The Limits of Stats

This is the best essay I have read in a long while articulately critiquing the obstinancy of baseball statisticians.

over 3 years ago Rev Halofan 13 comments 0 recs |

Comments

I thought this was a great read. Thanks for the link Rev. While I doubt it’ll really convince any of the “number crunchers” referred to in the article, the article really does a good job at blending how stats are indeed predictive and objectively important, but that the end of the day baseball is subjective and only limited to how we as the fans want to view it.

by HaloFanInDC on Oct 16, 2008 3:01 PM PDT reply actions

Good Stuff

I like how the article comes down to the winners being the champions in our hearts and minds, the heroes (or villains) of the story, not necessarily the best team overall. If these statistics and their impact mattered, we’d be crowning some team that made the best metrics. SI.com had an article about who’s got the strikeout staff, closer, whatever else wining teams have, and it was the Red Sox first, then the Angels, etc.

These things don’t matter as much in small series, and while that makes it arbitrary to some degree, it also makes it incredibly exciting.

by Kernel on Oct 16, 2008 3:11 PM PDT reply actions

Why not

just cut out all the meaningless stats? do they have a big enough impact on free agent signings and whether or nor to extend a certain player to keep them around? Or is it because lots of people will lose jobs because there will be no more number crunching?

to me it seems the game has gotten a lot more complicated than who went out there played harder and won the game. Yea, talent has to do with it but not numbers. I think as a whole baseball is already slow for outsiders and all the useless number crunching just makes it more of a reason for them to not watch. /

"Throw some CHED!"

by angelskid2210 on Oct 16, 2008 3:17 PM PDT reply actions

Fabulous read, Rev

Thanks. My favorite part…

“It happens because of great players, veteran leadership, tactical decisions, experience, and features from across the spectrum of what it means to be human, some of which are quantifiable, some of which are not.”

by ViolaHalo on Oct 16, 2008 3:55 PM PDT reply actions

my favorite part was this:

"My shit doesn’t work in the playoffs." B. Beane

by Downing Rules on Oct 16, 2008 5:11 PM PDT reply actions

As a numbers guy myself...

I don’t disagree with anything this guy says.

The numbers simply give us an idea of what we might expect. But, obviously, you can’t plug the numbers into a computer and have it spit out the results of the World Series. I think (and hope) that even the most hardcore stat-heads recognize that when it comes right down to it, the game is played on the field and that human emotion and error play a huge part.

I think when people like me say that the playoffs are a “crapshoot” (an overly broad statement), that’s basically what we’re saying. We’re admitting that you can throw the numbers out the door when trying to predict what will happen in the postseason. Yes, having a strikeout staff, a good closer, and good defense have historically made a team more likely to win, but again it comes down to little things here and there – who has a mental lapse or who happens to be “hot” at the time, for example.

It seems to me that the disconnect between the “stat-heads” and the “purists” is overblown. Maybe there are some stat-heads out there who don’t actually enjoy watching the game and really believe that the numbers are fully predictive of everything that happens on the field. But I think those people are to stat-appreciaters what Socialists are to Democrats (that is, on the fringes, and few and far between). Or maybe I’m naive.

by Dogman on Oct 17, 2008 8:55 AM PDT reply actions

Well said.

by snowhor on Oct 17, 2008 10:07 AM PDT up reply actions

Yo are being way too nice

Many stat-heads spend more time rationalizing why their numbers were not perfect indicators than they do actually seeking answers to the problematic scenarios of how to accurately measure greatness.

MANY of them.

by Rev Halofan on Oct 17, 2008 12:05 PM PDT up reply actions

At the same time, too many anti-stat-heads spend way too much time saying things like GA is a solid .285/15/85 guy without looking deeper to see that he is, in fact, no longer good.

It’s as if you’re either all stats or no stats. No common sense middle ground.

by snowhor on Oct 17, 2008 12:28 PM PDT up reply actions

Clarification: The perception is that you're either all stats or no stats, when I think most people are in some sort of middle ground.

And then we get caught up in pointless stat/anti-stat dialogues.

by snowhor on Oct 17, 2008 12:34 PM PDT up reply actions

I agree with you

I was going to write my opinion about this subject, but I knew it’d be rather long-winded, so I posted it on my blog. If anyone is interested, here it is.

Basically, I think most people are in both camps…they like and understand stats, and also believe there are elements to the game that can’t be measured, but have just as big of impact on the game as hits, runs, etc. So it kind of pisses me off when someone disregards statistics, or when someone thinks stats tell the whole story.

I think the article the Rev has referenced here suggests that, that there’s room for both.

I was uncool before uncool was cool.

by WiHaloFan on Oct 17, 2008 2:10 PM PDT up reply actions

agree with your take, WiHaFa

stats are the vitamin, not the meal.

by Rev Halofan on Oct 17, 2008 8:51 PM PDT up reply actions

Ridiculous.

What a long-winded and muddled argument!

“The words don’t predict or describe as well as precisely as the numbers, but that doesn’t mean they’re less important.”
That is EXACTLY what it means. Let’s put this in tangible terms. I have two players who I will sell your team for ten million bucks to DH, and you have to buy one of them.

Scenario 1
I describe those players to you in the terms that Joe Sheehan is criticizing and the blogger Intricately Independent is defending.
“Player Alpha is a champion – he plays with a lot of grit, always hustles, and wears his heart on his sleeve.”
“Player Beta is a detached, reclusive guy. He doesn’t show a ton of emotion, and he can have issues with managers and people in the clubhouse. Sometimes he doesn’t hustle enough.”

Scenario 2
I describe those players to you in statistical terms and with numbers.
“Alpha had a .261 Equivalent Average. He hit 1 hr, and got on base at a .354 clip.”
“Beta had Equivalent Averages of .309 and .398 on two separate teams. Altogether, he hit 37 HR and got on base at a .430 clip.”

The two players are David Eckstein and Manny Ramirez. Which one is the better DH for $10 million? If you made your decision based on the information in Scenario 1, you’re picking Eckstein. If your decision is based on Scenario 2, you’ll pick Manny. In Scenario 1, you’re throwing money away for qualities that won’t win you games.

When we evaluate MLB teams, players, and performance, stats and numbers are essential. The stock in trade of announcers – grittiness and other cliches – has almost no value.

There is a human tendency to emotionalize everything. “Grit,” “heart,” “character” – these are terms that everyone can understand, and they appeal to people on a visceral level. They are also extremely misleading and, if relied upon, lead to bad decisions.

"It's just a tiny little nick, but it hurts when I get champagne in there."
- Jason Bay, on getting spiked scoring the winning run in ALDS Game Four.

by 0157H7 on Oct 22, 2008 10:16 AM PDT reply actions