Sagarin ELO CHESS 9/29/2013

[WhiteMamba]

He has Bama#1, not 3.LMAO@ Daffy Ducks' 104 SoS.

Oregon's schedule is backloaded. 3 highly ranked teams.

#5 Stanford, # 15 Washington, # 13 UCLA, and 3 other teams receiving votes are left on schedule.

Bama has 2 ranked teams on the schedule all season as of today. #9 and #11 and 2 receiving votes.

And I am thinking Bama's SOS will take a prettty good hit with games vs Tennessee (common opponent), Kentucky, Chattanooga, and Georgia State left on the schedule, call me crazy... But that isnt that impressive man.

Bama's remaining schedule is a cake walk man.

Georgia State
@ Kentucky
Arkansas
Tennessee
bye
LSU
@ Miss State
THE Chattanooga Mocs
@ Auburn

and you are LOLing @ schedules?

 

[4down20]

Oregon's schedule is backloaded. 3 highly ranked teams.

#5 Stanford, # 15 Washington, # 13 UCLA, and 3 other teams receiving votes are left on schedule.

Bama has 2 ranked teams on the schedule all season as of today. #9 and #11 and 2 receiving votes.

And I am thinking Bama's SOS will take a prettty good hit with games vs Tennessee (common opponent), Kentucky, Chattanooga, and Georgia State left on the schedule, call me crazy... But that isnt that impressive man.

Bama's remaining schedule is a cake walk man.

Georgia State
@ Kentucky
Arkansas
Tennessee
bye
LSU
@ Miss State
THE Chattanooga Mocs
@ Auburn

and you are LOLing @ schedules?

Alabama has already played 2 top 25 teams, possibly 3. 3 of those teams have only 1 loss, each to Alabama. Probably one of the tougher schedules thus far, with only a few other SEC teams being able to say otherwise(Georgia/LSU). Still has 1, maybe 2 more to go minimum.

Your fellow Oregon fans started throwing these rocks without regards to their own glass houses, not us.

 

[Bamabino]

No. The ELO-Chess pretty definitively lists Bama as #3.

I see the difference now.

 

[WhiteMamba]

Alabama has already played 2 top 25 teams, possibly 3. 3 of those teams have only 1 loss, each to Alabama. Probably one of the tougher schedules thus far, with only a few other SEC teams being able to say otherwise(Georgia/LSU). Still has 1, maybe 2 more to go minimum.

Your fellow Oregon fans started throwing these rocks without regards to their own glass houses, not us.

And one of your fellow Bama brothers (although retarded) is who I was responding too.

 

[FSUmanager]

FSU will probably win, but if they play like they did against B.C., the Terps can easily pull off the upset.

Should have left the bolded word out of there.

 

[4down20]

His methodology is ridiculous.

West Virginia is 3-2 and was blown out 37-0 by Maryland.

Yet, West Virginia moves up because of the win over Oklahoma State and Maryland is undefeated and unranked, even though they destroyed West Virginia. Won't matter anyway, Florida State should get rid of the Maryland mystique next week.

The thing with computers is they don't buy into hype, they don't make assumptions, they just follow the data, and the patterns and trends the data shows. This early in the year computers are going to think some crazy things because the data is limited, and many teams have yet to play a quality game, which gives false impressions about them. Once we get deeper into conference play things will start to clear up, and things will get decided.

Then you have quality wins and so on.

Predictive systems also look at games they expect a team to win in the future as well, not just the ones they have won.

 

[NativeWebfoot]

Based solely on the games played so far, OR St should be listed above Utah. The Beavs went to Utah and beat them at their house. May have been ugly, but a win nevertheless and the Beavers are now 4-1, 2-0. OR St may not win 9 or more this season, but the computer shouldn't be looking at anything more than what's been played so far.

 

[4down20]

Based solely on the games played so far, OR St should be listed above Utah. The Beavs went to Utah and beat them at their house. May have been ugly, but a win nevertheless and the Beavers are now 4-1, 2-0. OR St may not win 9 or more this season, but the computer shouldn't be looking at anything more than what's been played so far.

FCS.

 

[Codaxx]

FCS.

