Writers of Pro Football Prospectus 2008

26 Jan 2006

Hawks Will Win, Shrink Calculates

John F. Murray, a sports psychologist in Florida has developed a system that has successfully projected the winner or the team that covered the point spread in the past three Super Bowls, and he's picking the Seahawks to win by 5 to 10 points this year. See if this sounds familiar:

"Mental Performance Index (MPI) is ... based on Murray's assessment of each play, quantifying the degree to which a team performs to perfection. For instance, a 3-yard rushing play might get an average score in one situation or an above-average score in another, if it came on a third-and-3. Defensively, if a team gives up a 12-yard reception, it might get a below-average score, or average to above average if there was a decent rush and good coverage. Murray emphasizes the importance of each play, including those on special teams, then adds it up, yielding conclusions and predictions."

When he adds it all up, Seattle scores a .566 to Pittsburgh's .530. The article doesn't get into specifics, but I have a few questions. First, does the model control for strength of opponent? Second, is .566 significantly better than .530? Third, does Murray know Jerome Bettis is actually from Detroit?

Posted by: P. Ryan Wilson on 26 Jan 2006

49 comments, Last at 30 Jan 2006, 1:12am by John Cramer

Comments

1
by White Rose Duelist (not verified) :: Thu, 01/26/2006 - 12:55pm

Seattle also leads in disrespect (.573 to .529), but trails severely in nastiness (.601 to .530).

2
by DJAnyReason (not verified) :: Thu, 01/26/2006 - 12:59pm

Quality of opposition struck me as well. Also, he's got sample-size issues. He's projecting Seattle at .566 (whatever that means) based on two games. Projecting Pittsburgh at .530 at 3 games is hardly better, but its slightly more robust. One would expect Seattle to regress towards the mean had they played a third game, based simply on... well, regression towards the mean.

His system has picked the last three winners. Get 8 guys in a room, and odds are you'll find someone who's picked the last three winners.

3
by pawnking (not verified) :: Thu, 01/26/2006 - 1:02pm

Actually, I was very intrigued by this article. Do you think FO can do an in-depth interview with this guy?

4
by RowdyRoddyPiper (not verified) :: Thu, 01/26/2006 - 1:02pm

My question is how does he know that there was good coverage, does he actually go to the games? Watching them on TV it's very difficult to tell.

5
by JC (not verified) :: Thu, 01/26/2006 - 1:18pm

Potential flaw:

"based on Murray's assessment of each play"

This guy's a psychologist!

6
by Sam (not verified) :: Thu, 01/26/2006 - 1:31pm

And is it defense-adjusted?

7
by Mike (not verified) :: Thu, 01/26/2006 - 1:41pm

Last three Super Bowls, eh? Everyone and their mother picked the Pats over the Panthers and the Eagles, so I think that's none too impressive. Leaving us with the Bucs over the Raiders. A clever pick, perhaps, but it's not like he picked the Pats over the Rams.

8
by Slippery Pete (not verified) :: Thu, 01/26/2006 - 2:09pm

#1:

Pittsburgh also holds a slight edge in swagger (.872 to .865).

9
by David (not verified) :: Thu, 01/26/2006 - 2:18pm

#7: Murray has predicted the winner against the spread all three years, which means he picked the Eagles last year and, if I'm remembering right, the Panthers in 2004.

10
by Nog (not verified) :: Thu, 01/26/2006 - 2:27pm

I suppose it also does not take into the fact that pittsburgh has soo much more superbowl experience than seattle (with all those wins in the 70's) and not to mention willie williams on their roster!

11
by Steve Z (not verified) :: Thu, 01/26/2006 - 2:29pm

The last three Superbowls, eh. That’s a, a, a...very small sample of the total population!

12
by brian (not verified) :: Thu, 01/26/2006 - 3:11pm

Regardless of the sample size, his methods are very similar to the methods and calculations used here (ie. 3rd down weightings), which, in all reality, are theories still in their infancy as well.

If someone actually could watch a game and be able to deterimine coverage and importance of a play, and a teams "perfection" that would provide data that would be the most importat aspect of any formula at FO.

Brian

13
by putnamp (not verified) :: Thu, 01/26/2006 - 3:18pm

Are we sure he's not just doctoring DVOA numbers?

