Posts of the Week
Each week, we will highlight a few entertaining and/or informative posts from the previous week. Please keep in mind that it is hard to keep track/prioritize all posts so we would welcome input from all Booties. You can make a "Bootie Selection" post as a response to any post that you deem worthy or you can email a link (to the nominated post) to me at firstname.lastname@example.org.
Below are the posts that made this week's list:
Poster: Opinionated Crank
Subject: Mercenaries (note correct spelling)
It may indeed be true that Bill Walsh fired up the Husky fans
for life with his remark. It may be true that Stanford will never
win in Seattle during my lifetime. But it is also true that Walsh
was absolutely correct. The last time I saw Bootleg stats,
Washington was dead last in the Pac-10 in graduation rate for
football players: 35%. And every single one of them redshirts as
a freshman. That's what the "mercenaries" remark was
about, not whatever NCAA sanctions the UDub fans keep whining
about. Washington owns Stanford since I came to the farm; they
"do it to us doggie style" repeatedly (usually run up
the score), but no matter how many times they do it, they are
still a football factory with a bunch of semipro mercenaries
playing for them. They have no class and never will.
Walsh was right.
Poster: 51 Dave
Subject: "Upon Further Review Of the Tape".....UCLA
1. K. Harris had a coming out party on Sat. Coleman will be in
the NFL and Harris gave him everything he wanted.
2. Kick coverage- Huge improvement from UO game. Some wicked hits on KO returns. Better on punts, as well.
3. As good as our interior guys played on the OL, Craig A. and Travis P. were almost as impressive on the DL. Very active and aggressive. Craig's best game, I thought. Tipton deserves a lot of credit for what's been accomplished with these guys.
4. Newberry got a beautiful kick out block on Allen's TD run. He'll continue to help us down the road.
5. I knew that Paul W. took a lot of snaps for Schindler on Sat. He also played for Eric H. on one of our long drives in the first half.
6. Officiating- We got some breaks. At UO, we got screwed. Sat., the Int. for Ryan F., the out of bounds call on Ryan Smith's catch both looked like blown calls. Pray that Jim Springer's crew (from the UO game) doesn't call the UW battle.
7. Coaching- One of the best jobs by the staff in TW's tenure. A smart aggressive defensive game plan. Great clock management in the 2nd half. BD did a terrific job of giving Chris Lewis a winnable game plan. The new OL coaches are making a difference. Seems to be more of a "take no prisoners" attitude this year.
8. Teyo needs the ball more versus UW.
9. Emotion swings- UCLA damn near pulled this one out but I got the feeling that a stop on either of the 4th and 1's in the Bruins first TD drive of the 3rd quarter might have turned this into a huge route. We actually played excellent defense on that drive, BTW. To the Bruins credit, they fought their way back into it but one more early stop and, believe it or not, something like 45-14 wouldn't have been out of the question. Just shows how quick things can change.
Subject: U of Washington Criminal Law 101, test question
Define a "minor offense."
Most common answer given:
Any violation of a law that has been committed 77 times or more.
Poster: Boston Card
Subject: Food for thought (long)
Some here (myself included) have suggested that the polls are
biased based on preseason expectations. I decided to test this
hypothesis. I compared the difference between the average of the
two major polls on Sunday versus the average of computer polls
calculated for the BCS (which throws out the best and worst poll)
with the difference between the average of the preseason polls
and the average of the current polls. In cases where the teams
were not in the top 25 in the preseason voting, I just used the
"others receiving votes" to sequentially rank teams.
Teams that didn't have any "other votes" were given a
ranking of the total number of teams that received at least 1
vote plus 1 (which overestimates their rank somewhat). What I
would predict is that those teams who were lower in the preseason
poll (compared to their current poll) would also show a gap
between how the computers rank them and how the pollsters do.
What I found was quite striking. The two teams that display the biggest difference between the computer and poll averages, Stanford and WSU, were unranked at the start of the season (WSU didn't even receive any votes). In fact of the seven teams that are ranked higher by the computers than by polls, only one, Oregon was ranked higher in the preseason than now, and the difference is minimal on both counts (computer - poll = 0.17 and preseason - now = 1).
On the flip side, Florida, ranked #1 in the preseason and currently basking in a 3.5 average poll, has an average computer ranking of 10.58, a difference between computers and humans of over 7. Virginia Tech, likewise is ranked 20.17 by the computers and 12 by the polls, but was ranked 9 in the preseason polls.
The one team that doesn't fit the pattern? BYU, unranked at the beginning of the season (42 in the coaches poll), but currently ranked 10 and 11 by the polls, yet only 17.83 by the computers. Guess the pollsters are impressed by the fact they are unbeaten. Miami ias also ranked better now than in the preseason, though by a single position.
Another hypothesis is that pollsters don't take strength of schedule into account enough when making polls, penalizing Stanford for losing once while playing the toughest schedule, while rewarding BYU and Miami for being undefeated, with cupcake schedules. (The flip side could argue that computers reward too much for SOS). This line also trends out quite well. As mentioned, Stanford with the toughest schedule in the mix also has the greatest difference gap between polls and computer rankings while BYU, with the easiest, has the second biggest "excess poll ranking" (behind Florida). 4 out of the 6 teams with the toughest schedules have gaps between computers and polls (the two others have essentially no difference between the computers and the polls). Of the 5 teams with the easiest schedules (ranked worse than 40), 4 have a better poll ranking than computer ranking (3 by more than 2 places, 2 by more than 4 places). The lone exception? WSU, which was discussed above.
The last thing I looked at was whether certain conferences tended to outpoll the computers. I must warn that the data set is limited here, since there are only 2 Big East and SEC teams and only 1 Mountain West team. However, it will surprise no one that the Pac-10 has the biggest gap between the polls and the computers with a mean difference of almost 3 slots (2.988 to be precise), mostly because of Stanford and WSU (though all but UW contribute). The other underrated conference is the Big-10 (with a gap of 1.92). Of the major conferences (the Mountain West doesn't count), the most overrated (no surprise here) is the Big East, which has poll numbers that are on average 5 better than its computer rankings. The SEC is next with an average difference of 3.625 between polls and computer rankings. The Big 12, by the way, is fairly neutral. It has an average difference between polls and computer rankings of 0.5. You can judge for yourself whether that spells "east coast bias".