Author Topic: CPU Predictos (Week 5)  (Read 5611 times)

0 Members and 1 Guest are viewing this topic.

Offline ew2x4

  • Pak'r Élitaire
  • ****
  • Posts: 3918
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #25 on: October 04, 2010, 06:46:47 PM »

The42Yardstick

  • Guest
Re: CPU Predictos (Week 5)
« Reply #26 on: October 04, 2010, 06:53:28 PM »


Using this too much.

This one was sarcastic. Did you listen to the podcast?

Offline CHONGS

  • The Producer
  • Administrator
  • Pak'r Élitaire
  • *****
  • Posts: 20114
    • View Profile
    • goEMAW.com
Re: CPU Predictos (Week 5)
« Reply #27 on: October 04, 2010, 08:16:56 PM »
Ok here's the "real" answer:

The margin of error for this game is pretty large relatively speaking, it is only 73% certain of a NU victory.  So with ~95% certainty the game will have a spread of +/- 11 (the cat's can win by 2 or lose by 20).

Offline Rage Against the McKee

  • Pak'r Élitaire
  • ****
  • Posts: 37959
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #28 on: October 04, 2010, 10:10:33 PM »
Ok here's the "real" answer:

The margin of error for this game is pretty large relatively speaking, it is only 73% certain of a NU victory.  So with ~95% certainty the game will have a spread of +/- 11 (the cat's can win by 2 or lose by 20).

I'm sure you're right, but it seems kind of odd that K-State has a 27% chance of winning by 1 or more, but only  ~5% chance of winning by 3 or more. I guess I'll just take the 27% and  :cheese:

Offline steve dave

  • Global Moderator
  • Pak'r Élitaire
  • *****
  • Posts: 88406
  • Romantic Fist Attachment
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #29 on: October 05, 2010, 08:47:08 AM »
I'LL PUNCH YOUR LIGHTS OUT!

Offline Dr Rick Daris

  • Global Moderator
  • Pak'r Élitaire
  • *****
  • Posts: 23381
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #30 on: October 05, 2010, 09:20:15 AM »
he's just making this stuff up as he goes along folks. you really just can't beat this.


btw,
i used the pythagorean theorem and one data set of points to the third power and ran last weeks weeks scores and came up with the prediction that chingon is a total dumbass. confidence level of 95% with margin of error plus or minus three. take it for what it's worth folks.




 :flush:
« Last Edit: October 05, 2010, 09:30:24 AM by Rick Daris »

Offline CHONGS

  • The Producer
  • Administrator
  • Pak'r Élitaire
  • *****
  • Posts: 20114
    • View Profile
    • goEMAW.com
Re: CPU Predictos (Week 5)
« Reply #31 on: October 05, 2010, 10:44:03 AM »
he's just making this stuff up as he goes along folks. you really just can't beat this.


btw,
i used the pythagorean theorem and one data set of points to the third power and ran last weeks weeks scores and came up with the prediction that chingon is a total dumbass. confidence level of 95% with margin of error plus or minus three. take it for what it's worth folks.




 :flush:
Here's a CPU:

Rick Daris gets his face stomped Thursday before the game. I know where you and your ugly little friends will be.  Your mouth just wrote a check even your fat ass can't cover.

Offline PoetWarrior

  • Katpak'r
  • ***
  • Posts: 2353
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #32 on: October 05, 2010, 11:44:15 AM »
I have a prediction:

The outcome of this game will largely be determined by the environment in which it is played.


Does this system take these factors into account?

Offline CHONGS

  • The Producer
  • Administrator
  • Pak'r Élitaire
  • *****
  • Posts: 20114
    • View Profile
    • goEMAW.com
Re: CPU Predictos (Week 5)
« Reply #33 on: October 05, 2010, 11:44:44 AM »
I have a prediction:

The outcome of this game will largely be determined by the environment in which it is played.


Does this system take these factors into account?
IT DOES!

Offline PoetWarrior

  • Katpak'r
  • ***
  • Posts: 2353
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #34 on: October 05, 2010, 11:49:57 AM »
It should be very accurate then.


But that assumes the outcome of a football game could be predicted, by some means, which is absolutely ridiculous.

Offline Dr Rick Daris

  • Global Moderator
  • Pak'r Élitaire
  • *****
  • Posts: 23381
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #35 on: October 05, 2010, 11:50:48 AM »
he's just making this stuff up as he goes along folks. you really just can't beat this.


btw,
i used the pythagorean theorem and one data set of points to the third power and ran last weeks weeks scores and came up with the prediction that chingon is a total dumbass. confidence level of 95% with margin of error plus or minus three. take it for what it's worth folks.