14
by underthebus (not verified) :: Thu, 01/26/2006 - 3:35pm

The Steelers will win because they want to win a Superbowl and the Hawks are just happy to be there.

15
by J-Diddy (not verified) :: Thu, 01/26/2006 - 3:48pm

Yeah Bettis being from Detroit will give him extra motivation. So instead of gaining 6 yards on 4 carries, he can gain 10 yards on 6 carries. If one of his carries lands him in the end zone, he will probably get MVP too. I can't think of a player who annoys me more.

16
by Fnor (not verified) :: Thu, 01/26/2006 - 3:52pm

J-Diddy hate popular situational backs! J-Diddy smash!

17
by Manteo (not verified) :: Thu, 01/26/2006 - 4:26pm

14 - Absolutely. I expect the Hawks to spend most of the time snapping photos of each other on the field and discussing afterparties. We shouldn't expect them to try to win the game.

The Steelers clearly lead in Trying to Win (.806 to .492).

18
by Becephalus (not verified) :: Thu, 01/26/2006 - 4:37pm

lol Fnor

19
by DJ Any Reason (not verified) :: Thu, 01/26/2006 - 4:38pm

The Steelers also maintain huge leads in QB Hair (.763 to .109) and respect for Tom Brady, having lost to him twice in AFCC games (.812 (genuflect) to .412 (genuflect))

20
by Kal (not verified) :: Thu, 01/26/2006 - 4:45pm

Sorry, but Seattle owns a huge edge in respect for Tom Brady: Seattle flew a #12 flag for three weeks, is still flying the flag currently, and if they win will force Pitt to fly the #12 flag in Pittsburgh.

After factoring that into the calculation, I believe that Seattle has a .721 to .412 edge on respecting Tom Brady.

21
by Zac (not verified) :: Thu, 01/26/2006 - 4:50pm

#9. No, he has picked the winner OR the team that covers the spread. I don't know how often each scenario happens (team he picks wins and covers, team he picks wins but doesn't cover, team he picks loses but covers, team he picks loses and doesn't cover), but if they are evenly spread out, then he's covered 75% of cases right there. He could be a great TV psychic.

22
by David (not verified) :: Thu, 01/26/2006 - 4:52pm

Oh, misread that. Yeah, that's substantially less impressive.

23
by DJAnyReason (not verified) :: Thu, 01/26/2006 - 4:59pm

And, as a final factor, Pittsburgh has a 1.00 to 0.00 edge in Tyrone Carter

24
by Bill (not verified) :: Thu, 01/26/2006 - 5:12pm

And how many of Seattle's parents will be at the game?

25
by Falco (not verified) :: Thu, 01/26/2006 - 5:23pm

This qualifies for news in the off week? Doesn't sound like much of an exclusive club to me. If you made a pick each of the last two years, congratulations, you too could be perfect under this criteria, by picking either the winner (NE) or team that covered the spread (CAR/PHI).

The QB drafted 6th round or later index points to Seattle. Brady, Brad Johnson, Warner, now Hasselbeck. (curses on you, Trent Dilfer, you should have been picked in the 6th and are an honorary member)

But then, the equally logical road playoff victories index points to Pittsburgh. The team with more road playoff victories has either won or covered the last five superbowls.

26
by andrew (not verified) :: Thu, 01/26/2006 - 5:25pm

The Seahawks have a significant edge in owner bank account.

27
by B (not verified) :: Thu, 01/26/2006 - 5:26pm

24: From what I can gather of the media coverage of Seattle, all thier players are orphans. It must be the case, cause they never appear on television.

28
by DJAnyReason (not verified) :: Thu, 01/26/2006 - 6:12pm

Re: 27

In fact, 105 of the players in the Superbowl are orphans. Of all the players in the NFL, only Jerome Bettis and Donnovan McNabb have living parents. A true tragedy.

29
by Steve Z (not verified) :: Thu, 01/26/2006 - 6:51pm

Re: #12

The fact that our good Doctor Murray uses a method that’s superficially similar to those used by the FO crew does not thereby increase the confidence we can place in the evidence he cites: His recent Superbowl predictions. Three games are just not enough to lend credence to any claim that his method actually works.

