team lost, along with multiple other undefeated or highly ranked teams. How do the power rankings shake out now? Did Alabama maintain it's hold on the top spot? Here's the new top 25:

Rank Team W L Rating SOS
1 TCU 7 0 524.22 31.3
2 Missouri 6 0 490.84 48.4
3 Stanford 5 1 473.64 74.72
4 Boise St 6 0 449.27 27.2
5 Oregon 6 0 419.57 34.53
6 Alabama 6 1 403.09 59.61
7 Nebraska 5 1 319.86 39
8 Ohio State 6 1 303.76 35.69
9 Oklahoma 6 0 300.61 76.6
10 Utah 6 0 258.74 17.08
11 South Carolina 4 2 237.86 90.85
12 Iowa 5 1 237.3 29.55
13 Arizona 5 1 232.79 43.07
14 Oklahoma St 6 0 218.07 37.29
15 Auburn 7 0 216.93 61.84
16 Southern Cal 5 2 216.06 69.06
17 Michigan St 7 0 208.68 37.87
18 LSU 7 0 205.74 58.16
19 Florida St 6 1 202.71 35.96
20 California 3 3 195.18 72.59
21 Arizona St 3 3 180.81 72.79
22 Arkansas 4 2 171.01 64.79
23 Georgia 3 4 170.42 62.36
24 Florida 4 3 170.24 80.02
25 Texas A&M 3 3 165.08 80.38

With all the commotion up top, it's the team that has routinely taken care of business at number one, TCU. Their 31-3 handling of BYU showed that despite a weak schedule, they're a force to fear. Alabama managed to drop all the way to number 6 due to a weak win over Ole Miss, while surprising Missouri moved up towards the top. Next week the Tigers can prove their for real with a Big 12 showdown against Oklahoma.

Some other interesting teams:
37 Texas 4 2 110.13 64.93
48 Washington 3 3 73.02 110.18
62 Northwestern 5 1 46.5 17.79
417 Georgetown DC 3 4 0.52 0.71

For fun, here's how the BCS compatible top 25 shake out:
Rank Team W L Rating SOS
1 Oklahoma 6 0 43 10.75
2 Missouri 6 0 33.51 8.37
3 LSU 7 0 31.46 7.86
4 Boise St 6 0 27.34 6.83
5 TCU 7 0 27.21 6.8
6 Auburn 7 0 27 6.75
7 Oklahoma St 6 0 22.75 5.68
8 Michigan St 7 0 22.36 5.59
9 Stanford 5 1 20.63 8.18
10 Oregon 6 0 20.32 5.08
11 Alabama 6 1 18.73 6.95
12 Utah 6 0 18.26 4.56
13 Kansas St 5 1 17.88 7.09
14 Texas 4 2 17.33 10.91
15 Arizona 5 1 17.09 6.78
16 Florida St 6 1 16.45 6.11
17 Nevada 6 1 16.28 6.05
18 Nebraska 5 1 15.62 6.2
19 South Carolina 4 2 15.02 9.46
20 Wisconsin 6 1 14.03 5.21
21 Iowa 5 1 13.87 5.5
22 Air Force 5 2 13.82 7.63
23 Mississippi St 5 2 13.75 7.59
24 Ohio State 6 1 13.54 5.03
25 Southern Cal 5 2 13.11 7.24

Looks like Oklahoma vs Missouri may be the game of the week next Saturday. But more intriguing are Boise St and TCU ranked 4 and 5. With just a couple of the right bounces we could see a non-BCS team in the championship game this year. As there should be, per the first rankings.

Saturday, October 16, 2010

Games of October 16th-predictions

Some of these may already have happened, but I was busy winning at IM flag football last night (3-0! Don't worry, I'll post rankings). It's just math though, so there's no bias. Here are the games:

Iowa 32 Michigan 23
Hawkeyes should take this one in the Big House, but it's not looking great so far.

Nebraska 35 Texas 9
Lou Holtz and I agree, one of these teams is really good, the other not so much.

Auburn 28 Arkansas 25
Two weeks ago the SEC battled for the best team, last week it was a game for second. Is the winner here the third best? Auburn's home field puts them overtop.

Oregon St 35 Washington 28
The Huskies will keep it closer than expected, but a win might be asking too much.

Georgetown 23 Bucknell 7
This is already in progress though, and the Hoyas need a comeback to prevent Bucknell's first win of the season.

Ohio State 31 Wisconsin 21
Although they're an undeserving number 1, the Buckeyes should still win this one.

Last Week: 3-3
Season: 12-7

Tuesday, October 12, 2010

BCS Rankings and Top Units

Next week the BCS will officially be released (although it can already be calculated easily). How are the computer rankings going to look, and can Boise St. really be number 1? To analyze this I am debuting my margin of victory free rankings (meaning they follow the BCS rules). Here's the top 10 (top 50 are under the new tab up top).

Rank Team W L Rating SOS
1 Oklahoma 5 0 42.5 10.62
2 Boise St 5 0 39.51 9.87
3 LSU 6 0 38.88 9.72
4 Missouri 5 0 33.97 8.49
5 TCU 6 0 31.83 7.95
6 Auburn 6 0 25.26 6.31
7 Nevada 6 0 24.62 6.15
8 South Carolina 4 1 23.9 10.4
9 Oregon 6 0 23.05 5.76
10 Oregon St 3 2 22.11 16.76

Surprisingly, this ranking is very different than the MOV rankings, but doesn't necessarily benefit just those with perceived strong schedules. Boise St actually gets quite a bump (explaining why they're BCS number 1 this week) as does Nevada. The biggest boost though goes to number one Oklahoma (up from 23). In the power rankings their close wins over teams like Air Force and Cincinnati anchored the Sooners around 25. Close wins don't make a difference without MOV though, and that sends the Sooners shooting to the top. Also benefitting is Oregon St, who may be just 3-2 but has clearly played the toughest schedule in the nation under both rankings.
Some notable teams don't fare as well though. Here's just a few, with some explanation:
11 Alabama - May have played four games against top 25 teams, but the other two were Duke and San Jose St., lowering their strength of schedule.
19 Ohio State - Their SOS is just slightly lower than Stephen F. Austin's. But they at least have Wisconsin up next week.
32 Iowa - Weak opponents such as Ball State and Eastern Illinois hurt their rankings, but with three tough games coming up their rank could swing in a hurry.

Also for fun, here are the top five units in college football by points line rating:
Offenses
1. Oregon
2. Oklahoma St
3. Boise St
4. Nevada
5. Stanford

Defenses
1. Arizona
2. Alabama
3. TCU
4. Missouri
5. California