Page 1 of 1

Laxpower versus the Human Pollsters, 2007 Edition

PostPosted: Wed Apr 18, 2007 3:08 pm
by Dan Wishengrad
Laxpower Top 25 (Poll Rank):

1. BYU (1) 84.90
2. Oregon (3) 84.32
3. Colorado State (2) 83.46
4. Minnesota-Duluth (8.) 82.20
5. Michigan (5) 81.36
6. Colorado (6) 80.46
7. Sonoma State (7) 80.21
8. Northeastern (9) 80.10
9. Arizona (12) 80.08
10. UC Santa Barbara (10) 79.60
11. Arizona State (4) 79.39
12. Simon Fraser (22) 79.10
13. Chapman (NR) 79.07
14. Georgia (17) 78.82
15. Boston College (21) 78.51
16. Florida State (11) 78.29
17. Utah (13) 77.92
18. Lindenwood (14) 77.08
19. Loyola-Marymount (NR) 76.79
20. Michigan State (20) 76.14
21. Texas A&M (16) 76.06
22. Florida (18) 75.70
23. Virginia Tech (19) 75.67
24. Buffalo (NR) 75.64
25. Cal-Poly (15) 75.53

next five:
26. Claremont (24) 75.43
27. Georgia Tech (23) 74.93
28. UC Davis
29. Oregon State
30. Texas

and the lone ranked team lower:
34. Illinois (25) 73.47

Some notes on discrepancies between computer rankings and poll rankings:

Since the previous poll was announced, Arizona beat Arizona State 17-10. Laxpower reflects this game but the poll does not. ASU fell six spots in the P.R.s after that loss, while UA jumped at least four or five spots.

Chapman is not eligible for poll votes, but Laxpower ranks the team regardless.

Boston College's win over Northeastern is reflected on Laxpower, but not yet in the polls. BC and especially Simon Fraser are the most underrated team in the polls, according to the computer, which also says Buffalo should be ranked. The computer also says UMD and Florida State are over-ranked by the humans but that Georgia is under-ranked. The computer also agrees with the pollsters on quite a few teams with identical rankings.

Personally I find an analysis of the computer numbers to always be interesting and a debate might be fun here. As a voter I always study the computer rankings and personally believe -- in my own humble opinion -- that Laxpower's numbers are pretty damn good in 2007. I will be happy to read from dissenters as well those who concur here.

PostPosted: Wed Apr 18, 2007 3:15 pm
by unclaxman23
i would agree that the computer polls are pretty good, i just think that the sos is kinda messed up. i think that the computer should not just average the opposition. For example unc had one of the hardest schedule and the computer does not show that.

Re: Laxpower versus the Human Pollsters, 2007 Edition

PostPosted: Wed Apr 18, 2007 3:44 pm
by Dulax31
Dan Wishengrad wrote:Laxpower Top 25 (Poll Rank):

4. Minnesota-Duluth (8.) 82.20

"The computer also says UMD and Florida State are over-ranked by the humans"


Ummm, the computer says UMD is under-ranked by the pollsters (humans)....

Re: Laxpower versus the Human Pollsters, 2007 Edition

PostPosted: Wed Apr 18, 2007 4:48 pm
by Dan Wishengrad
Dulax31 wrote:
Dan Wishengrad wrote:Laxpower Top 25 (Poll Rank):

4. Minnesota-Duluth (8.) 82.20

"The computer also says UMD and Florida State are over-ranked by the humans"


Ummm, the computer says UMD is under-ranked by the pollsters (humans)....


Yep, sorry, stated that incorrectly... too much data in without sufficient memory... and my computer's slow also! :lol:

PostPosted: Wed Apr 18, 2007 5:39 pm
by CATLAX MAN
I think our pollsters do a much better job of ranking the teams consistently than the Laxpower computers do. Their rankings are very unreliable because computers don't have common sense, judgement and the ability to interpret results.

Re: Laxpower versus the Human Pollsters, 2007 Edition

PostPosted: Thu Apr 19, 2007 1:08 am
by Dr. Jason Stockton
Dan Wishengrad wrote:Laxpower Top 25 (Poll Rank):

12. Simon Fraser (22) 79.10

BC and especially Simon Fraser are the most underrated team in the polls, according to the computer, which also says Buffalo should be ranked.


Obviously the computer has no anti-Canada bias. . . :lol:

Re: Laxpower versus the Human Pollsters, 2007 Edition

PostPosted: Thu Apr 19, 2007 9:57 am
by jlax
Dan Wishengrad wrote:Laxpower Top 25 (Poll Rank):
Personally I find an analysis of the computer numbers to always be interesting and a debate might be fun here. As a voter I always study the computer rankings and personally believe -- in my own humble opinion -- that Laxpower\\\'s numbers are pretty damn good in 2007. I will be happy to read from dissenters as well those who concur here.


Dan,
I appreciate your effort to move SFU up in the ranking to secure an at-large when Oregon blows them out again. I can also understand your effort to get Oregon State into the rankings. (You are reaching) SFU has played no one except Oregon. I could see if it were close but a 9 goal loss to the only ranked team they play gets SFU ranked? I have a feeling some voters are ranking SFU much higher than they should be to get them into the polls. SFU should not be ranked. Nine games with only two out of conference and 1 ranked team. I thought A teams had to have 3 ooc games to be ranked and get a at-large.