30
by Trogdor (not verified) :: Thu, 01/26/2006 - 7:03pm

"Nonetheless, I remained skeptical until Murray kiboshed all doubts by disclosing a sure sign of credibility -- he spent a year in Pullman on a sport psychology internship and worked with the 1997 Pac-10 champion and Rose Bowling Cougs."

So.... this guy was the shrink that produced Ryan Leaf? Clearly, this man is brilliant, and should be trusted unconditionally.

31
by MRH (not verified) :: Thu, 01/26/2006 - 7:06pm

Seattle leads in the "coach on his 2nd team wins SB" index.

Since 1997, only Billick has won a Super Bowl in his 1st NFL HC job (Shanahan, Vermeil, Belicheck, and Gruden were all at their 2nd stop). Go back one more year and you find Holmgren, 1st stop. Before that, the index completely breaks down.

32
by Mshray (not verified) :: Thu, 01/26/2006 - 7:49pm

16 - Fnor, there is nothing funnier than Hulk-speak on message boards. Seriously. Unless of course it's Thor-speak or Dr. Strange-speak. By the Hoary Hosts of Haggoth!

28 - but you said parents, does that mean you've seen Donovan's dad on TV? I don't watch many Eagles games, but I haven't.

31 - If you look at the coaches who took 2 teams to the SB index, none have won with both teams. Parcells won with his first team & lost with his second, Shula & Vermeil the opposite, and Reeves lost with both. Or look at it another way, nine head coaches besides Holmgren have gone on to guide other franchises after winning a Super Bowl. But most never even sniffed the playoffs again, beginning with Vince Lombardi, who was at the Washington Redskins for one year following the Packers. Same for the Chiefs' Hank Stram (Saints), Colts' Don McCafferty (Lions), Raiders' Tom Flores (Seahawks), Bears' Mike Ditka (Saints), and 49ers' George Seifert (Panthers).

Two more, the Cowboys' Jimmy Johnson (Dolphins) and Rams' Dick Vermeil (Chiefs), made the playoffs with their new clubs but did not get close to the Super Bowl. Only Bill Parcells has won a Super Bowl & made it back with a new team, and of course who did he lose to? Holmgren.

33
by Nolan (not verified) :: Thu, 01/26/2006 - 8:00pm

Who are you people kidding about Seattle just excited to be there. Take a look at the 2 locker rooms after their respective games. Pittsburgh was the ones who were happy to be in the super bowl! Seattle was clearly aware that they had plenty of work to do!!

34
by Jed (not verified) :: Thu, 01/26/2006 - 8:04pm

The only other parent I've heard anyone talk about is Hasselbeck's dad who was a former player. That's the most amazing one of all - talk about an immaculate conception...

35
by Balaji (not verified) :: Thu, 01/26/2006 - 8:24pm

Well Nolan, if I were a Steelers player I would have been celebrating too, since my only prior experience would have been losing in conference championship games. With that history, I think they deserved a little celebration.

And I don't think being glad to get over that hump means they aren't aware of what needs to be done - remember, Cowher has been to the Super Bowl before, so he knows what's at stake. And all the interviews of Steelers so far have contained a variation of "we're not satisfied just to be here."

36
by DJAnyReason (not verified) :: Thu, 01/26/2006 - 8:34pm

Re: 32

The most recent Chunky Soup commercial features his dad:

Dad: "That's my son!"
Mom: "That's my play!"

37
by Joey Freakin\' Porter (not verified) :: Thu, 01/26/2006 - 8:42pm

Mental Performance Index (MPI) is … based on Murray’s assessment of each play, quantifying the degree to which a team performs to perfection.

Joey Porter disdains your "mental" performance index. Murray wants the game to be a thinking game, the Steelers want the game to be a football game.

38
by Paul (not verified) :: Thu, 01/26/2006 - 8:53pm

Tomorrow, the staff astrologer gives us her reading on things, and on Saturday, we'll examine a chicken's entrails for additional clues.

39
by putnamp (not verified) :: Thu, 01/26/2006 - 9:35pm

#33,35

I think you guys are missing the joke here.

40
by FastEddy (not verified) :: Thu, 01/26/2006 - 10:21pm

Falco #25: Extremely good point, which I missed initially. Geez. Picked the winner OR the winner ATS? Literally everyone got the last 2 years right. And even I, with my slight problem since the lobotomy, picked TB to crush Oakland.

That is so bogus. The bogosity needle is off the scale. Bogons are flying every which way.

