The Human Element

But aren’t these statistics supposed to predict the future? They didn’t see the Padres coming! Not exactly. The idea is that, after identifying the problems above and answering them to a degree, the new statistics can better predict what’s going to happen. They never said they can predict with absolute certainty what will happen. Injuries, performance, randomness, etc. fluctuate due to all sorts of factors, and there is no way to definitively state what will happen. However, sabermetrics can say with a greater degree of accuracy what will happen. This is important when fans evaluate GMs … and even when GMs evaluate themselves. If you have to make a decision and we’ve agreed that nothing on earth can definitively state what will happen, you essentially have to take the option that has the greatest chance of occurring. It’s playing the odds like you would anywhere else. And yes, something else might happen, which is why we say ODDS, but that does not mean that the idea/theory is wrong. The hope is that your team will make more good decisions and that they will outweigh the bad decisions other teams are making.

But I don’t care about the future. My statistics say what actually happened! Well, not exactly. Lots of people have discussed why pitcher wins, ERA, BA,  etc.

This has nothing to do with anything, but it came up when I typed in "regression to the mean" into Google.

don’t really measure what we are told they measure. We’re told they mean one thing, but they actually mean that thing along with a few other things that combine to mean something else. I talked about this in reference to Joba a few weeks ago, and if you want more of an explanation of the topic, I suggest you go there before you read any further. The short explanation is that sabermetrics DO measure what has happened. We often talk about regression in association with these statistics, but that is simply to aid in understanding the new statistic in terms of the older ones (In my opinion, this was done with the best intentions, but I often wonder if it worked out long-term. Sabermetricians tried to relate the new stats to older stats in order to make comprehension easier, but when they did so, it created another misunderstanding due to the somewhat complex nature and some propaganda. I don’t know definitively that it was detrimental, but it’s just a thought). For example, “His ERA will regress to this because of his FIP”. The pitcher is pitching to his FIP, and we expect that his ERA will eventually reflect that FIP. People think that means that FIP indicates what a pitcher should have done because they still see ERA as the measurement of pitcher performance, but FIP actually measures pitcher performance. ERA regresses because defense and luck even out over the course of many years, leaving only pitcher performance. But the new statistics DO measure current performance, but because they are better at doing so, they are also better predictors of future outcomes. Sabermetrics, in addition, has created several PROJECTION systems such as PECOTA, ZIPS, etc., but they include newer statistics along with historical trends of similar players in order to try to predict the future. But they usually tell you that they are low, medium, high, or the percentile of performance, meaning that they recognize the uncertainty. What this means for you is that you’ll have to learn some new acronyms. Sorry, but it’s not that hard.

But they don’t include intangibles! Okay, I said I’d leave this alone, but I won’t go too far. Most sabermetricians won’t completely disregard intangibles, but they will warn you that they may not affect the game as much as you think they do. The caution is due to the uncertainty. We don’t know how they affect certain players, and they don’t affect each player equally. Ascribing a value to it is arbitrary and not helpful. In addition, intangibles are difficult to determine because we’re talking about the top percentile of all baseball players, and what would affect us may not affect them. Sabermetricians aren’t atheists. They’re more agnostic.

But I just want to watch the game! Sabermetricians do, too. Just because they can manipulate formulas does not mean that they do not appreciate human skill. The desire to appreciate that human skill is the reason that they do what they do. They want to make sure everyone is appreciated in the way they should be because they know this game is hard, but they want to give credit where credit is really due. They know few can do it. It’s their appreciation of that human skill that brought them to the game. It’s watching the game that got them hooked. They don’t have a lesser appreciation for the “human element”. They just see it differently.

But I like the human element. Don’t boil baseball into a bunch of computers! Ah, we’ve arrived at my main point. Sabermetrics was not created by a bunch of computers. It was invented by humans. Humans realized the logical disconnects. Humans set about trying to identify the problems. Humans identified the problems. Humans came up with the methodology to solve the problems. Humans came up with the formulas (yes, computers do the calculating, but humans could, too. Computers just do them faster). Sabermetrics should be a CELEBRATION of the human element, not a detriment to it. People often ask what the difference between humans and animals is. This is it. No, not sabermetrics specifically. CRITICAL THINKING is. The ability to think deeper is our greatest achievement. Our behavior is animalistic, but where we distinguish ourselves is in the ability to reason. Sabermetrics is simply an example of it.

