Talking Football With Stat Geeks
Aaron has found the perfect film clip to describe what it feels like to talk football with a stat geek.
By PackerAaron
Longtime readers of this site know my love-hate relationship with football fans who reach for "innovated statistics" when arguing a point about something that happened on the football field. More often than not, this exchange will manifest itself with a devotee of the (admittaedly fantastic) site Football Outsiders.
I won't rehash my inherent mistrust of those that instantly reach for numbers when talking football (You can get my thoughts here if you'd like to get up to speed) But earlier this week I had the misfortune of getting caught up in a spirited argument about the Packers offense and I found myself thinking about this moment from Ghostbusters. It perfectly captures the frustration of talking football with people who insist on seeing every argument through the prism of DVOA or DYAR:
Now, lets be clear. "Innovative statistics" have their place and can be an invaluable tool, when used correctly. But they are not the be-all-and-end-all of football analysis. And I don't think Aaron Schatz and company think that they are - but boy, some of their readers sure do...
PLEASE SUBSCRIBE TO OUR CHEESEHEAD NATION WEEKLY NEWSLETTER HERE.
__________________________
Comments (28)
November 27, 2010 at 04:37 pm
Haha, this cracks me up. Right on point too. There's a guy on another forum I frequent who LOVES to point to DVOA as the be-all, end-all argument for everything.
Frankly, I don't care what the hell DVOA is or what it stands for. I go by what I see with my own two eyes. Football is not a game that can be reduced to stats like baseball as there's a lot of intangibles stats can't account for. I really wish people would stop trying to make it as such.
November 27, 2010 at 05:08 pm
It seems like the argument is either that stats tell the story or they don't. I agree, they certainly don't tell the whole story...but they also don't tell NONE of the story. It seems like people pull stats when it's convenient, then poo-poo them when it works against their beliefs. Football CAN be quantified to an extent. Again, I'm not saying that it tells the whole story. When Bill Parcells took over the Cowboys, there was a time when they had the number one offense and defense. I don't recall the outcome of the season exactly, but I believe they stumbled into the playoffs and must have lost quickly (considering they haven't won a playoff game in several years before last season).
It may not be an exact science, but it's at least a soft-science.
Without hard data, we may be able to see things the numbers dont...but we also may be clouded by emotion and "gut feelings." Data are valuable in football, but overall I agree - people can get carried away...but don't dismiss it completely either.
Rant concluding.......NOW.
November 27, 2010 at 06:08 pm
Not all stats are worthy of mention but if relevant stats are ignored then a person cannot objectively (uninfluenced by emotions or personal prejudices) analyze most performances (player or team) .......
AN has issues with stats because stats often prove his theories wrong ......
PackerAaron
November 27, 2010 at 06:32 pm
(rollseyes)
People like yourself like to use stats to "prove" that Brandon Jackson "sucks" - yet those who care to watch him, you know, actually play the game of football down in and down out - even when doing things that don't produce "stats" - know how valuable he is to the Packers offense.
November 27, 2010 at 07:36 pm
I knew this was all about BJ (your boy) & not really about stats ..... Actually, I've defended BJ more than once since Grant went down .....
But you're correct ..... Ignore all stats & BJ looks pretty good ...... Also, watch Nance get more carries again versus Atlanta ..... If Nance shows even more than he has thus far, BJ will be back to 3rd down back .....
That happens when you stutter-step your way to 28 yards on 14 carries ..... Geez, sorry, those irrelevant stats again.
November 27, 2010 at 08:21 pm
I'm just going to point out that B-Jax has a 4.0 avg, 3 tds, 231 yards receiving (good for 4th on the team), and one receiving td. I believe Nagler's article was "BJAX is good enough." Frankly, if you read that article again, I think it's hard to say he's been wrong. Jaxson isn't a stud, no one's claiming he is, and the stats bare that out...just sayin'...
November 27, 2010 at 08:43 pm
GB 20 .... Falcons 16 ..... Four minutes left in the game ...... Atlanta has just punted to GB after a stalled drive..... T. Williams fair-catches at GB's 15 yard line .......
Do you give the ball to BJ 3 times in an effort to gain a 1st & ultimately run out the clock? ..... I sure wouldn't.
November 27, 2010 at 08:54 pm
WoodyG - Neither would I. I guess that's what I'm saying. He's serviceable, which is what the stats would bare out. I don't think anyone's arguing he's an elite back who can take over a game. I, however, would argue he's a "good" overall back as his running is average, he's good catching the ball out of the backfield, and, in my opinion, is outstanding in pass protection.
November 27, 2010 at 09:08 pm
Tommyboy
We actually agree .... It's Nagler again who's distorting the reality .... Here's my quote from another earlier blog ....
