Started By
Message

The computers pick South Carolina over Georgia.

Posted on 11/1/17 at 4:41 pm
Posted by Lonnie Utah
Utah!
Member since Jul 2012
24103 posts
Posted on 11/1/17 at 4:41 pm
They really did.

USC 35 UGA 34
Posted by WG_Dawg
Hoover
Member since Jun 2004
86572 posts
Posted on 11/1/17 at 4:41 pm to
Well this is a day ruiner
Posted by I Bleed Garnet
Cullman, AL
Member since Jul 2011
54846 posts
Posted on 11/1/17 at 4:42 pm to
Posted by Lonnie Utah
Utah!
Member since Jul 2012
24103 posts
Posted on 11/1/17 at 4:43 pm to
On a serious (non trolling) note, it's a very interesting analysis.

Basically, both guys at the end say their model in wrong in this case. But it does set up the scenario where SC could keep it close.
Posted by DirtyDawg
President of the East Cobb Snobs
Member since Aug 2013
15539 posts
Posted on 11/1/17 at 4:44 pm to
Just like FoxSports giving Florida a 52% chance to beat us
Posted by SCLibertarian
Conway, South Carolina
Member since Aug 2013
36285 posts
Posted on 11/1/17 at 4:44 pm to
They're going rogue. I've seen this before.

Posted by CaptainBrannigan
Good Ole Rocky Top Tennessee
Member since Jan 2010
21644 posts
Posted on 11/1/17 at 4:46 pm to
I used to think computers were shite. But they have figured out that Gus Malzahn is a moron who lost to Coach O. So I'll trust them
Posted by U_south_cack
Member since Sep 2016
464 posts
Posted on 11/1/17 at 4:47 pm to
COCKS ON TOP!!!!!!!!!!
Posted by Lonnie Utah
Utah!
Member since Jul 2012
24103 posts
Posted on 11/1/17 at 4:47 pm to
quote:

Just like FoxSports giving Florida a 52% chance to beat us


Except for Fl not showing any ability to score up to this point in the season, yes.

More interesting than the analysis, is the comments section. Most are predicting a massive blowout. If you watch the video, you'll understand why that isn't likely to happen. Both team have a propensity to shorten the game. In fact, that's both teams game plan. That strategy limits possessions for both teams and keeps the score low.
Posted by RTRinTampa
Central FL
Member since Jan 2013
5532 posts
Posted on 11/1/17 at 4:48 pm to
BS in. BS out.
Posted by fibonaccisquared
The mystical waters of the Hooch
Member since Dec 2011
16898 posts
Posted on 11/1/17 at 4:49 pm to
Hang that shite on the bulletin board...
Posted by CockyTime
Baton Rouge
Member since Nov 2015
3154 posts
Posted on 11/1/17 at 4:49 pm to
print the fricking shirts
Posted by Lonnie Utah
Utah!
Member since Jul 2012
24103 posts
Posted on 11/1/17 at 4:50 pm to
quote:

BS in. BS out.


That's why their model is 15-3 against the spread this year.
This post was edited on 11/1/17 at 4:51 pm
Posted by fibonaccisquared
The mystical waters of the Hooch
Member since Dec 2011
16898 posts
Posted on 11/1/17 at 4:52 pm to
quote:

Both team have a propensity to shorten the game. In fact, that's both teams game plan. That strategy limits possessions for both teams and keeps the score low.


Haven't watched enough SC this year to know, but will take your (and their) word for it. Assuming that is true, absolutely it could limit scoring opportunities. My only major concern in this game is that the Gamecocks have been particularly good in the turnover margin, particularly in SEC play. If this is in fact due to their style of play and not simply something that will regress to the mean, then I'd say that could certainly keep them in the game. It might also necessitate that we not shorten the game though as well. Ultimately, would rather not find out. :) Dawgs roll!
Posted by PortlyDawg
GA
Member since Aug 2011
2400 posts
Posted on 11/1/17 at 4:52 pm to
Their model indicated the fewer the yards SC gains the more points they score. Seriously.
Their algorithm might have few flaws. I love their weekly YouTube predictions though, good stuff.
Posted by SummerOfGeorge
Member since Jul 2013
102699 posts
Posted on 11/1/17 at 5:05 pm to
I'd really like to know what fatal flaw in their model tossed out 35 pts for South Carolina.

35 pts would be tied for the 2nd most pts Carolina has scored all season (48 vs Arkansas).

35 pts would clearly be the most pts Georgia has given up all season (28 vs Missouri, 19 vs Notre Dame).

I assume the disconnect has something to do with all the non-offensive TDs Carolina had against NC State and Arkansas, as they have them scoring 35 pts on 4.10 YPP (which is bad).
Posted by SummerOfGeorge
Member since Jul 2013
102699 posts
Posted on 11/1/17 at 5:06 pm to
Now I'm curious so I'm gonna run my (probably not as good generally) model and see what it spits out.
Posted by DustyDinkleman
Here
Member since Feb 2012
18176 posts
Posted on 11/1/17 at 5:09 pm to
So do I
Posted by MikeyFL
Las Vegas, NV
Member since Sep 2010
9619 posts
Posted on 11/1/17 at 5:10 pm to
"If I gave South Carolina less yardage, they would be predicted to score more points."


Funny how computers interpret data.
Posted by fibonaccisquared
The mystical waters of the Hooch
Member since Dec 2011
16898 posts
Posted on 11/1/17 at 5:16 pm to
quote:

Their model indicated the fewer the yards SC gains the more points they score. Seriously.
Their algorithm might have few flaws. I love their weekly YouTube predictions though, good stuff.


Could potentially be calculating short fields/turnovers? Otherwise, yeah... probably flawed model.
Page 1 2 3
Jump to page
first pageprev pagePage 1 of 3Next pagelast page

Back to top
logoFollow SECRant for SEC Football News
Follow us on Twitter and Facebook to get the latest updates on SEC Football and Recruiting.

FacebookTwitter