2021 Pre-Season Probability Poll Discussion Thread

coit

Smarter than Diseqc
Joined
Nov 29, 2007
Messages
71,485
The PPP is back for 2021, baybee! The season is just about 6 weeks away, and I think most all of us (with some notable exceptions) are excited for the season to get underway! As in past years (other than 2020), I will be tracking everyone's performance through the season and handing out kudos at the end for top performers, so be sure to vote in every poll if you want to participate in that.

Links to each week's poll will be added here as each game's poll is released:

TBD

---------------

In previous seasons we've developed some scoring systems that attempt to incentivize picks that are close to the actual result. We have taken to calling these the TIAScore and the GGTScore weighting factors. These systems are described below.

The @ThisIsAtlanta / @GoGATech Scoring Systems

TIA Score Weighting:


For a Win, your score equals your win probability.
For a Loss, your score equals 1 minus your win probability.
  • If we win a game you predicted as 0.9 you get 0.9
  • If we lose a game you predicted as 0.9 you get 0.1
  • If we lose a game you predicted as 0.1 you get 0.6
  • If we win a game you predicted as 0.1 you get 0.1
This rewards those who realistically vote low probabilities for the games we end up losing and those who vote high probabilities for the games we end up winning.


GoGATech / 18in32 Scoring Modification:

Score starts with the TIA scoring described above. Each game is given a result value according to the following index, which is based on the point spread of the final score:
  • 0.4 = three possession (>16 pt) win
  • 0.3 = two TD (>11 pt) win
  • 0.2 = two possession (>8 pt) win
  • 0.1 = TD (>3 pt) win
  • 0.0 = win or loss by < 3 pts
  • -0.1 = TD (>3 pt) loss
  • -0.2 = two possession (>8 pt) loss
  • -0.3 = two TD (>11 pt) loss
  • -0.4 = three possession (>16 pt) loss
The modified score is calculated by multiplying the TIA score by 1 minus the absolute value of each vote probability minus 0.5 minus the game result value above.

GGTScore = TIAScore * (1 - abs(VoteProb - 0.5 - GameResultValue)

This rewards vote probabilities that are closer to final outcome by giving them a higher weight than results that are further from the actual final outcome.

The final result of these two weighting systems produces a score that reasonably reflects how close to the actual result a voter's predictions are on a game by game and seasonal basis.

If you'd like to reminisce about past polls, links are below.

2019 Results Thread: https://stingtalk.com/board/threads/2019-preseason-probability-poll-results.101870/
2018 Results Thread: https://stingtalk.com/board/threads/2018-preseason-probability-poll-results.97791/
 

coit

Smarter than Diseqc
Joined
Nov 29, 2007
Messages
71,485
Ok, I'm putting the finishing touches on the schedule for publishing the probability polls now, and really just now taking a hard look at our schedule. Wow.

Using the ESPN FPI data:
  • Our FPI is positive for the first time in awhile!
  • #2 schedule strength
  • 6 straight games to finish the season, the final 4 against 3 top 10 FPIs
By any measure, it will be a brutal season, but I'm sure many of you know this already.

So, here's my plans to get the games published for voting:

Game 01 - Aug 1
Game 02 - Aug 3
Game 03 - Aug 5
Game 04 - Aug 8
Game 05 - Aug 10
Game 06 - Aug 12
Game 07 - Aug 15
Game 08 - Aug 17
Game 09 - Aug 19
Game 10 - Aug 22
Game 11 - Aug 24
Game 12 - Aug 26

I'll leave them open for 6 days each, so hopefully everyone will have time to catch up on their voting.
 

ee8384

Cynic At-Large
Joined
Jan 8, 2005
Messages
11,121
The PPP is back for 2021, baybee! The season is just about 6 weeks away, and I think most all of us (with some notable exceptions) are excited for the season to get underway! As in past years (other than 2020), I will be tracking everyone's performance through the season and handing out kudos at the end for top performers, so be sure to vote in every poll if you want to participate in that.