And celebrating it does not mean simply embracing it. Ask questions. Demand better research. You should because humans need to be pushed to reach our potential. But don’t interrogate simply to prove someone wrong. Ask questions because you don’t understand. Ask questions because there are things that haven’t been discovered yet. If we can have civilized, intelligent discussion, it can go farther. And then, it doesn’t have to be the “sabermetric” movement. It can be a HUMAN movement. I realize this sounds a bit idealistic, and that’s okay. It probably is, but I have also resigned myself to the fact that some people just refuse to change. But I also think that the reason some haven’t changed, that some refuse to listen is because it’s never been explained in a good enough way. And that’s what this post is about. It’s about trying to explain what’s going on in a different way because it might, just might, hit home with someone. I think we give up too easily sometimes. It doesn’t take much to reach frustration, but I think frustration is a poor reason to give up. So, I’ll continue to try.

12 thoughts on “The Human Element

  1. The majority of the issue isn’t with sabermatricians over the old or the new. There has always been that debate in baseball over every issue, in every season, and on every play.

    The issue most people (to include myself) is the arrogant attitude of the majority of the saber crowd, where they claim that they are the only ones who right, anyone who enjoys the old stats are idiots who don’t know anything about the game, and that the game can only be measured in terms that they champion.

    And all 3 assumptions are wrong.

    They are not the only ones who are right. There are many ways to evaluate a player, and sabermetrics, while most probably the best way, is not the only way.

    Just because I like pitcher wins and rbi’s doesn’t make me an idiot. I know that they are only numbers, but that doesn’t mean they can’t be used in some contexts.

    Just because they claim they are right doesn’t mean they are. Baseball is a very subjective sport, and each fan has the right to use what he feels is appropriate, and not be abused for it.

    When sabermatricians gain more personality and the ability to deal with the entire fanbase, they’ll gain more converts.

    Of course, that’s just my opinon. Sadly, I didn’t use any part of the Quadratic Formula, I didn’t solve for ‘x’, and I didn’t show my work. So I’m probably wrong.

  2. I think Ron’s comment is useful, because it illustrates that much of the friction between the sabermetric crowd and the traditional crowd is based upon resentment.  People do not like being looked down upon, which is how traditional baseball fans might feel from the sabermetric community.  It’s like sitting at a table where everyone is speaking Russian and you don’t.  Similarly, while I cannot see into the minds of the people collectively referred to as “the sabermetric community”, it may be true that the sabermetric community resents tradition MSM for occupying great jobs while adding little insight.

    However, I’m not sure that “baseball is a very subjective sport” is a very meaningful statement.  In fact, I’m not sure what it means at all.  Of course fans may subjectively hold opinions about the sport, but if those opinions are based upon statistics that misrepresent or fail to tell the whole story of a player’s performance (such as a pitcher’s record), then the opinion is assailable, isn’t it?  If someone told me that Phil Hughes has had a better season than Felix Hernandez because he has more LNBs (leads not blown, aka wins), then that person is wrong. 

    And there is a sense in which baseball is the least subjective sport.  Over a 162 game season where ballplayers come to the plate hundreds of times, there is a repetition of events that allows us to compile enough data that we may meaningfully compare statistics to measure performance.  And this may be done objectively.  By contrast, it doesn’t make much sense to compare the number of tackles Kenny Phillips has versus Troy Polamalu because football games are short and few.  But every baseball season presents piles of data by which we can meaningfully analyze player performance.  Why not do so in the most objective way possible?  Fans certainly have the right to analyze the game however they want, but if what if they use metrics that aren’t the most meaningful?  I may subjectively like Robinson Cano as a player because I think he has a great looking swing and he’s a pleasure to watch.  But if someone wants to know if he’s better than Pedroia, I’d want to look past his batting average and RBIs.

  3. First I'd like to say that this was a terrific article. This is my first time posting on the site, and I have to admit I read this site regularly despite not even liking the Yankees because of how much I love and respect the writing and analysis that's given.

    "In my opinion, this was done with the best intentions, but I often wonder if it worked out long-term. Sabermetricians tried to relate the new stats to older stats in order to make comprehension easier, but when they did so, it created another misunderstanding due to the somewhat complex nature and some propaganda. I don’t know definitively that it was detrimental, but it’s just a thought."

    I disagree that it was detrimental for sabermetricians to scale new stats to match older ones. I understand why you would wonder if it was, but I really don't think there's any way scaling did more harm than good, assuming sabermetricians wanted their findings to eventually become more mainstream in the baseball world.

    From personal experience, I can certainly say that I had no idea what I was looking at a couple years ago (when I first got into sabermetrics), and so I ignored them for a little while. But I found that certain writers I liked reading (like Rob Neyer, who appeared very contemplative in nature) kept using these numbers, so eventually I did a little homework to figure out what some of these numbers meant, and things then fell into place because I was able to have some frame of reference for what the numbers were trying to tell me.