"However, BJ is good with the dump-off passes in open space, doesn’t put the ball on the ground & most importantly is great at picking up the blitz & giving AR that split second he may need to find GJ ……
PackerAaron
November 27, 2010 at 09:11 pm
No - it was actually a pretty lighthearted post - one which you took upon yourself to say something snide about. What. A. Shock.
November 27, 2010 at 09:26 pm
Another one of your "corrections" after the fact ...... You can fleece yourself & others but not me ...... I was talk'in analysis, strategy & statistics when yo' momma was still wiping your ass ..... (Lighthearted post plus a few Leines) ..... LOL.
PackerAaron
November 27, 2010 at 09:37 pm
Aaron Schatz liked the post - that's good enough for me. (Sticks out tongue) ;)
November 27, 2010 at 11:09 pm
Well, if we agree, then really, all three of us agree. Then, WTF is going on? We REALLY suck at agreeing. At one point I felt like I was gonna hear someone pull a "HOLD ME BACK! HOLD ME BACK!"
Hmmmm...so let's see...my theory on this whole orgy of discontent is....oh look, a beer! I gotta go.
November 27, 2010 at 06:51 pm
I for one believe that emotions/gestalt gut feelings can be a hell of a lot MORE accurate than statistics if you fine-tune your "perceptual-judgment apparatus." Besides, regardless of whether you utilize innovative statistics, they still have to go through the filter of your own brain. The notion that using statistics will make you more objective is preposterous.
November 27, 2010 at 08:13 pm
Frankly, the idea that gut feelings make you more objective is just as preposterous. A balance, as usual, seems to be the most logical solution.
November 27, 2010 at 07:04 pm
"An unsophisticated forecaster uses statistics as a drunken man uses lamp-posts - for support rather than for illumination."
-Andrew Lang
"Trying to convince the public of something? Use statistics. You can make them say anything you want."
-Thomas Sowell
November 27, 2010 at 07:27 pm
Why even keep score, right? ..... Fans could call in their analysis to determine the victors of each game .... LOL
November 27, 2010 at 08:26 pm
I think all contests should ultimately be decided by interpretive dance. I simply canNOT be alone on this.
November 27, 2010 at 08:29 pm
You completely miss the point of the quotes.
Say for instance that the statement is "The Falcons are a good defense against the run because they are 6th in the NFL in yards allowed." That is a bad arguement because...
A) It says very little about whether or not they are good (only in a relative way) and absolutely nothing about WHY they are or aren't good....and
B) The Falcons are 21st in the NFL in yards per carry allowed. Another simple stat that is entirely contradictory to the previous.
Using stats as the ENTIRE bais of an argument (especially in football where there are hundreds if not thousands of variables) is simply weak and lazy.
November 27, 2010 at 08:35 pm
Honestly, I think you're misinterpreting the quotes. It clearly calls for more interpretive DANCE.
...k, I'll stop drinking.
November 27, 2010 at 07:35 pm
Statistics show correlation. Correlation does not necessarily show cause and effect.
November 27, 2010 at 08:24 pm
...of course, that totally depends on how you calculate the statistics. You could easily run ANOVAs or t-tests with a lot of the stats...or you could run correlational data.
...not that any of this matters as what we all see are descriptive statistics 99% of the time.
November 28, 2010 at 10:01 am
Variance analysis is a great form of statistical analysis with a million practical applications.
In football, stats are a great thing to use when looking backwards to explain previous results. But like anything else, they're not foolproof when trying to predict the future.
November 27, 2010 at 09:01 pm
Here's a stat (take it or leave it) ..... Both GB & Atlanta are on 4 game winning streaks ..... Atlanta has given up 10 TD passes in their 4 victories while GB has only given up 2 TD passes in their 4 victories .......
AR decisiveness & fearlessness in the pocket has returned ...... He gets 3 versus the Falcons ..... Bank on it.
November 28, 2010 at 12:11 am
I'm gonna sound like a broken record, but I'm just so relieved to see Aaron poised in the pocket again, and not scared of the rush. For a moment in the season I really thought we could've lost our future QB.
A scared QB is a bad QB. Ask David Carr...
November 28, 2010 at 12:09 am
Really? All this turmoil over the concept of stats? Guess I'll jump in!
I love stats. I'm a statistical whore.
But, no, stats don't tell the whole story. Some stats are really not important.
85% of people knows that...
November 28, 2010 at 08:39 am
Statistics are like Wikipedia: they are a good place to start, but you shouldn't depend on them for all the right answers.
Looking at statistics can reveal some things you might not have seen otherwise; however, once you find them, you have to dig deeper for what they mean and why they show what they do.
Statistics can be just as subjective as observation. It's all in how they're used.
November 28, 2010 at 11:21 am
Threads like these are when I just sit back and laugh at all you boys. ;)