Monday, February 4, 2013

FRANCO is unhinged

I don't mean to scare you, but I'm ready to strangle the next person who uses the word momentum to describe the course of a football game.

First, some back story.

This guy, Brian Burke runs a very credible site called Advanced NFL Stats.  He also contributes to the New York Times, and put up this blurb before the big game.   In it he says that the 49ers are the safer pick, simply because they're the better team and there isn't much to sustaining game-to-game hot streaks, or "momentum"-- contrary to what Ravens' backers would prefer to think.  His theory didn't pan out, but his methodology remains valid.  Specifically, I call your attention to this bout of reasoning:
I might be going off the deep end here, but I think fans and analysts alike are susceptible to the idea of momentum because our brains are geared to detect patterns from nature. Football teams and their win-loss records are very, very abstract constructions, but our brains aren’t built for such abstractions.
Boom.

As for his theory being in tact, I would say he makes allowances for things like special teams play being a game changer, he just refuses to account for it in predictions due to its flukey nature.  Baltimore fails to pick up the 108 yard return at the beginning of the second half, and it's SF lifting the trophy.  Baltimore fails to run one back on the Steelers in the regular season, and my parents never meet, I'm never born, you're reading a better blog right now.  Special teams are flukey!

So anyhow, Brian Burke's talking about making predictions week-to-week, and discounting hot streaks as an indicator of future things.  How about using momentum as a point of analysis inside a game?  This is where I'm going to take over.

Momentum is mass times velocity.  It's some physics shit that says my bowling ball is going that way in a hurry because it's heavy and I rolled it.  You could use the word to describe the physics of a football game.  "That fatty is running at that other guy really fast, and when he hits him it's going to hurt because of all that momentum!"  It is NOT some abstract quality that shifts back and forth to whichever team is putting together a good drive at the moment.

Well why not, FRANCO? 

I'll tell you!  Because as soon as we start using the word to mean whoever looks good at the moment, then it's completely meaningless.

But what about lead changes, FRANCO?

What about them!  They're changes in the lead, not "shifts in momentum."  If you want some perspective on how ludicrous this phrase sounds, try using it to describe the weather.  "This snow storm sure has a lot of momentum!  It's going to keep snowing until there's a shift in momentum and it stops snowing."  If that somehow enhances your understanding of meteorological patterns, may I recommend you do your sudoku's in pencil.

Back to the Super Bowl.

I spent the first half on the road, so the game was brought to me over the radio.  Boomer Esiason was the color guy, and he must have said "momentum" a dozen times.  He also maintained that the 49ers were going to come back and make the game very exciting.  On this point, I agreed completely.   Not because Baltimore was going to somehow lose their mental toughness or because San Francisco was going to be psychologically healed, but because the 49ers are fucking good at scoring points.  They're playing in the damn Super Bowl because of this!  If I was a science man, I might think of this as likely regression.

Anyway, the Ravens do their fake field goal thing, it fails, but the 49ers are pinned on their own 6 yard line or something.  Boomer talks about how all the momentum is back on San Fran's side.  49ers go 3-and-out.  Then more instances of this, which I won't document here.  Beyonce comes out, the momentum is on her side, then it shifts to Destiny's Child, then she retakes the momentum to finish strong.  Then the second half starts, and Boomer tells me that he strongly believes the 49ers are ready to go, that they're about to steal the momentum.

11 seconds later, and he is telling the world that he is stunned.  No kidding.

Hey! Then the power goes out!

Around this time, I get to my destination, which is my aunt and uncle's house, where I meet up with them and my parents.  They're watching the game, which has just resumed, and they're talking about...wait for it... momentum.

I'd like to brag that I'm related to smart people, but last night they were not bringing it.  My aunt was going on about how she was sure the 49ers were about to take over the momentum, and my dad thought that was a pretty convincing analysis.  Always on the ready, I whipped out my laptop and showed them a live WPA graph of the game, pointing out that Baltimore had reached a solid ~95% of winning the game after the special teams TD.  I went Joe Namath on these poor people, guaranteeing everything and trying to lick everyone.

Then SF scored two TDs and everyone was like "See?? Momentum!"  And you know.  I did my thing: I  pointed to a fucking graph, talked about regression not being enough to overcome the odds, muttered something about Nate Silver wanting to see all of you burn in hell.

Hey, what do you know.  Math was ultimately right.  It got dicey there for a minute, but when you're looking at ~95% odds of something happening, it's a pretty good vantage point for making guarantees and not giving a shit about abstract shifts in aura or feel.

So I complained a bit to Nils, and then went to bed confident that we have the technology for a better world, but not the population to employ it.  Woke up this morning, ready to get on with my life.  That's when my aunt greeted me with "Now do you believe in momentum?" She then brandished her New York Times, which I now understand to be a liberal claptrap. 
In a sporting event that has had spectacular finishes as well as an infamous wardrobe malfunction during a halftime performance, the electricity at the Superdome stole the show on Sunday night, interrupting the third quarter for more than a half-hour and seemingly shifting the momentum of the game in a dramatic way.
 The equally leftist and untrustworthy Yahoo! Sports was also on board with this nonsense. 
But it's hard not to intertwine the blackout and the complete momentum changed that happened afterward.
Oh my god, people, it's so easy to be reasonable about this!!!  First of all, I do believe halftime and the blackout helped the 49ers.  Jim Harbaugh is a smart asshole, and with his team getting the shit beat out of it, he needed every extra minute he could buy to regroup, analyze, and make strategic adjustments.  No one mentioned how an intermission or two could, in this specific way, be an advantage to the team needing to make the adjustments. 

Secondly, we need to look at this momentum shit again.  Just because the 49ers scored points did not change their likelihood of winning by much.  If we wanted to use the M word correctly, we could posit that the Ravens had built up enough momentum in their likelihood to win, that the 49ers were incapable of stopping it, even if they managed to slow it down.  (I feel dirty.)

Hey, look at a graph.  I'm not embedding it here, but go back to Advanced NFL Stats.  That charts the likelihood of Baltimore winning the game after any given play.  Being as possessed by math fury as I am this morning, I averaged all the little points on the graph before the power surge and again after.  Baltimore stood a 74.7% chance of winning before the lights went out, and an 80.6% after they came back on.

We all love to think that anything can happen in a contest.  That's why we watch.  But sometimes shit gets insurmountable.  Regression is not momentum.  Regression is regression.  Momentum is some physics shit.

Next up on FTC: how to use the word "surreal" correctly.

No comments: