Continuing the Conversation …

There’s too much math for me to completely understand, but I get the theory.

Apparently, as one gets more readers, one finds that others respond to one’s post with their own. Earlier this week, The Common Man did so, and earlier this morning, Bill, over at The Daily Something, did so as well. This is the sort of thing I wanted to be involved with when I started — a conversation between rational people about a game I love. Bill responded to my post on the battle between sabermetrics and traditionalists. If you haven’t checked out his blog, you really should.

Anyway, in his post, he makes his case for “compassionate sabermetricism” (a good way to put it, I think), where one doesn’t go about the name-calling and just tries to find the answers. In a lot of ways, I agree with him. I tried to sit the fence a little in my original post. I didn’t want people to think I sided with one side or the other to force them to look at the big picture. Simply, I wanted people to realize that there are, in fact, two sides with logical reasons, but the way in which they treated each other was getting in the way. By doing so, I left out where I think the game is headed.

About two years ago, I really, really started paying attention to baseball (as opposed to just really paying attention and knowing who the best players were and followign the Braves), looking at everything from statistics to how to manage a team and its payroll (and etc.). I had never heard of sabermetrics before, and I was a little overwhelmed by the thought of it. How could someone take such a beautiful game and turn it into a pile of numbers? But as I began looking through it (one of the few times I can truly say that I had an open mind about something), I began to see a lot of good things about it. Like Bill said, traditional stats really don’t take much into context. They’re bare numbers that must be applied somehow, and many of the advanced metrics coming out, do try to put them into context. But here’s my question and the crux of my point — if you make a new metric and it still doesn’t take everything into context, is it a better tool?

The answer is yes. It at least gets closer, and theoretically, it will continue to get ever closer to the truth. For the most part, I think sabermetrics has the on-field, bare numbers down. Most of the hitting metrics describe the difference between good and bad hitters fairly well. Same for pitchers. Defense? Well, it’s getting there, but that’s a hard one. However, if you continue to miss something (like missing a step in a math problem), you will never quite get the right answer. Here’s where I think sabermetrics misses, and traditionalists hit (well, sort of but bear with me).

As I said, sabermetrics has all the bare numbers down, and for the most part, they’ve put those bare numbers into supernumbers that put a player’s performance into context and evaluate it. But I think they are still missing something, and that something prompted my first post. Sabermetricians have had to fight for what they’ve gotten, and in many ways, it was against the traditionalists. My theory is that, in trying to make their argument, they have discarded everything (well not everything, but a lot) to do with traditionalists. Sure, they’ve kept home runs, RBI’s, and runs, but they put those numbers inside OPS and run production, which is fine. All of that, I believe, gets us somewhere. But the point of these new stats is to measure value, the true value of a player, in order to define who the best player is, how much he should be paid, and when it’s time to say good-bye. If you continue to leave something out of that measurement (something that traditionalists use and have been discarded by sabermetricians), will you ever get to that all-knowing stat?

That is why I was calling for a compromise. No, I don’t think traditionalists have a better argument than sabermetricians. My logical, rational mind won’t allow that when given evidence to the contrary. But what if the traditionalists have something right, and in a fit of tearing them down, sabermetricians dismiss it? Eventually, they’d probably figure it out, but why not do it now?

What is this something? I’m not completely sure, but here are a few of my thoughts. There are “mental” or non-physical aspects to this game (I think all of the phycial ones have been questioned, subpoenaed, and cross-examined).

– Stolen bases have been shown to not help as much as previously thought, but what happens to a pitcher when a speedy runner gets on and threatens to steal? Does it affect him? If it does, shouldn’t that figure into the player’s value?

– If a player helps another with a hitting stance or approach, should that added value (or at least, part of it) be tacked on to the first player’s value? For instance, Brian McCann has long extolled Chipper’s mentoring. McCann’s OPS was .797 in the minors. In the majors? .858. Should Chipper receive some credit for helping McCann improve?

– Is there such a thing as team chemistry? We wonder which came first — the chicken (team chemistry) or the egg (winning) (or is it the other way around?). But what do we do about it? Mostly, I’ve seen it dismissed. But can’t we get a psychologist or two to follow teams from Spring Training through the end of the season? Then, rinse and repeat a number of times to see if we get an answer?

– Is there such a thing as leadership? If a player can motivate his teammates, shouldn’t he get extra points? If he’s a distraction, shouldn’t he lose points? Are players really emotionless drones that can just put that aside when game time comes?

There are more of these, but I think you get the point. There are things that are hard to quantify. But just because they are hard (maybe even impossible, but if Bill James got onto it, I see great things), it does not mean they should be dismissed.

But are these things negligible? I know you’ve probably been screaming this for about eight paragraphs. I’m not sure, but I don’t think anyone else is, either. Mainly, I think it’s just been thrown to the side and assumed dead. But if sabermetricians do that, are they better than the traditionalists, throwing away potentially helpful information? These are hard things to quantify. Maybe they won’t lead us anywhere. My argument, though, is that we try.

Bill is right. The answer is more metrics and more research. But if we fail to research everything, are we ever going to be right?

Traditionalists are a dying breed. Newspapers are going under. More readers convert to websites and blogs. More convert to sabermetrics because, let’s face it, it is the future. It’s progress. But does progress always include the positives the old-timers had?

Advertisements

6 Responses to “Continuing the Conversation …”

  1. Zach Sanders Says:

    I want to address some of the points at the end.