SFU 15 Oregon State 10
SFU 9 Oregon 18
Idaho 2 SFU 29
Gonzaga 8 SFU 18
Boise State 7 SFU 22
Montana 13 SFU 21
Rhode Island 6 SFU 14
Texas Tech 11 SFU 23
Washington 14 SFU 20

I hope voters are not letting Laxpower dictate how they vote.

PostPosted: Thu Apr 19, 2007 10:57 am
by x1dschm
First of all I am not advocating for a change to a computer system.
however, I do think that as the year goes on the laxpower rankings get better and better as the computers have more and more data to work with. Also they don't carry an east coast west coast bias or US versus everyone else bias. The computer system has many faults, but the human system isn't perfect either.

Re: Laxpower versus the Human Pollsters, 2007 Edition

PostPosted: Thu Apr 19, 2007 11:11 am
by Dan Wishengrad
jlax wrote:Dan,
I appreciate your effort to move SFU up in the ranking to secure an at-large when Oregon blows them out again. I can also understand your effort to get Oregon State into the rankings. (You are reaching) ... SFU should not be ranked. Nine games with only two out of conference and 1 ranked team. I thought A teams had to have 3 ooc games to be ranked and get a at-large. ...
I hope voters are not letting Laxpower dictate how they vote.


Well, I asked for comments! Let me respond.

First, who is letting Laxpower "dictate" how they vote? How did we move from a discussion comparing Laxpower to the human polls to the notion that Laxpower is somehow "gospel"? There is a big difference between reviewing a source of information, like Laxpower, and relying on it to shape one's vote. But I would argue that a voter who doesn't even bother studying Laxpower -- as an available source of info -- is doing himself and the MCLA a disservice.

Second, I made no attempt to promote a team, to "get Oregon State ranked" or some such nonsense. If I wanted to vote for the Beavers I remain free to do so. But I have NOT done so in 2007, however, because I don't feel that OSU's record this year warrants such a vote.

Finally, as regards SFU, you are partially correct on the facts but certainly entitled to your opinion even when you have your facts wrong. No, Simon Fraser is not eleigible for an at-large bid to the MCLA tourney as they have not played the required three OOC games. But yes, they ARE indeed eligible to be ranked and can go to the tourney if they win the PNCLL's A.Q.. Yes, I think SFU is a pretty good team, one deserving of a higher ranking than they currently hold (#22). The computer agrees. You don't have to, you are free to think they are terrible if you want. Nor does there have to be consensus among us pollsters. We cast our votes and Sonny tallies them up. But it always dismays me to read that any team with a tough schedule deserves a high ranking while a team with an easier schedule doesn't even deserve to be ranked. This is nonsense, like all that talk about Boise State not deserving a BCS bowl game because of the team's weak schedule. Well, BSU sure proved they deserved the game by beating Oklahoma. A team with a weaker schedule might not be as well prepared to compete at nationals, but it doesn't mean they don't belong there. It just makes it tougher to rank such teams.

It is obvious that most of my colleagues don't agree with my ranking of SFU. This is perfectly acceptable, and does not bother me in the slightest. One of the reasons that Sonny expanded the poll to 40 voters was to mitigate some of the regional biases. I merely point out that the computer has SFU as one of the more under-rated teams in the MCLA this year. You can read whatever you like into this, including what's not being said. It's a free country!

PostPosted: Thu Apr 19, 2007 11:23 am
by SDSULAX
You do not have to play any OOC games to be ranked, you have to play the appropriate number of OOC games to be eligible to get an At-Large bid to the National Championship games.

Re: Laxpower versus the Human Pollsters, 2007 Edition

PostPosted: Mon Apr 23, 2007 5:42 pm
by nhoskins
jlax wrote:
Dan Wishengrad wrote:Laxpower Top 25 (Poll Rank):
Personally I find an analysis of the computer numbers to always be interesting and a debate might be fun here. As a voter I always study the computer rankings and personally believe -- in my own humble opinion -- that Laxpower\\\'s numbers are pretty damn good in 2007. I will be happy to read from dissenters as well those who concur here.


Dan,
I appreciate your effort to move SFU up in the ranking to secure an at-large when Oregon blows them out again. I can also understand your effort to get Oregon State into the rankings. (You are reaching) SFU has played no one except Oregon. I could see if it were close but a 9 goal loss to the only ranked team they play gets SFU ranked? I have a feeling some voters are ranking SFU much higher than they should be to get them into the polls. SFU should not be ranked. Nine games with only two out of conference and 1 ranked team. I thought A teams had to have 3 ooc games to be ranked and get a at-large.

SFU 15 Oregon State 10
SFU 9 Oregon 18
Idaho 2 SFU 29
Gonzaga 8 SFU 18
Boise State 7 SFU 22
Montana 13 SFU 21
Rhode Island 6 SFU 14
Texas Tech 11 SFU 23
Washington 14 SFU 20

I hope voters are not letting Laxpower dictate how they vote.


As much as I'd like to think Dan is a big-time SFU fan, he hasn't ordered himself an SFU hoodie or t-shirt yet. And while supporting teams in your conference happens, I don't think Dan or any of the other Dawgs are biased towards SFU.

I do however think this SFU team is under-rated, but as jlax pointed out "If you wanna be the man, you've gotta beat the man."

SFU has the opportunity to do that this weekend, if they can fight through a tough OSU squad for a rematch with the Ducks. I'll be there to see it, and if the pollster's can forgive CSU for a bad loss (UO 18-6), I'm sure SFU is capable of bouncing back from a similar loss (UO 18-9).