Nate Silver Predicts Super Bowl Matchup

[19th Hole]

Nate Silver Predicts Super Bowl Matchup Featuring Patriots vs. Seahawks




This undated image released courtesy of Brian Silver shows author and statistician Nate Silver in New York. The 34-year-old statistician, unabashed numbers geek, author, and creator of the much-read FiveThirtyEight blog at the New York Times, correctly predicted the presidential winner in all 50 states, and almost all the Senate races. (AP Photo/Robert Gauldin)
Apparently unwilling to rest on his laurels after garnering wide acclaim for his remarkably accurate predictions for the U.S. Presidential election (aka the "Super Bowl" of politics) at his Five Thirty Eight blog at The New York Times, statistician Nate Silver has unleashed his mind on a decidedly more (ahem) hard hitting matter: the actual Super Bowl.
In an interview Thursday, Silver told ESPN's "First Take" that he's predicting a Patriots-Seahawks Super Bowl matchup. Citing data from "Football Outsiders," a prominent football statistics site that attempts to "bring objective analysis to football," Silver prognosticated that the No. 2 seed from the AFC and the No. 5 seed from the NFC would reach New Orleans.
While a pick of New England to represent the AFC won't likely surprise anyone, why Seattle over the Atlanta Falcons or San Francisco 49ers? As Silver explains it, balance is the key:

San Francisco... played very well for much of the year,” said Silver, “In terms of the balance overall between offensive and defensive play, Seattle had that best balance throughout the year and... in the second half, were blowing people out by 50-to-7 scores and so forth. So, they had the best stats from the regular season.

And as for the Patriots, Silver predicts a Denver-Patriots playoff game will become the de facto Super Bowl, but ultimately the Patriots will come out on top and advance to the biggest game.

[detroitlionsfan]

correctly predicted the presidential winner in all 50 states ?

if true thats incredible

[Shafted69]

he should stick to predicting elections lol

[gregm]

He actually got started writing for baseball prospectus and he handled the pecota projections for a long time. If you are interested in sabermetrics, you have definitely seen him around before but probably didnt recognize the name until his blog became popular in the mainstream media.

[Dfjay9]

Yeah 50/50 and 49/50 in 08 I believe. The dudes amazing

[goofyre]

Dude is amazing. But he stated before that football is the hardest sport to predict because of so many variables. Baseball, Basketball, then football as the hardest.

[nvrlose37]

correctly predicted the presidential winner in all 50 states ?

if true thats incredible

Not really when 36 states are basically guaranteed before the election and even the swing states had leans. Had to make maybe four tough predictions with Virginia, Florida, and Ohio.

Stick to politics pal

[TiredPro]

All 50 states eh ?? sharp dude.......I'm sure he considers everything...........even voter fraud.

[Dfjay9]

Not really when 36 states are basically guaranteed before the election and even the swing states had leans. Had to make maybe four tough predictions with Virginia, Florida, and Ohio.

Stick to politics pal

You're a retard to think that 50/50 and 49/50 isn't amazing. Not to mention 31/33 Senate races this past election. Oh and the state he didn't get right in 08 was by less than 1%

[SEAHAWKHARRY]

I like it

[nvrlose37]

You're a retard to think that 50/50 and 49/50 isn't amazing. Not to mention 31/33 Senate races this past election. Oh and the state he didn't get right in 08 was by less than 1%

It's pretty good but not amazing. Like I said, only about 14 swings, all with heavy leans except for like 4 to 5 toss ups. Basically 15-1 odds and that's being generous.

Losing Missouri in 08 is tough but like the others is a tossup

[Dfjay9]

Not even. The polls have been all over the place pal. No other pollster predicted a landslide for Obama.

When I saw Silver's predictions I curled up into a ball, hugged my money and cried.

[nvrlose37]

http://www.realclearpolitics.com/epolls/2012/president/2012_elections_electoral_college_map.html

Check it, 39 guarantees, 11 swings. FL, OH, VA, NC and maybe NH the only hotly contested of the 11. All of those five had leans towards one side. Hence 15-1 at best.

[lamanax]

Guy has been amazing. Bill Simmons had a good podcast with him recently. Guy is impressive. If he gets this right... his legend will get ridiculous

[detroitlionsfan]

Guy has been amazing. Bill Simmons had a good podcast with him recently. Guy is impressive. If he gets this right... his legend will get ridiculous

he might be right

[McBa1n]

Nate Silver is very interesting to read. I really enjoy his work. But sheesh, if the Pats have to go to Denver? That's a tough game. Also, I'd be shocked if Seattle got by Atlanta, let alone won 3 in a row on the road. If Seattle made it to the Superbowl, it'd be very impressive. They have everything you need to win but a veteran QB, but the way Wilson plays, he gives them as good of a chance as anyone. Seems one team shocks every year. Can't wait for the dang games start. I'm goin nuts waitin over here.

[Lo$t]

Fuk Silver. Any book would love to take his math action. F'ing ferry. Going to be Peyton vs Rodgers.

[pavyracer]

He picked blue states to go to the Superbowl. Using election data to predict the Superbowl teams is moronic.

[gregm]

His writings about baseball sabermetrics and the political blog is totally different than this prediction. He even mentions the small sample size in a football season versus the types of data he usually deals with, and he is using the Football Outsiders system which he didnt work on directly. He said he thought the two best teams were the Patriots and the Broncos, I think he is going a little overboard with the media appearances.

[suicidekings]

His writings about baseball sabermetrics and the political blog is totally different than this prediction. He even mentions the small sample size in a football season versus the types of data he usually deals with, and he is using the Football Outsiders system which he didnt work on directly. He said he thought the two best teams were the Patriots and the Broncos, I think he is going a little overboard with the media appearances.