Links to each week's poll will be added here as each game's poll is released:

TBD

---------------

In previous seasons we've developed some scoring systems that attempt to incentivize picks that are close to the actual result. We have taken to calling these the TIAScore and the GGTScore weighting factors. These systems are described below.

The @ThisIsAtlanta / @GoGATech Scoring Systems

TIA Score Weighting:


For a Win, your score equals your win probability.
For a Loss, your score equals 1 minus your win probability.
  • If we win a game you predicted as 0.9 you get 0.9
  • If we lose a game you predicted as 0.9 you get 0.1
  • If we lose a game you predicted as 0.1 you get 0.6
  • If we win a game you predicted as 0.1 you get 0.1
This rewards those who realistically vote low probabilities for the games we end up losing and those who vote high probabilities for the games we end up winning.


GoGATech / 18in32 Scoring Modification:

Score starts with the TIA scoring described above. Each game is given a result value according to the following index, which is based on the point spread of the final score:
  • 0.4 = three possession (>16 pt) win
  • 0.3 = two TD (>11 pt) win
  • 0.2 = two possession (>8 pt) win
  • 0.1 = TD (>3 pt) win
  • 0.0 = win or loss by < 3 pts
  • -0.1 = TD (>3 pt) loss
  • -0.2 = two possession (>8 pt) loss
  • -0.3 = two TD (>11 pt) loss
  • -0.4 = three possession (>16 pt) loss
The modified score is calculated by multiplying the TIA score by 1 minus the absolute value of each vote probability minus 0.5 minus the game result value above.

GGTScore = TIAScore * (1 - abs(VoteProb - 0.5 - GameResultValue)

This rewards vote probabilities that are closer to final outcome by giving them a higher weight than results that are further from the actual final outcome.

The final result of these two weighting systems produces a score that reasonably reflects how close to the actual result a voter's predictions are on a game by game and seasonal basis.

If you'd like to reminisce about past polls, links are below.

2019 Results Thread: https://stingtalk.com/board/threads/2019-preseason-probability-poll-results.101870/
2018 Results Thread: https://stingtalk.com/board/threads/2018-preseason-probability-poll-results.97791/
Is there a link to the predictions so we can be sure that we got them all in?
 

coit

Smarter than Diseqc
Joined
Nov 29, 2007
Messages
71,485
Is there a link to the predictions so we can be sure that we got them all in?
I'm working on the spreadsheet now. You are 100% thru the Miami game. I'll get it set so folks can submit any that they missed in this thread before I finalize.
 

ee8384

Cynic At-Large
Joined
Jan 8, 2005
Messages
11,121
I'm working on the spreadsheet now. You are 100% thru the Miami game. I'll get it set so folks can submit any that they missed in this thread before I finalize.
I thought I did them all.

Just clicked through them and have check marks by all, including BC, ND and Gaggers.
 

coit

Smarter than Diseqc
Joined
Nov 29, 2007
Messages
71,485
Ok folks, here's the results of this year's polls. The contest is only for folks that have picked results for every game, so if you want to participate, take a look at your votes and reply here with your picks for missing polls.

All picks must be submitted by 9 AM Saturday morning.

 

BrentwoodJacket

Dodd-Like
Joined
Dec 24, 2007
Messages
4,217
How is the season predictions spreadsheet set up? The column for Northern Illinois does not look right,
 

mookie43

Flats Noob
Joined
Dec 6, 2006
Messages
979
@coit My voting results tab line doesn't match my seasons results tab line. Please advise
 

1982Jacket

Jolly Good Fellow
Joined
Sep 26, 2020
Messages
1,883
I was an IM major* so I can't follow all that complicated KenPom style math. I'll just have 0.5's across the board.


* OK, jk. BEE '82 with highest honors. Still . . .
 

coit

Smarter than Diseqc
Joined
Nov 29, 2007
Messages
71,485
I did this in a rush, so let me check and I'll update everyone.
 
Top