So, you want to fade the public?

[Joe Public]

I don't know if this really is your chance. But I found myself in another confidence pool outside of the one here. Yes, I'm cheating on Big Al, he hogs the covers in the winter and won't introduce me to his parents. Now at least it's out in the open.

Anyway, I have to believe this thing is a very good representation of the public.

There are only two downsides, one, no spreads. That's the biggest one I see as I think it will take a number of the big number games out of the running. The other downside, there are only 210 people in the thing. I'm no statistics guy, but I'd feel better with a 500 person sample size.

At any rate, if you guys are interested, and if someone can give me an idea of how to post a table here, I'll put up the times picked, that percentage, and average points wagered per contest.

 

[redbearde]

supposedly, 30 is the magic number known as statistically significant.

And I, for one thing, would like to see such a list for ML dog purposes.

SJSU is the only one I've really identified so far...

 

[Joe Public]

Yeah, I'm guessing the guy who's running this enjoys his Excel, and bless his little heart for it. Because he's done a lot of the work for me. It's just a matter of either converting these things into HTML tables, or just sort of cutting and pasting/retyping these things. Which I could do, I'm not 'that' lazy.

It would just be nice to format them somehow.

Interesting point, I actually have the results from last year's version of this pool as well. The winner went all of 20-8 with 320 points. This year there are what, four more bowl games so I suppose that adds another, what would it be, 120 points to the potential total.

Dammit, remedial algebra, where are you when I need you--somebody, what's the equation for figuring that out?

At any rate, I'm not sure 320 last year impresses me all that much. Not that I'm saying I can do better, just that 20-8 SU, especially when you're going to have at least three or four double digit spreads, well, you see what I'm saying.

 

[Joe Public]

F**k a duck, I missed an entire page of games, I thought it felt light. Motherf**ker. Can somebody delete that post and I'll re-rack this thing?

 

[G-Man]

supposedly, 30 is the magic number known as statistically significant.

And I, for one thing, would like to see such a list for ML dog purposes.

SJSU is the only one I've really identified so far...

If you are looking for accurate public opinion on wagering, the national number is 3000 people surveyed or more on any given event.

 

[Joe Public]

If you are looking for accurate public opinion on wagering, the national number is 3000 people surveyed or more on any given event.

Good info, Mr. Scrooge.

Numbers to follow anyway.

=)

 

[Joe Public]

Here are your SU perecentages:

CMU - 98%
Rutgers - 94.3%
Texas - 93.3%
Clemson - 92.4%
Ohio St. 91.4%
Hawaii - 90.9%
Auburn - 90%

WVA - 86.7%
Lou. - 85.7%
UCLA - 85.3%
TCU - 85%
S. Florida - 84%
T.Tech - 83%
BC - 81%
So. Car. - 80.5%

LSU - 78%
Cincy - 77%
Mich - 76.7
Tenn. - 76%
Ore. St. - 73.4%

VT - 69.6%
Rice - 68%
Cal - 68.2%
OK - 67.2%
Wisky - 63.9%
Miami - 63%
BYU - 62%
So. Miss. - 62%

Utah - 54.9%
Maryland - 54.5%
OKST - 53.5%
New Mexico - 50.7%

 

[Joe Public]

Now, the whole deal. What's posted is the following. The percentage of people on one side followed by the numbers from which that percentage was taken.

Next is confidence point differential, the difference in average confidence points put to each team. Interesting thing here, two teams that won on percentage, Utah and Wisky, lose on confidence points. Meaning that although more people picked them to win the game, the people who picked Tulsa and Arkansas put more confidence points on their pick. In the case of Tulsa it's not that big of a variance, but with Wisky/Ark. it does seem significant.

Last is total points received and the division lines are the day breaks.

Again, who knows if this thing helps or is any sort of accurate berometer of anything. But maybe somebody can take something from these.

TCU - 85% - (180-31)
point diff.: 12 - (22.5 -10.5)
Total pts: 4050-325

BYU - 62% - (131-80)
point diff.: 4.8 - (14.1-9.3)
Total pts: 1847-744

Rice - 68% - (145-66)
point diff.: 6.7 - (12.3-5.6)
Total pts: 1786-371

-------

S. Florida - 84% - (178-33)
point diff.: 4.9 - (13.3-8.4)
Total pts: 2373-276

New Mexico - 50.7% - (107-104)
point diff.: .8 - (11.0-10.2)
Total pts: 1178-1061