Wouldn't you? He's in a perfect spot to make a bold prediction, knowing that the Superbowl is the most watched event of the year. If Seattle was to pull off a win this weekend he'll come out looking good, and all of a sudden have the attention of a whole new audience that would never normally follow him. Good for business.

[StackinGreen]

He'll be wrong about this as he has been on most of his sports predictions. The Chicago Tribune tried to give him a ton of credit (even retrospectively, flat out lying) for predicting a White Sox season 2007 perfectly (I believe 72 wins that year). The problem was that in every flanking year they outdid his prediction, which was just around that number, including when they won the World Series in 2005 (which was what the stupid Trib reporter lied about).

If he doesn't get this right, it'll be Football Outsiders fault, watch (Bad Data claim). I respect the pick, but don't think it'll happen, and certainly don't believe that it has anything to do with elections predictions or any real football analysis. This is classic media trying to convert one thing to the other and hoping it'll be successful --- it won't be.

So as not to be a naysayer, I'll make my prediction: Packers over Patriots.

I realize it is easier always to say someone's going to be wrong, so I"ll at least throw that out there in order not to be annoying.

[marcoloco]

interesting

[rm18]

he should of took the bookies for millions with those election prediction

[face]

makes me want to bet the falcons. fighting my simplistic contrarian reaction

[pavyracer]

What is his record predicting the Superbowl teams when the last 8 teams are left?

I'd bet you 30% of SBR posters can predict the Superbowl teams when the last 8 are left.

[SEAHAWKHARRY]

Pittsburgh Ny giants come to mind teams that had to get 3 road wins to get to SUPERBOWL..come to think of it they won as well.hummmmmm

[hankcream]

Hopefully he can move the line down, I'd love to pound Atlanta ML tomorrow.

[gregm]

I have been using football outsiders for a while but seattles on the road stats vs. seattle at home stats are what stand out to me and I am on Atlanta, but here the team efficiency ratings from football outsiders. You need to understand the DVOA, Defense-adjusted Value Over Average (DVOA) system. http://www.footballoutsiders.com/info/methods#dvoa

TEAM	TOTAL DVOA	LAST YEAR	NON-ADJ TOT VOA	W-L	OFFENSE DVOA	OFF. RANK	DEFENSE DVOA	DEF. RANK	S.T. DVOA	S.T. RANK
1	SEA	38.3%	19	30.6%	11-5	18.5%	4	-14.1%	4	5.7%	3
2	DEN	36.6%	24	38.8%	13-3	22.1%	2	-13.8%	5	0.7%	13
3	NE	34.9%	3	37.6%	12-4	30.8%	1	1.3%	15	5.5%	4
4	SF	29.9%	6	24.0%	11-4-1	17.0%	5	-14.3%	2	-1.5%	20
5	GB	26.6%	1	18.8%	11-5	19.5%	3	-7.3%	8	-0.2%	18
6	CHI	20.5%	15	18.2%	10-6	-10.9%	26	-26.8%	1	4.7%	6
7	NYG	13.5%	12	12.3%	9-7	12.9%	7	1.4%	16	2.0%	10
8	BAL	9.8%	7	11.6%	10-6	3.0%	13	2.2%	19	9.0%	1
9	WAS	9.6%	21	13.7%	10-6	15.3%	6	1.8%	17	-4.0%	27
10	ATL	9.1%	8	14.7%	13-3	6.1%	12	-2.9%	12	0.1%	16
11	HOU	6.6%	5	14.7%	12-4	0.1%	16	-14.2%	3	-7.7%	32
12	CIN	6.1%	17	12.1%	10-6	-1.8%	17	-3.8%	10	4.1%	7

[C-Gold]

He is a square who picked a ROOKIE QB to win 3 road playoff games on a team that was 3-5 on the road. He is predicting Russell Wilson to win as many games in the playoffs on the road as he did in the entire regular season, against playoff teams. Every time he says he thinks Seattle is going to the super bowl replace it with, " I think a rookie QB will win 3 road playoff games".

The football outsiders suck. Good idea bad execution. Their predictions are as bad as the popular magazines they poke fun at. They say they care about statistics but then pick games because they FEEL this is the better team.

[19th Hole]

[QUOTE=C-Gold;17460664]He is a square who picked a ROOKIE QB to win 3 road playoff games on a team that was 3-5 on the road. He is predicting Russell Wilson to win as many games in the playoffs on the road as he did in the entire regular season, against playoff teams. Every time he says he thinks Seattle is going to the super bowl replace it with, " I think a rookie QB will win 3 road playoff games".
QUOTE]


~~~~~~~~~~~~~~~~~~~~~

You wouldn't know a square if it fell on your head.
Square and Sharp have been so over used they have lost
any meaning.
The only words that mean anything are WON or LOSS.

But... Good Luck and Enjoy the games...

[C-Gold]

Sharps don't pick a rookie QB to win 3 road playoff games especially given 8/1 odds.

I love how stat boy makes his pick now when there are 8 teams left. He might as well wait until the championship games to make his super bowl prediction.

[Grits n' Gravy]

The guy may be wrong with his SB picks but his work on analyzing data and formulating predictions are truly great reads. Likely over the heads of most guys here though.

[detroitlionsfan]

Pittsburgh Ny giants come to mind teams that had to get 3 road wins to get to SUPERBOWL..come to think of it they won as well.hummmmmm

last yrs giants played the wildcard at home

you might be thinking of 2010 paclkers

[detroitlionsfan]

in a way he was wrong by predicting a patriot bronco afc champ

but by denvfer losing its really setting up for a pats hawks

[pavyracer]

This guy is really good. How did he know?