    I've now become a lot more interested in sabermetrics overall, but I'm not sure that would have happened without the scaling. Old stats like ERA are very simple to understand and calculate yourself, so that number can actually hold meaning to a non-genius fan. To our brilliant minds at Fangraphs and all those sites, I'm sure they could make sense of stats like FIP without scaling, but those numbers would have little meaning to people outside of that small circle. And without understanding what a number is even supposed to mean when you see it, anyone but the most mathematically brilliant people would never jump on board the sabermetric train at all.

    I also think that while there may be some confusion (you addressed those confusions pretty thoroughly in the article) about what sabermetrics are trying to show, overall the scaling has allowed an evergrowing number of people to climb on board because they can actually understand BOTH what the numbers are trying to tell you, and what flaws in the older stats they attempt to correct. This has allowed me to explain to my friends why I've started using these stats and they've also begun to come around as a result.

    And for the group of people that appear to thoroughly misunderstand what the newer scaled stats mean, I think most of them are probably people that would reject such stats no matter what.

    So long story short: While sabermetrics are still despised by some, it's a movement that is clearly growing. And while this shift from old to new stats may have happened anyway, I would guess that it's happened MUCH faster because of the decision to scale certain stats for the benefit of people like me. There are a lot more baseball fans like me (reasonably intelligent, but not a math whiz) than there are geniuses.

  4. “The issue most people (to include myself) is the arrogant attitude of the majority of the saber crowd, where they claim that they are the only ones who right, anyone who enjoys the old stats are idiots who don’t know anything about the game, and that the game can only be measured in terms that they champion.”
    This is a total strawman. I don’t think I’ve ever heard any saber proponent claim that you don’t have the right to enjoy the game however you want. However, when you decide to inject yourself into the conversation/debate, whether by commenting or blogs or plying a trade as a baseball writer, that’s another matter. You can have your opinion, but you have to defend it. You can’t just fall back on, “I’ve got a right to my opinion and you’re all mean bullies if you think my opinion is dumb” if you choose to put your opinion out there. You can’t have it both ways, in other words.
    “Just because I like pitcher wins and rbi’s doesn’t make me an idiot”
    It depends on what you mean by “like.” If you mean, “i think it’s quaint and old-timey and makes me smile and I just like to see it in the box score after I watch a game,” then more power to you. If you mean, “I think they’re a good way to evaluate a player’s performance objectively,” then yes, it does make you an idiot. I’m sorry, it just does.
    Put another way, I respect your right to have an opinion and to enjoy the wonderful game of baseball however it is that you get the most enjoyment out of it. Hell, I encourage it! I really do. I want as many people as possible to enjoy the game, to watch the game, to have their kids playing little league, etc. But that doesn’t mean there’s a right to put yourself in the conversation and throw out opinions based on flawed/un-scientific statistics and not have that opinion challenged or to have anyone else take it seriously.
    Look, I don’t hate on “traditional” stats because I’ve got a personal investment in newer numbers or because I want to buck tradition. I hate on them because, by and large, they’re un-scientific “statistics” that are terribly flawed on their own merits. Take batting average, for example. The problem with batting average is that it doesn’t account for walks at all, and treats all hits as though they’re the same. That means there’s a lot of aspects of “hitting” that it doesn’t incorporate, nor does looking at a player’s batting average give you a very good indication of how good of a hitter they are, because it gives you no indication of their overall ability to get on base or hit for power. And obviously, those are pretty important components to hitting.
    “Of course, that’s just my opinon. Sadly, I didn’t use any part of the Quadratic Formula, I didn’t solve for ‘x’, and I didn’t show my work. So I’m probably wrong.”
    A word of advice; when you start your argument with the premise that the other side is arrogant and unjustifiably dismissive of your position, it’s probably best not to conclude with a passive-aggressive ad hominem.

  5. Interesting points being made.
    I mostly, though not surprisingly, agree with Brien. You’re allowed to think wins and RBIs are important in the grand scheme of baseball history (it’s part of history, how it has affected our view of baseball, etc.), but RBIs usually aren’t used in that context. Articles need to be written about the power of the RBI in our minds and how it’s shaped our thinking, but if you want to talk about who is the better hitter, there are so many better stats than RBI that using RBI doesn’t help your case.
    One point that is made that I very much agree with is attitude. One thing I will note, however, is that it comes from both sides. I don’t know who started it, but I don’t think that matters. Both sides definitely attack the other side and “talk down” to them. I don’t really think it’s partial to sabermetricians, though I agree that their style is often inferior to journalists (hey, what do you expect from math majors?). Style is important. The way you come across in explaining things will influence the audience’s response to it. I tried to be as helpful as I could in this post at doing just that, but you all have to be the judge on that.