-
Yes.. You will see Oregon #1 in some polls that throw out FCS. I really would not pay too much attention to computers until about week 7. You need at least 5 data pts to get anything close to reasonable. Once again the people are complaining are actually upset that the computers do not have a SEC Bias.

 

[Codaxx]

here is a computer model I follow. It has been pretty reasonable. There are still some odd results here (tenn??):

1 Oregon 2 Baylor 3 Clemson 4 Alabama 5 Miami (FL) 6 Louisville 7 Washington 8 Georgia 9 Florida 10 Florida St. 11 Stanford 12 Texas Tech 13 UCLA 14 Ohio St. 15 LSU 16 Northwestern 17 Maryland 18 Virginia Tech 19 South Carolina 20 Tennessee 21 Illinois 22 Texas A&M 23 Ohio 24 Arizona St. 25 Central Florida

 

[occupant]

here is a computer model I follow. It has been pretty reasonable. There are still some odd results here (tenn??):

1 Oregon 2 Baylor 3 Clemson 4 Alabama 5 Miami (FL) 6 Louisville 7 Washington 8 Georgia 9 Florida 10 Florida St. 11 Stanford 12 Texas Tech 13 UCLA 14 Ohio St. 15 LSU 16 Northwestern 17 Maryland 18 Virginia Tech 19 South Carolina 20 Tennessee 21 Illinois 22 Texas A&M 23 Ohio 24 Arizona St. 25 Central Florida

It's been well established that ADJUSTEDSTATS.COM uses Fisher Price computers.

 

[Codaxx]

It's been well established that ADJUSTEDSTATS.COM uses Fisher Price computers.

-
it has been a pretty solid tool for betting. served me well.

 

[NativeWebfoot]

FCS.

Just saying "FCS" doesn't explain why a computer that is only supposed to be comparing games played to this point doesn't have OR St in it's top 25. All the teams have been playing FCS teams and OR St lost their opener to the #2 FCS. Not the cellar dwellars that everyone else has been beating. All OR ST's other opponents were FBS and 2 of them IC. You can try telling me how bad Utah and Colo are IC, but some teams on the list haven't played anyone better. The computer program isn't supposed to be looking ahead or at popularity, or conf affiliation. The computer isn't supposed to be able to discern favorites or which teams are supposed to finish in the NCG or the BCS bowl games.

 

[4down20]

It's been well established that ADJUSTEDSTATS.COM uses Fisher Price computers.

Codex is right, it's too early to put too much faith in what the computers say. It's all based on limited data at this point and due to the large number of teams and differences between teams it takes a little while for them to become accurate.

5 data points is what I like to have personally, and even then it's far from perfect and will have oddities. 5 is just a start and good overall. But by the end of the year they are usually pretty spot on, provided the formula is good.

A little common sense and understanding goes a long way when looking at computer models at this point.

 

[4down20]

Just saying "FCS" doesn't explain why a computer that is only supposed to be comparing games played to this point doesn't have OR St in it's top 25. All the teams have been playing FCS teams and OR St lost their opener to the #2 FCS. Not the cellar dwellars that everyone else has been beating. All OR ST's other opponents were FBS and 2 of them IC. You can try telling me how bad Utah and Colo are IC, but some teams on the list haven't played anyone better. The computer program isn't supposed to be looking ahead or at popularity, or conf affiliation. The computer isn't supposed to be able to discern favorites or which teams are supposed to finish in the NCG or the BCS bowl games.

You lost to a FCS team. You're only win of note so far this year is against Utah, which isn't a very big note. Utah has beaten BYU and Utah St so far, which is a step up for them, but it's not like they are really great wins either. The FCS loss by far outweighs that Utah win, and the rest of your wins are not quality wins by any stretch.

Furthermore, you likely have maybe 3 more wins left in the season. You will probably beat Washington St and Cal. You will likely beat USC, but that's not really a given. I would guess you should be favored. Meanwhile, you will likely lose to: Stanford, Oregon, Arizona St and Washington. Leaving you with an overall record of 7-5, bowl eligible and in the mid 30's for rank.

As for computers, I disagree almost completely. I don't know of any that use popularity or conference affiliation, but predictive models are usually the better ones. At any rate, if Oregon St proves to be better and gets more wins, it will show up and they will be recognized for it. But right now, not so much. I have Oregon St going 7-5 right now, and that's not even taking the FCS loss into account in terms of future predictions.