 :flush:
Here's a CPU:

Rick Daris gets his face stomped Thursday before the game. I know where you and your ugly little friends will be.  Your mouth just wrote a check even your fat ass can't cover.

 :pray:

Offline CHONGS

  • The Producer
  • Administrator
  • Pak'r Élitaire
  • *****
  • Posts: 20114
    • View Profile
    • goEMAW.com
Re: CPU Predictos (Week 5)
« Reply #36 on: October 05, 2010, 11:54:27 AM »
It should be very accurate then.


But that assumes the outcome of a football game could be predicted, by some means, which is absolutely ridiculous.
Not really.  It won't be 100% accurate, but it can give you a substantially better than uninformed prediction.

Offline kso_FAN

  • Global Moderator
  • Pak'r Élitaire
  • *****
  • Posts: 29506
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #37 on: October 05, 2010, 11:55:23 AM »
Skipper44 asked a very good question in a previous thread (IMHO); how does the CPU take into account special teams play?

TIA.

Offline CHONGS

  • The Producer
  • Administrator
  • Pak'r Élitaire
  • *****
  • Posts: 20114
    • View Profile
    • goEMAW.com
Re: CPU Predictos (Week 5)
« Reply #38 on: October 05, 2010, 12:05:06 PM »
Skipper44 asked a very good question in a previous thread (IMHO); how does the CPU take into account special teams play?

TIA.
Sorry about that, it slipped my mind.

Short answer:  it is indirectly accounted for in the Def and Off efficiencies. 

Longer answer:  Since yardage is NOT the rubric used to compare offenses (or defenses) and ONLY  adjusted points scored per play, a score on a special teams play is equal to a score on a regular offensive play (as well as an interception on a defensive play).

Offline kso_FAN

  • Global Moderator
  • Pak'r Élitaire
  • *****
  • Posts: 29506
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #39 on: October 05, 2010, 12:09:06 PM »
Skipper44 asked a very good question in a previous thread (IMHO); how does the CPU take into account special teams play?

TIA.
Sorry about that, it slipped my mind.

Short answer:  it is indirectly accounted for in the Def and Off efficiencies. 

Longer answer:  Since yardage is NOT the rubric used to compare offenses (or defenses) and ONLY  adjusted points scored per play, a score on a special teams play is equal to a score on a regular offensive play (as well as an interception on a defensive play).

That makes sense.  It would be interesting to see the numbers run on adjusted points per special teams play.  Probably a ton of more work for you though.

Offline slimz

  • PCKK7DC Survivor
  • Katpak'r
  • *******
  • Posts: 2128
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #40 on: October 05, 2010, 01:47:41 PM »
It appears the CPU is used to set the Vegas lines and O/U: 12 point spread, 52 O/U.

Offline steve dave

  • Global Moderator
  • Pak'r Élitaire
  • *****
  • Posts: 88406
  • Romantic Fist Attachment
    • View Profile
Re: CPU Predictos (Week 5)
« Reply #41 on: October 05, 2010, 01:50:55 PM »
Skipper44 asked a very good question in a previous thread (IMHO); how does the CPU take into account special teams play?

TIA.
Sorry about that, it slipped my mind.

Short answer:  it is indirectly accounted for in the Def and Off efficiencies. 

Longer answer:  Since yardage is NOT the rubric used to compare offenses (or defenses) and ONLY  adjusted points scored per play, a score on a special teams play is equal to a score on a regular offensive play (as well as an interception on a defensive play).

That makes sense.  It would be interesting to see the numbers run on adjusted points per special teams play.  Probably a ton of more work for you though.

It would be interesting to see the numbers run by someone who wasn't A SQWUAK HUSKER NAZI!

The42Yardstick

  • Guest
Re: CPU Predictos (Week 5)
« Reply #42 on: October 05, 2010, 01:52:31 PM »
Skipper44 asked a very good question in a previous thread (IMHO); how does the CPU take into account special teams play?

TIA.
Sorry about that, it slipped my mind.

Short answer:  it is indirectly accounted for in the Def and Off efficiencies. 

Longer answer:  Since yardage is NOT the rubric used to compare offenses (or defenses) and ONLY  adjusted points scored per play, a score on a special teams play is equal to a score on a regular offensive play (as well as an interception on a defensive play).

That makes sense.  It would be interesting to see the numbers run on adjusted points per special teams play.  Probably a ton of more work for you though.

It would be interesting to see the numbers run by someone who wasn't A SQWUAK HUSKER NAZI!


Are you going to want to fire Snyder if we win?