  6. Swift,


    I'll first second (that was confusing) Brien comment on the double space in between paragraphs. (Using double space)


    Next, I think you make a great point about scaling. Scaling has obviously caused some issue in regard to comprehending the actual measurement vs. what the player should have done argument, but I think you are right in that the benefit of scaling has outweighed that. I brought it up, and I'm glad you answered.


    And you're allowed here as a non-Yankee fan. The Yankees are probably my least favorite MLB team, but I write here because, as you mentioned, I have an immense amount of respect for my fellow writers on the site, who do a great job in covering a wide range of topics. Still doesn't mean I won't take a cheap shot now and then. :)

  7. Thanks Brien and Mark.
    I agree, Mark, that the scaling has caused some confusion, but those that do suffer from that confusion generally are going to fall into one of two categories:
    (a)  Those who aren’t that interested in newer stats in actuality and are just looking for some way, any way, to demonstrate the flaws with them (of course while simultaneously ignoring all the even bigger flaws with the older stats they hold dear) so they can go on pretending like sabermetrics is some far-out-there craze by people who have their head in the clouds and are obsessed with computers.
    (b)  Those who are truly interested in learning better ways to evaluate players, but have problems with certain newer stats because they are still towards the very early stages of learning about sabermetrics and understanding what they truly are.
    In the case of (a), they’re not much of a concern for sabermetricians, since people in (a) are basically the same as fans (and writers) who ignore sabermetrics altogether and aren’t truly open to learning how to objectively evaluate players at all. In a way, this group reminds me of creationist “scientists” who refuse to accept evolution.
    And in the case of (b), the fans who really are trying to learn, they eventually will come to “get it” after a while. I personally also started out thinking of some of these stats as projections, and it took me a few months of reading articles at Fangraphs and other sites for me to learn much of what this article is trying to teach. And actually, it might have been an article similar to this one which pushed me over the edge. Articles like this seem to pop up once a month or so on this or that site, and for people that have been involved with sabermetrics for years, such articles probably seem repetitive and unnecessary. Even I feel like I’ve read (some version of) this article at least a dozen times before (not to take away from it, it’s very well articulated and probably the most comprehensive one I can remember).
    But I think the real people who benefit from articles like this one are the guys who are still in “transition” from group (b). And since the sabermetric community is growing all the time, it’s a good thing that people do keep writing this kind of article so that it continually shows up at the top of those websites they are probably just starting to read.
    There might be a group (c) that has people who are genuinely interested in sabermetrics but will never transition from seeing them as projections. These would be the (so-called) victims of sabermetricians’ decision to scale their newer stats. But I frankly think that the number of people who will be permanently stuck in this group is really small. The overwhelming majority of the time, I think the people you deal with will fall into (a) or (b). With (a), just shrug your shoulders and don’t worry about them. And for those in (b), all we need is a little patience.

  8. So let me get this straight Brien. I say the problem with the saber community is their arrogance and willingness to call people idiots who like the old stats, and then you call me an idiot for saying I like wins and rbis. 

    Thanks for proving my point.

    By the way, what I said was “Just because I like pitcher wins and rbi’s doesn’t make me an idiot. I know that they are only numbers, but that doesn’t mean they can’t be used in some contexts.”

    Don’t be afraid to actually read the comments people make before insulting them about those same comments. And for the record, I’ve been following sabermetrics for 30 years, and probably know more about them than you do.

    Another thing, sabermetrics is not science. It is mathematics. While many fields in science use math, mathematics can, and does, stand on it’s own in my areas, to include actuarial tables, time tables, and  yes, even statistical analysis.

    Also, even though I was a Political Science major, I don’t feel the need to use poly-bable. Write to your audience, not your vocabulary level. Baseball isn’t the place to be pretentious.

    Also, my ‘not solving for x’ statement was a joke. And a good one. Go on You-Tube and watch “Who’s on First”.  You’ll love it.

  9. Well, i hedged a little. But yes, if you think you can evaluate individual performance with wins and RBI’s, that’s just ignorant, and there’s no excuse for it given the plethora of examples out there of people explaining why those are flawed stats. I respect your right to an opinion, but you don’t have a right to have the substance of that opinion challenged if you refuse to acknowledge basic rules of using statistics. You can call me arrogant if you want, but frankly, it only makes me think less of your argument. Not because it’s insulting, but because it just doesn’t speak very highly of you if you actually think people challenging misplaced beliefs about older statistics using the basics of statistical analysis is “arrogance.”