Utah - 54.9% - (116-95)
point diff.: Tulsa .5 - (9.1-8.6)
Total pts: 997-865

-------

Hawaii - 90.9% - (192-19)
point diff.: 12.3 - (21.0-8.7)
Total pts: 4024-166

-------

CMU - 98% - (207-4)
point diff.: 13.8 - (20.2-6.0)
Total pts: 4173-24

-------

UCLA - 85.3% - (180-31)
point diff.: 5.7 - (15.7-10.0)
Total pts: 2817-309

-------

OKST - 53.5% - (113-98)
point diff.: .7 (10.9-10.2)
Total pts: 1234-1003

Cal - 68.2% - (144-67)
point diff.: 4.4 (15.3-10.9)
Total pts: 2207-727

Rutgers - 94.3% - (119-12)
point diff.: 12.6 (23.0-10.4)
Total pts: 4582-125

-------

Clemson - 92.4% - (195-16)
point diff.: 8.8 (21.7-12.9)
Total pts: 4225-206

Ore. St. - 73.4% - (155-56)
point diff.: 2.5 - (13.1-10.6)
Total pts: 2034-594

So. Car. - 80.5% - (170-41)
point diff.: 3.2 - (18.2-15.0)
Total pts: 3100-614

T.Tech - 83% - (177-34)
point diff.: 7.3 - (18.3-11.0)
Total pts: 3245-373

Maryland - 54.5% - (115-96)
point diff.: 2.1 - (13.2-11.1)
Total pts: 1522-1061

-------

BC - 81% - (171-40)
point diff.: 7.3 - (18.6-11.3)
Total pts: 3184-450

Texas - 93.3% - (197-14)
point diff.: 14.8 - (26.0-11.2)
Total pts: 5115-157

VT - 69.6% - (147-64)
point diff.: 6.6 - (17.3-10.7)
Total pts: 2537-685

-------

Miami - 63% - (1873-1042)
point diff.: .7 - (14.1-13.1)
Total pts: 1873-1042

-------

Tenn. - 76% - (161-50)
point diff.: 5.9 - (18.6-12.7)
Total pts: 2992-634

Auburn - 90% - (190-21)
point diff.: 8.3 - (18.2-9.9)
Total pts: 3456-208

WVA - 86.7% - (183-28)
point diff.: 8.3 - (20.8-12.5)
Total pts: 3807-349

Wisky - 63.9% - (135-76)
point diff.: Ark. 1.4 - (14.8-13.4)
Total pts: 1811-1128

Mich - 76.7% - (162-49)
point diff.: 3.2 - (16.1-12.9)
Total pts: 2607-633

OK - 67.2% - (142-69)
point diff.: 7.1 - (20.7-13-6)
Total pts: 2938-939

-------

Lou. - 85.7% - (181-30)
point diff.: 9 - (22.9-13.9)
Total pts: 4144-418

-------

LSU - 78% - (165-46)
point diff.: 6.5 - (21.9-15.4)
Total pts: 3616-709

-------

Cincy - 77% - (163-48)
point diff.: 5.7 - (17.3)
Total pts: 2814-557

-------

So. Miss. - 62% - (131-80)
point diff.: 5.5 - (16.3-12.8)
Total pts: 2130-1022

-------

Ohio St. - 91.4% - 193-18
point diff.: 13.7 - (25.9-12.2)
Total pts: 4996-219

 

[SHSUHorn]

Interesting take.

 

[Dreamer]

Interesting. The "public" is 2-1, so far this bowl season....

 

[Joe Public]

True, Dreamer, though I think the consensus in this contest is on every single favorite SU.

 

[Seabass]

there isn't any set number that makes something statistically significant-it depends what you are aiming to prove. i.e, if 90% of 30 people all believe in one side, there is probably solid basis to believe that the public as a whole are on that side too. But if 60% of 30 people believe in one side, odds are that it could just be chance from a random sample of a 50% population...

 

[HUNT]

good stuff joe, I deleted the post you wanted gone for you.

 

[Joe Public]

Thanks, HD.

Yeah, Seabass, I agree. I also don't know how much regional bias figures in. I believe this thing is being run by someone out of Ohio, but that doesn't tell you about the whole pool. I'm out in California, I'm sure others in this thing are as well as this was sent along to me. But if say 75% of the people were out of Ohio it wouldn't be surprising to see such a high number on Ohio St. vs. the Gators.

Actually, funny thing about that game, in this pool a lot of people actually have Ohio St. as their 32, but like two people have Florida as their 32. What's interesting to me about that isn't just that Florida's a seven point dog in the game, but that these people would go that high on Florida to win the game outright.

Maybe it's a strategy thing and they figure that if they hit that at 32 and everybody else loses at anywhere between 20 and 32 that swing will give them a shot. Still I found it interesting.

I think some of these games the above numbers are going to be sort of meaningless. Like for the CMU game. Hell the line is 10 and they're playing MTSU, so that nearly everybody picked CMU isn't surprising.

But that they're so one-sided on Michigan is interesting. Same with UCLA.