Where is Gauss? - Printable Version
+- The CardBoard (https://thecardboard.org/board)
+-- Forum: C-House! (https://thecardboard.org/board/forum-4.html)
+--- Forum: The CardBoard (https://thecardboard.org/board/forum-5.html)
+--- Thread: Where is Gauss? (/thread-9838.html)
+- The CardBoard (https://thecardboard.org/board)
+-- Forum: C-House! (https://thecardboard.org/board/forum-4.html)
+--- Forum: The CardBoard (https://thecardboard.org/board/forum-5.html)
+--- Thread: Where is Gauss? (/thread-9838.html)
Where is Gauss? - BostonCard - 09-14-2015
Someone needs to keep score so that we know who to award the fabulous and expensive prizes for winning. Since Gauss is MIA, I tallied the prediction results up for the UCF game, and when I have time will do the same for the NW game. As I recall, the rules were you get one point for each point difference you had between your predicted score and the actual score. Points are awarded based on the number of players you did better (so if there are 39 entrants and you were the second best prediction you get 38 points). I wasn't sure how Gauss dealt with ties so I awarded the highest number of points eligible (that is, if two people were tied for first place, then they each got 39 points if there were 39 entrants). I recall a penalty being applied for guessing the wrong winner, but I don;t know how many points that was. Regardless no one guessed wrong this week.
Without further ado, here is the tally
Stanford UCF Error Modified Error Points
Trumpcard 30 10 4 5 39
martyup 30 10 4 5 39
Publius 30 13 7 9.5 37
Renaissance an 31 14 7 10.5 37
Big Brother 27 10 7 7.5 37
Las Vegas 32 13 7 8.5 37
Kathy 27 10 7 7.5 37
Papa John 26 10 8 9 32
Extra Point 28 13 9 10.5 31
washingtonismoney 28 13 9 10.5 31
ThePassionOfTheChryst 38 10 10 9 29
Durty Nelly 27 13 10 11 29
leftcoast 35 13 10 7 29
CTcard 31 17 10 15 29
Hank 91 27 13 10 11 29
thunder_chik 24 13 13 13.5 24
pcgoode 28 17 13 16.5 24
2 for 2028 38 13 13 7.5 24
amartinsu13 27 17 14 17 23
pefloresjr 27 17 14 17 23
BigEasyCard 24 14 14 14 23
82lsju 19 10 15 19.5 18
GoodGrief 42 3 15 18.5 18
Brannerite 21 12 15 17.5 18
Mick 38 17 17 11.5 15
Braniac_Infinite 24 17 17 18.5 15
ferrari 24 17 17 18.5 15
hurlburt88 24 17 17 18.5 15
PrinceLightfoot 26 20 18 22 11
Winflop 24 19 19 21.5 10
CrazedZooChimp 45 13 20 18 9
CornFed 41 17 20 10 9
garvin 17 14 21 24.5 7
Boston Card 20 17 21 21.5 7
fullmetal 16 13 21 25.5 7
SUBuddha 35 24 21 23.5 7
NYCCardinalFan 13 10 21 28.5 7
SeattleTree 9 7 22 33 2
SamAtoms1980 48 13 23 22.5 1
==========================
Median 27 13 10 11
Actual Score 31 7 0 0
In case people are wondering, the modified error takes a slightly different calculation and is intended to reward people for getting the spread right even if the individual scores were off. Basically, instead of abs(predicted Stan - actual Stan) + abs(predicted UCSF + actual UCF), the formula is abs(predicted sum of scores - actual sum of scores)/2 + abs(predicted difference in scores - actual difference in scores).
It doesn't make a brobdingnagian difference as you guys can see, but it would help someone who maybe predicted Stanford to win 24-0 (nailing the difference) over someone who predicted 24-14. In both cases, they were off by a touchdown for each team, so the old system would have given them 14 points. But by the new system, the second score is penalized for missing the difference by 14 points.