It is frightening that people give credence to jerks like this. Next thing you know someone will be saying that there's WMDs in Iraq and that warrantless eavesdropping is a *good* thing. For, um, security... yeah, security.

41
by vr (not verified) :: Thu, 01/26/2006 - 11:46pm

one key statistic that was not mentioned is that the steelers have a huge advantage in hair (.823 to .651). Look, with Polamalu and Roethlisberger alone compared to the Hawk superstars like Hasselbeck and Alexander, this could get ugly...

42
by someone (not verified) :: Fri, 01/27/2006 - 12:04am

re.41. Its a good job Roeth and Troy are on different sides of the ball respectively. Otherwise, their lead in Potential Velcro Effect (.995 to 0.104) could be disastrous for them.

43
by ulric (not verified) :: Fri, 01/27/2006 - 5:18am

The team with the weaker DVOA-rated regular season strength of schedule has won 5 of the last 7 Superbowls, and is 4-for-4 in games where Bill Bellicheck is not coaching one of the participating teams. Since the Patriots are not playing next Sunday, the Seahawks will win. Expect a final score in the area of 87-13.

44
by Make Every Miles Davis Count (not verified) :: Fri, 01/27/2006 - 6:37am

Is it really a good thing to be nicknamed "Football Freud"? Reminds me of an Onion blurb:

INDIANAPOLIS—According to sources close to Ashley Manning, wife of Colts quarterback Peyton Manning, the All-Pro's constant audibles on the playing field are mirrored in his intimately personal life, a fact that Mrs. Manning sometimes finds exasperating. "I mean, sometimes when we, you know, 'take the field,' and the 'formation' is all 'lined up,' and Peyton gets all set 'under center,' he will start barking out new signals," Mrs. Manning said. "Suddenly, it's 'Flip Right Hawk 62! Deuce Right, Waggle 15H Throwback C-Posts! Dice Right 218 Bastard! Yes! Yes! Yes!' And I'm like, Peyton, we're at 'third and inches' here, honey. Please, just 'snap the ball.'" Mrs. Manning did note, however, that her husband was a remarkably talented "field general" with a "rifle arm" who "led the league" in "finding the end zone" last season.

45
by Craigers (not verified) :: Fri, 01/27/2006 - 10:42am

“Suddenly, it’s ‘Flip Right Hawk 62! Deuce Right, Waggle 15H Throwback C-Posts! Dice Right 218 Bastard! Yes! Yes! Yes!’ And I’m like, Peyton, we’re at ‘third and inches’ here, honey. Please, just ’snap the ball.’�

Oh, but when he starts doing his "chicken dance"... ohhhh baby, it's magic.

46
by MJK (not verified) :: Fri, 01/27/2006 - 12:08pm

Re 44: Damn it, why do you have to make me laugh so hard when I'm at work?

Re 32: Your list only includes coaches who went twice as head coaches. I imagine it gets much longer if you include times a coach went with another team as a coordinator as well, and actually would include some wins. Belichick, for example, won a SB with the Giants as the DC, and then with the Pats as a HC.

47
by Björn (not verified) :: Sat, 01/28/2006 - 12:29pm

You guys are forgetting about Clinton Portis' mom. He took her to the prom.

48
by NF (not verified) :: Sun, 01/29/2006 - 6:19pm

Seattle is clearly ranked too high because Hasselbeck's shiny dome makes him look like John Malkovich. Polling random people on the streets of Detroit is way better than this. 411 y0r 13a5e r be1ong 2 u5!

49
by John Cramer (not verified) :: Mon, 01/30/2006 - 1:12am

This Football Shrink has actually "beaten the spread" each year, which is better than "beating the spread or picking the winner," and what he has said in each of the past 3 years is actually kind of how the game played out. I've listened to a couple of his national radio reports and just heard another one Saturday. He predicted a "Tampa Bay blow-out win" 3 years ago and "close games around 3 points" in each of the past two years. The odds of beating the spread 3 years in a row are 1/8. This year he says that Seattle is a stronger team and should win. Since Pitt is favored by 4, if Seattle wins his estimates of performance would only happen 1/16 times by chance ... which would actually be pretty impressive. We'll wait and see. If he gets it right again ... I might start believing .. but in the meantime - Go Steelers!