 

[Codaxx]

You lost to a FCS team. You're only win of note so far this year is against Utah, which isn't a very big note. Utah has beaten BYU and Utah St so far, which is a step up for them, but it's not like they are really great wins either. The FCS loss by far outweighs that Utah win, and the rest of your wins are not quality wins by any stretch.

Furthermore, you likely have maybe 3 more wins left in the season. You will probably beat Washington St and Cal. You will likely beat USC, but that's not really a given. I would guess you should be favored. Meanwhile, you will likely lose to: Stanford, Oregon, Arizona St and Washington. Leaving you with an overall record of 7-5, bowl eligible and in the mid 30's for rank.

As for computers, I disagree almost completely. I don't know of any that use popularity or conference affiliation, but predictive models are usually the better ones. At any rate, if Oregon St proves to be better and gets more wins, it will show up and they will be recognized for it. But right now, not so much. I have Oregon St going 7-5 right now, and that's not even taking the FCS loss into account in terms of future predictions.

There are some computers that do have bias. I believe I read that Sagarin puts some human inputs in for the first few weeks to try to "clean up" the numbers. I think a few others do that for the first few weeks.

 

[4down20]

There are some computers that do have bias. I believe I read that Sagarin puts some human inputs in for the first few weeks to try to "clean up" the numbers. I think a few others do that for the first few weeks.

Possible, they have to do something otherwise it's basically impossible to do early rankings. Personally I just use previous years stats, and I'm guessing those human inputs eventually go away as do my use of last years stats once there is reliable data to fall upon.

 

[NativeWebfoot]

4D20, you're missing the point and going back to your biased human viewpoint again. Alabama is supposed to be the fav and get the #1-3 ranking based on expectations, which are based on prev years finishes. Who have THEY played so far this season? The computer program is NOT supposed to look at anything but wins and losses THIS season so far. Getting beaten by a top FCS in a close game should count for moire than beating a bottom FCS (equivalent of a div 3)by 40. How many of the teams listed are unbeaten? Were their losses to top FCS or above average FBS teams? The computer can only compare data given. Faulty data in produces faulty results.

 

[Codaxx]

4D20, you're missing the point and going back to your biased human viewpoint again. Alabama is supposed to be the fav and get the #1-3 ranking based on expectations, which are based on prev years finishes. Who have THEY played so far this season? The computer program is NOT supposed to look at anything but wins and losses THIS season so far. Getting beaten by a top FCS in a close game should count for moire than beating a bottom FCS (equivalent of a div 3)by 40. How many of the teams listed are unbeaten? Were their losses to top FCS or above average FBS teams? The computer can only compare data given. Faulty data in produces faulty results.

They played 2 bowls teams from last yr in A&M and Ole Miss. SEC starting their conference schedule early certainly helps in the early polls. I am happy to slam Bama, but I do not think a schedule of VT, A&M, Ole Miss, and Cupcake U is anything but solid.

 

[4down20]

4D20, you're missing the point and going back to your biased human viewpoint again. Alabama is supposed to be the fav and get the #1-3 ranking based on expectations, which are based on prev years finishes. Who have THEY played so far this season? The computer program is NOT supposed to look at anything but wins and losses THIS season so far. Getting beaten by a top FCS in a close game should count for moire than beating a bottom FCS (equivalent of a div 3)by 40. How many of the teams listed are unbeaten? Were their losses to top FCS or above average FBS teams? The computer can only compare data given. Faulty data in produces faulty results.

Who has Alabama played? Are you joking? They have some of the most impressive wins of the season so far. 2 ranked teams beaten already, possibly 3 if you count VT. They have 0 FCS wins, and they have had 1 cupcake game so far in the 1st 4.

The problem with Alabama's schedule in terms of strength is that future games are a good bit easier and many of the tougher games are behind. Still got LSU and possibly the SECCG, but relatively easy until then unless Arkansas, Auburn or Miss St start becoming world beaters.

The only team that can really make any claim to having an equal schedule so far is Georgia, who dropped the game to Clemson or would likely be #1 for many. Maybe Notre Dame with having already played Michigan and Oklahoma, but they dropped both those.

I really don't know where you get your ideas about what computers are "supposed" to do.