BC
Re: Where is Gauss? - Rally - 09-14-2015
(09-14-2015, 08:37 PM)Boston Card link Wrote:Someone needs to keep score so that we know who to award the fabulous and expensive prizes for winning. Since Gauss is MIA, I tallied the prediction results up for the UCF game, and when I have time will do the same for the NW game. As I recall, the rules were you get one point for each point difference you had between your predicted score and the actual score. Points are awarded based on the number of players you did better (so if there are 39 entrants and you were the second best prediction you get 38 points). I wasn't sure how Gauss dealt with ties so I awarded the highest number of points eligible (that is, if two people were tied for first place, then they each got 39 points if there were 39 entrants). I recall a penalty being applied for guessing the wrong winner, but I don;t know how many points that was. Regardless no one guessed wrong this week.
Without further ado, here is the tally
Stanford UCF Error Modified Error Points
Trumpcard 30 10 4 5 39
martyup 30 10 4 5 39
Publius 30 13 7 9.5 37
Renaissance an 31 14 7 10.5 37
Big Brother 27 10 7 7.5 37
Las Vegas 32 13 7 8.5 37
Kathy 27 10 7 7.5 37
Papa John 26 10 8 9 32
Extra Point 28 13 9 10.5 31
washingtonismoney 28 13 9 10.5 31
ThePassionOfTheChryst 38 10 10 9 29
Durty Nelly 27 13 10 11 29
leftcoast 35 13 10 7 29
CTcard 31 17 10 15 29
Hank 91 27 13 10 11 29
thunder_chik 24 13 13 13.5 24
pcgoode 28 17 13 16.5 24
2 for 2028 38 13 13 7.5 24
amartinsu13 27 17 14 17 23
pefloresjr 27 17 14 17 23
BigEasyCard 24 14 14 14 23
82lsju 19 10 15 19.5 18
GoodGrief 42 3 15 18.5 18
Brannerite 21 12 15 17.5 18
Mick 38 17 17 11.5 15
Braniac_Infinite 24 17 17 18.5 15
ferrari 24 17 17 18.5 15
hurlburt88 24 17 17 18.5 15
PrinceLightfoot 26 20 18 22 11
Winflop 24 19 19 21.5 10
CrazedZooChimp 45 13 20 18 9
CornFed 41 17 20 10 9
garvin 17 14 21 24.5 7
Boston Card 20 17 21 21.5 7
fullmetal 16 13 21 25.5 7
SUBuddha 35 24 21 23.5 7
NYCCardinalFan 13 10 21 28.5 7
SeattleTree 9 7 22 33 2
SamAtoms1980 48 13 23 22.5 1
==========================
Median 27 13 10 11
Actual Score 31 7 0 0
In case people are wondering, the modified error takes a slightly different calculation and is intended to reward people for getting the spread right even if the individual scores were off. Basically, instead of abs(predicted Stan - actual Stan) + abs(predicted UCSF + actual UCF), the formula is abs(predicted sum of scores - actual sum of scores)/2 + abs(predicted difference in scores - actual difference in scores).
It doesn't make a of exceedingly large proportions difference as you guys can see, but it would help someone who maybe predicted Stanford to win 24-0 (nailing the difference) over someone who predicted 24-14. In both cases, they were off by a touchdown for each team, so the old system would have given them 14 points. But by the new system, the second score is penalized for missing the difference by 14 points.
BC
I may understand this better, BC, after another shot of Ardmore, but if this is how they score FanDuel I'm out :)
Re: Where is Gauss? - 81alum - 09-14-2015
BC,
Looking back at the thread I predicted
Quote:Stanford 31
UCF 13
Not all that great a prediction, but you seem to have missed it.
Thanks for doing this!
Re: Where is Gauss? - washingtonismoney - 09-14-2015
Gauss is just a few standard deviations away.
Re: Where is Gauss? - winflop - 09-14-2015
Is there a prize for last because that's what I'm shooting for
Re: Where is Gauss? - Rally - 09-14-2015
(09-14-2015, 09:09 PM)winflop link Wrote:Is there a prize for last because that's what I'm shooting for
I am new here but my understanding is the winner gets a virtual tshirt and second place gets two. Season winner gets a virtual hand signed boda bag from Andrew Luck. Or so they say.
Re: Where is Gauss? - needle - 09-15-2015
Thanks BC!