    1. I believe there was a study done on how pitchers do with threat on first. I can’t find it now, but if I remember correctly, it said that it really didn’t affect them too much.

    2. The Chem-Winning debate is a great one. I will say this. If you put bad chemistry players on a good team, chances are they will shape up because they are winning. Winning cures most ills.

    3. Leadership will not be measureable in statistics until a high level math professor (eg. Charlie from Numb3rs) takes the time to cruch numbers with extensive formulas.

    You do need scouts, no question about it. Unless we start breaking down players mechanics with numbers (which will happen at some point), scouts will need to be there to find these things, tell coaches, and then correct them.

  2. tHeMARksMiTh Says:

    Those were just a few things to throw out there to ponder. I don’t doubt that some studies have been done, especially in regard to looking at statistics in a different way. But I still think there are some psychological (that’s the word I was trying to think of before) aspects that are overlooked. How much does a manager really affect the game? Does placing in the order have an effect on players (I know the studies that say that lineup order doesn’t matter, but I wonder if players change certain things in different situations depending on where they hit)? There are lots of questions like this.

    In some ways, I see this almost as a situation similar to defense. It’s hard to quantify, and thus, largely ignored. But it still plays a role (maybe not as much as batting, pitching or defense), and isn’t .1 or .5 on win value enough to make it worth studying?

  3. Ron Rollins Says:

    Careful, Mark, you’re talking about – oh, no – ‘INTANGIBLES’!!!!!

    And of course every sabermatrician in the world will tell you they don’t exist. In fact, they will curse you and call you names for even suggesting that they might exist.

    Here’s a study I want to see done.

    How do players do when they have stolen the signs and know what pitch is coming, versus not knowing which pitch is coming.

    Because wouldn’t that just change everything?

    Great discussion beteen the three of you. It’s good to see stuff like this.

  4. Bill Says:

    There are definitely sabermetricians out there who will tell you that “intangibles” don’t exist. But I have no doubt that there are things like that–leadership and so forth–that affect the game in very real ways. The thing about them is that they are, by definition, intangible, so there’s no point in trying to measure them. You make your little adjustments based on what you observe and think you observe, but you can’t go building your team around things you can’t measure. Put the best team together that you possibly can, and then if Chipper Jones turns out to be not only one of the best third basemen on the planet but also a great mentor and leader, then bonus points for you.

    My real problem with this line of thinking, though, is that traditional-stats guys try to claim “intangibles.” Like, “the stat guys don’t get what it really takes to win–the little things and the intangibles–but we do.” And that’s unfailingly baloney. We’re all “stat guys”–it’s just that those guys are using batting average and fielding percentage instead of whatever we’re using, and when their stats don’t measure up, the fallback is “oh, but you can’t measure what he gives you in [leadership, etc.].”

    Case in point: a couple weeks ago, Joe Morgan was talking about somebody. I don’t remember who. And I’m making up the numbers, but what he said otherwise went exactly like this: “Yeah, his on-base percentage is only .320. But let me tell you this: baseball is not about numbers. And this guy is a great baseball player. He has 7 home runs and 24 RBI and he’s hitting .335 with runners in scoring position.”

    Seriously, that’s exactly how he said it. “Baseball is not about numbers–now here are some NUMBERS that I can throw at you to prove my point, and if I say it eloquently enough maybe you’ll think I’m talking about intangibles or the little things or something when I’m really just making a bogus statistical argument.” I’m sure if you watch the Joe ‘n’ Steve show tomorrow night on ESPN during whatever game happens to be going on in the background, you’ll see more examples of the exact same thing.

    I guess the point is this: yes, I absolutely believe those things exist, those things that sabermetricians can’t measure, and that they have an effect on the game. But the traditional counting stats guys can’t measure (or whatever the equivalent of “measuring” is in this case) those things, either. They are, by definition, unknowable. So it’s hard to see how they favor either side in this debate.

  5. Zach Sanders Says:

    Some intangibles I want measured by stats:

    1. The effect of pre-game meal choice on player performance.
    2. Players home life.
    3. Amount of sleep vs production.

    Anbody who says intangibles don’t exist is ignorant. Some of the greatest baseball minds have been masters of intangibles (i.e. Schuerholz). Bill is correct in saying that you can’t build a team on them, for the most part. Once you know a guy is a leader, you can count on that when signing him. But, you can’t predict them like other stats, you have to wait and see.

    I never hear this on MLB draft day: “I think he’ll hit .300 in the Majors, and add 2 wins a year with his leadership ability”. Never will hear it either

    And keep in mind, this is coming from someone who considers himself on the statistics side of things.

  6. tHeMARksMiTh Says:

    Okay, we throw around “intangibles”, and, as Bill says, by definition they are unknowable. But are they? Is it just a misnomer? How do we know that they can’t be measured if we don’t try? How do we know they don’t have much of an effect if we don’t even look at it? To me, this is the exact same thing as saying “We don’t include defense into the MVP discussion because it’s obviously nowhere near as important as offense.” After looking into it, we’ve debunked that. Why can’t we do that for “intangibles” (that was the other word I was searching for yesterday)? Maybe they are negligible, but until we look into it, we may never know.

    The point of writing my first post was that all the name-calling was causing us to just throw these things away without even looking at them.

    But let’s say you have a guy worth 3.4 wins and a guy worth 3.0 wins that you are considering. Player A is a terrible distraction (Manny-like) and is worth -.3 in leadership. But Player B is John Smoltz-like and worth .2. Now Player B is worth more than Player A.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: