They’re Not Getting Lucky

The St. Louis Cardinals, to date, have been the best team in baseball, at least in terms of getting wins and avoiding losses. It’s not surprising, in that the Cardinals have proven themselves to be something of a powerhouse, but it is surprising given some of their injuries. The Cardinals have the best record in the National League by two games. They have the best run differential in the NL by 32 runs. A big part of the explanation is pure talent. The Cardinals have talented players. There’s also the matter of timing. At the plate, with the bases empty, the Cardinals own an 86 wRC+, good for 24th and sandwiched between the Mariners and the Pirates. With man or men in scoring position, they own a 140 wRC+, good for first by a mile. The Cardinals have scored 35% of their baserunners, and that’s tops in all of baseball.

This didn’t escape Buster Olney’s attention. Olney has written and tweeted about the Cardinals’ ability to drive runners home, noting that they have the highest team OBP with runners in scoring position since at least 1974. A lot of this came up Sunday night, and a lot of the response, as you can imagine, is that the Cardinals have been lucky. Olney wasn’t buying this argument, and you shouldn’t buy it, either. “Luck” — what we’re dealing with here isn’t luck.

Look at the Cardinals’ splits. They don’t make sense. Good-hitting teams shouldn’t be bad at hitting in certain situations, and bad-hitting teams shouldn’t be good at hitting in certain situations. The Cardinals almost certainly aren’t going to keep hitting this poorly with no one on. Nor are they going to keep hitting this well in run-scoring situations. This is a situation that just screams out for regression, a somewhat complicated mathematical principle that’s weaved its way into ordinary parlance, and we can identify that by eyeballing it. We can declare, with an absurdly high degree of confidence, that the Cardinals’ splits are going to get more normal. But when talking about what’s already happened, there’s a need to be careful. There’s a need to pay attention to semantics.

In this situation, and in all such situations. I am far from the first person to write about this, but the word “luck” has been bothering me more than usual lately. I think people now are more responsible than they used to be, but they’re still over-aggressive, and it’s important to understand the difference between “luck” and “sustainability.” They are not the same thing, even if it feels like there’s an awful lot of overlap. Lucky outcomes are unsustainable, but not all unsustainable outcomes are lucky.

If one were to ask me, I’d say the Cardinals’ splits are unsustainable. They are not going to continue as they’ve proceeded to date. That is, they will not be sustained. But the Cardinals haven’t been flipping coins and getting a bunch of heads or tails. They’ve been the players hitting the baseballs, and we can’t strip them of responsibility. Think of the case of Yasiel Puig. At this writing, Puig’s got a 1.447 OPS. His slugging percentage is almost four digits. Obviously, Puig isn’t going to threaten to bat .500 for the duration of his career, but it wouldn’t be fair to say he’s been lucky. Perhaps he’s been lucky to have faced the Padres a bunch, but that line of his reflects his performance.

The thing about “luck” is, we know what people mean by it, usually. The purpose of language, to me, is to convey certain ideas, and as long as that’s accomplished, one needn’t be pedantic. One needn’t correct another who has the gall to say “decelerate.” But there is that important difference between “lucky” and “unsustainable,” and one makes it easier for statheads and sabermetric research to be dismissed. See, “luck” is off-putting, and inaccurate, and hardly necessary.

The big problem with using “luck” when talking about baseball is that it’s essentially a reassignment of agency. Baseball is a game between teams of individual players, and when you start tossing around “luck,” you start removing credit and blame from the equations, even though it’s the players who’re playing. It’s the Cardinals’ hitters who’ve hit the ball harder and found more holes with runners in scoring position. They did that — they swung at particular times, with particular swings, against particular pitches. They batted those baseballs, and they generated those results. They’ve been trying to get hits, and they’ve been largely successful.

The fact of the matter is that we don’t know how responsibility should be divvied up. This is a big issue when it comes to, say, BABIP. But when you start using a term like “luck,” you basically strip away all responsibility, and that’s dangerous. And, probably, wrong. Jhonny Peralta has a .417 BABIP, and that won’t continue. Adam Dunn has a .175 BABIP, and that won’t continue. But to simply say that Peralta’s been lucky or that Dunn’s been unlucky is to ignore a presumably real and significant performance difference. I’m guessing Peralta’s hit the ball harder. I’m guessing Dunn’s hit the ball worse, and perhaps more predictably.

When people talk about luck with regard to baseball statistics, they mean relative to some ordinary baseline. A player or team should have X, but instead they have Y. We can, again, use the BABIP example, where people are accustomed to seeing BABIPs around .290-.300. Someone’s lucky or unlucky if they’re a good deal away from that. Somebody’s normal if he has a BABIP of, I don’t know, .295. But what is the agency breakdown, then? Is the player considered responsible for the three hits out of ten balls in play? Could a player then not be responsible for two or four hits out of ten balls in play? When the Cardinals’ scoring-position splits regress, people will say the Cardinals hit like the Cardinals. Is it not possible for them to be responsible for temporary over-performance?

And that’s what this is really about — temporary over- or under-performance. People aren’t always performing right on their true talents, and variation can’t just be written off to luck. That misses too much of the truth, and while one can’t deny the presence and influence of luck, one should be careful not to exaggerate the impact. It’s going to make people less willing to listen, and it’s often going to be wrong, too. There’s almost never a reason to say “luck” when you can use some variety of “sustainable,” and though maybe that’s more of a mouthful, we’re already pronouncing names like “Adeiny Hechavarria” so it’s not like we’re worried about syllable economy.

One of the easiest ways to dismiss statheads is by saying they don’t watch the games, or they don’t enjoy them, or they don’t understand them. One of the easiest examples is the overuse of the word “luck,” as if baseball is nothing but dice in a cup. Mind your words, all of the time. Mind your words, in this case in particular.




Print This Post



Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.

71 Responses to “They’re Not Getting Lucky”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. CONSTERNATIONNATION says:

    Fuck the Cardinals, though.

    Vote -1 Vote +1

  2. Steve says:

    Thank you :)

    Vote -1 Vote +1

  3. JuanPierreDoesSteroids says:

    The purpose of language, to me, is to convey certain ideas, and as long as that’s accomplished, one needn’t be pedantic.

    WHich is why I get annoyed by the “irony police”. Or for that matter, anyone who says that a word is misused, despite there being mutual understanding of what is meant to be said.

    Vote -1 Vote +1

    • KDL says:

      I don’t correct folks who use the I-word incorrectly, because I get what they mean. But I think it’s more than fair to judge them for misusing a word. Especially since they are almost certainly using that word to simultaneously convey something about their cleverness or intelligence.

      I’ve never understood the point of using a word incorrectly (irony), when another word that already exists captures the idea correctly (coincidently -most of the time). When there is no word for a concept…go to town playing with language to find something that works. For example, I like the use of ‘irony’ regarding things like loving William Hung or William Shatner or will.i.am, because there wasn’t really a term for that type of interaction with pop culture previously. But almost every misuse of irony, and certainly the most popular, has a proper word.

      And the funniest thing of all, to me, is that it’s the “irony police” who get viewed as snobbish. Meanwhile those using a ‘big’ word pretentiously (why else wouldn’t you use any number of other words or phrase to get your idea across?) are viewed as the ‘down to earth’ crowd. At the very least both the irony police and the irony misusers are pretentious sides of the same coin.

      Vote -1 Vote +1

  4. Damaso's Burnt Shirt says:

    All I know is that AA rolled the dice and the Jays got stuck in the Britta of Timelines. Damn you Abed, why didn’t you stop him?

    +22 Vote -1 Vote +1

  5. _David_ says:

    Also, in addition to runners-on situations selecting for worse pitchers, each instance of success brings up another hitter in that beneficial situation, upping the portion of those plate appearances.

    Vote -1 Vote +1

  6. atoms says:

    Jeff, I think you’re turning this way too much into a black-and-white thing than it is. You can say the Cardinals have been a bit lucky (or fortunate, if you want a better word), without saying that their results have been ENTIRELY due to luck. As the saying goes (supposedly from Branch Rickey no less), luck is the residue of design.

    So to take your Jhonny Peralta/Adam Dunn example, it’s entirely conceivable and even probable that Peralta has hit the ball harder than Dunn, and therefore deserves a higher BABIP. But at the same time it’s also conceivable that a batter hitting the ball like Peralta has done should generally expect maybe a .360 BABIP and not a .417 BABIP, and a batter hitting the ball like Dunn should generally expect a .250 BABIP and not a .175 BABIP, so it’s still fair to say that one has probably been more fortunate than the other, especially as we know that these statistics are subject to unpredictable bounces of the ball, the fielders are not positioned in the same place all the time or gifted with the same abilities, and that a couple of millimeters of difference in the angle of contact can send the ball many feet in a different direction.

    +7 Vote -1 Vote +1

    • atoms says:

      I guess what I’m saying is that the word “luck” is a useful term for conveying the degree to which more-or-less random variation influences the results.

      Vote -1 Vote +1

    • Absolutely, I think the best = good + luck, and the worst = bad + luck. It’s probably not a surprise that I see gray areas in everything. But when people say a guy has been lucky, even if they mean he’s been partially lucky, the message conveyed is that it isn’t just partial. It’s a stronger word than it might seem like.

      +7 Vote -1 Vote +1

      • Jason B says:

        But its difficult to express the concept of “partially lucky, but skill, or lack thereof, also plays some real, but indeterminate, role” succinctly. I would hope that when we use, or read, the term “luck” that we’re using it as shorthand for that nebulous concept. We shouldn’t use it as “100% lucky, 0% skill involved” or read it as such. Maybe just read it as short for “luckier than most” or “luckier than is likely to be sustained”, like in the case of Peralta’s BABIP.

        Appreciate the article, and the willingness to accept and explore the gray. (Area. Not the wolf movie. Well…that too. Half and half.)

        Vote -1 Vote +1

      • LK says:

        “But when people say a guy has been lucky, even if they mean he’s been partially lucky, the message conveyed is that it isn’t just partial.”

        I find this statement interesting. Are you saying that the message, as understood by the majority of non-stats-oriented fans, is that it isn’t just partial? If so, then I would mostly agree, and I’d say that when you’re writing for such an audience it’s important to include the caveat that “lucky” and “good” aren’t mutually exclusive.

        However, if you’re saying that there’s something inherent about saying a guy has been “lucky” that conveys the luck isn’t just partial, I have to disagree. There’s nothing in the message that says that, just the audience’s interpretation of the message. And the Fangraphs audience, in particular, I don’t think really needs to be reminded of such things in most instances.

        Puig has been lucky, and I don’t really think that’s up for debate. If we could replay the past week of baseball games an infinite number of times, he’d perform worse than he has in the vast majority of instances. He just happens to have been awesome as well as lucky.

        Vote -1 Vote +1

        • Anon21 says:

          Yes, exactly. And when you’re writing an actual article, you should try to specify how lucky you think your subject has been; in the case of the Cardinals, you would not project their wRC+ with RISP to regress all the way back to league average, but maybe just to their own season average.

          But I completely reject Jeff’s interpretation of “lucky” as an absolute statement. That’s idiosyncratic and wrong, and if there’s a significant group of fans or commentators out there who would incorrectly interpret it in absolute terms, the solution is to educate them, not swear off the very useful concept of luck.

          Vote -1 Vote +1

        • I think luck comes with certain unavoidable connotations. I think there are better ways of conveying a similar or even identical idea. I think it’s easy to tune out a group of people who talk about luck in baseball, and I think this has been demonstrated, and I think this can be at least partially addressed by staying mostly away from an unnecessary word. Luck can work, but it can be misinterpreted, and we can make things easier and more accessible without, I don’t know, pandering.

          Vote -1 Vote +1

        • TCQ says:

          I think this probably could get parsed further. Taking Puig: I don’t think it’s lucky that he’s hit four homeruns, because I can look at HitTracker and see that they were all pretty good (although one was a Just Enough). So, unsustainable-not-lucky. But he’s also gotten two infield hits (plus a bunt), which, on a rate basis (noting that a rate basis is ridiculous for two hits, but whatever) would lead the league. That’s probably just lucky, even granting that he’s fast. And a one or two hit difference is noticeable over 29 PA.

          With that in mind, I’m going to say that I think this article has a really good point, but poor examples: hit sequencing (the Cardinals example) is basically luck, I think. BABIP is probably a mixture: luck in random variation of dinks; commendable unsustainability in hitting the ball abnormally hard.

          Puig’s case with homeruns, or a pitcher (say, Kris Medlen in the second half of ’12) with ridiculous peripherals…those are more to the point, as I read it, in terms of a player excelling at the things he controls, and deserving credit for it, even if it doesn’t budge ZIPS ROS that much.

          Vote -1 Vote +1

        • LK says:

          I have a different take on the situation. I think that the concept of luck (not the word but the concept) has been rejected by many so-called “old-school” fans not because people like Fangraphs writers use the word “luck” specifically, but because they don’t like the idea. I think if people like Fangraphs writers stop using the word luck, and start using “randomness,” or whatever else you’d want to call it, then those people will by and large start disparaging the concept of randomness.

          There are going to be a lot of Giants fans out there who think that if you replayed the 2012 season a million times, the Giants would win the 2012 World Series a million times (I’m not trying to pick on Giants fans at all here, it’s just the most convenient example). You can use whatever word you like, if you try to say that’s not the case, lots of people are going to disagree now matter how you phrase things.

          Vote -1 Vote +1

        • Bab says:

          I wouldn’t say Puig has been spectacularly lucky. Even bad Padres pitchers can strike you out etc.

          IMO, to really get down to brass tacks with understanding lucky players, you’ll have to get almost absurdly granular with BABIP data. Even after that, what point has been proven? Would a ‘lucky’ label increase or decrease the value of a player?

          The Cardinals may just have team dynamics in which the lineup rides hot simultaneously. Perhaps they have players with shallower performance cycle fluctuations. Perhaps on field management is especially astute with the lineup . . .

          In any case, luck as an analytical tool would likely have to be almost entirely exogenous.

          Vote -1 Vote +1

      • Bab says:

        Wouldn’t worst = bad + no luck?

        Vote -1 Vote +1

      • Baltar says:

        I respectfully submit, Mr. Sullivan, that you are better at writing about baseball than writing about semantics.

        Vote -1 Vote +1

  7. Requiem says:

    Nice article, Jeff.

    I never understood why “luck” was used as the buzzword for the exact reason you state. There’s a lack of agency implied in the word.

    I prefer the word “randomness,” because IMO, there is no implication of agency or non-agency in the word, but that just might be me.

    Vote -1 Vote +1

    • PackBob says:

      I like the term variability, and with it comes results that we consider for our purpose to be good, or lucky, and bad, or unlucky. Part of the problem, to my mind, is that there is the tendency to think that analysis of the whole describes the individual. It really describes a range of likelihood for an individual.

      Vote -1 Vote +1

      • Baltar says:

        This is a good comment, and I voted you a +. Words like “chance” and “random” are much better than “lucky,” because “lucky” implies a characteristic of an individual.
        However, it is awkward to use those better words, so I say “lucky” and, if necessary, explain further.

        Vote -1 Vote +1

  8. Drew says:

    It’s flukey that a bunch of good hitters have, so far this season, on average, only hit well in certain situations.

    It’s lucky for Cards fans that those certain situations involved runners on base.

    +7 Vote -1 Vote +1

  9. Nick O says:

    I get your point, but I think the reason that people refer to these events as “luck” is because those sort of outlier events happen exactly as often as they would in a baseball simulation game.

    Vote -1 Vote +1

  10. Aaron Franklin says:

    When someone says results are “unsustainable,” this is answering a fundamentally different question than the one luck can answer. When the question is asked “Why are these results unsustainable?” we must answer that it is because the difference between the unsustainable results and sustainable results is that unsustainable results rely factors which are not within the players’ control. You say:

    “It’s the Cardinals’ hitters who’ve hit the ball harder and found more holes with runners in scoring position. They did that — they swung at particular times, with particular swings, against particular pitches. They batted those baseballs, and they generated those results.”

    But, I think this may be an over-attribution of agency. It is a matter of our semantics that we assign subjects to words like “hit,” and “found a hole,” but the syntactic requirements of conveying these verbs do not necessarily reflect the agency at work. So while you can say that the Cardinals hitters have done things which are necessary to there being a certain range of results, I think those who say luck is involved might take issue with your claim that they “generated” these results. Perhaps they are taking actions in such a way so as to produce these results at a greater probability than other players, but the difference between the sustainable results given their choices as agents and the reality is perhaps what people have in mind when they say they’re “lucky.” Luck is the answer to the question of why this result is unsustainable. If this is the case, then I think it should have a place in the way we talk about baseball, despite how it may irk some people.

    +8 Vote -1 Vote +1

    • Jason B says:

      Well said. It can be equally wrong, and equally irksome, to overattribute agency as to underattribute it. Although I think for a fan of a particular team (like a zealous cards fan responding to claims that they’ve been lucky with hit sequencing or with RISP) underattribution of agency is more bothersome; they like to think their guys have some fundamentally special and repeatable skill, not that they got a little lucky.

      Vote -1 Vote +1

  11. Martin says:

    Correct me if I’m wrong, but the assumption underlying the entire discussion is that a team shouldn’t have an OBP vs. non-OBP split, right? I see the logic behind the assumption, but logic doesn’t necessarily mean it is true. I guess another way to pose the question is: “are players that perform better with RISP randomly distributed across the population?” After all, for every Hank Aaron (.288 no-RISP, .306 RISP) there is a Rickey Henderson (.282 no-RISP, .257 RISP) and a Pete Rose (.301 and .302). But I guess I just want to believe that Rennie Stennett was actually a clutch hitter (.263 no-RISP, .302 RISP)…

    Vote -1 Vote +1

  12. Joe says:

    I think the obvious answer here is that the Cards’ runners are very good at relaying signals.

    Vote -1 Vote +1

  13. Joe L says:

    Jeff, I completely agree that luck is certainly a bad word to use in baseball. When hitting a baseball, there are so many additional/unknown causes for the results on the field — even when we eliminate fielding from the picture.

    For starters, there is in fact no such thing as luck. It’s not a real “thing.” Every event has an actual cause — many, independent and dependent actual causes — and none of them are are “luck.” They are physical, chemical, and perhaps even emotional forces. They are real, tangible and intangible. Some you control, some you do not control but you can learn to navigate succesfully, and some are entirely beyond your control or even your knowledge.

    Luck is just a shorthand for saying that (1) forces that are entirely beyond your control or knowledge and your ability to navigate played a greater role — or, perhaps, the entire role — in your success, and these forces were unusually favorable to you compared to how favorable they are to others who are similarly situated.

    But of course, the problem is that no others are ever in exactly the same situation. And even if they were, we do not have enough data to know whether they are similarly situated. So we’re just making lazy generalizations.

    Let’s take a coin flip. It’s simialr to hitting because there are no fielders, but even simpler because there is no pitcher either. Fewer factors outside of your control. And yet, if you flipped a coin 100 times, winning $1 for heads and losing $1 for tails, we would call you lucky if the coin flip came out heads 75 times. But that would be presumptious, because would not know why you received those 75 heads unless we watched you extremely carefully, monitored you, interviewed you beforehand, etc. For each coin you flipped, there are many actual causes that led to your coin landing heads. Unless we make an incredibly careful study, we do not know if you somehow, for part of the exercise developed a repeatable flipping motion that increased the likelihood that the coin would land heads. We do not know what you knew about the temperature, humidity, or wind — or whether you knew about. We have no idea how much of what you did was attributable to your skill of coin-flipping, your current physical and/or mental state, or the conditions. We have no idea whether you are totally obvlivious to everything.

    However, with a coin flip, since there are probably few people who have devoted a lifetime to learning how to flip (and are paid millions of dollars to flip) a coin to result in heads, we can feel pretty safe in saying you were lucky when you flipped 75 heads.

    Hitting a baseball is like flipping a coin but far, far more complicated. Potential causes for performance multiply. It’s a highly physical tasks, so physical condition matters more than when flpping a coin (presumably). Plus the hitter has been training, so he may have developed some techniques that may or may not be working that particular day. We have all the normal range of mental and emotional factors that affect human performance, but we also have the interaction with coaches, agents, fans, etc. that can change our usual mental or physical state. Perhaps the hitter is working on a new swing and has it “just right” today, but not tomorrow? We don’t have any of that data. Of course, there also is a pitcher and a ballpark — and even if we controlled for that, and even if we measured the speed of his pitches, the location of his pitches — and even if we had some stat geek watch for the type of pitch and how much the curveball broke versus that pitcher’s typical curveball — the pitcher himself might have various physical or mental conditions on that daty. Not to mention that there so many more weather factors (perhaps less so in a Dome) than we tend to measure. Not just wind speed and direction, temperature and humidity — how about where the shadows are, not just all of the time but at that particular time of the year.

    So, sure, is easy to say that if someone has a BABIP of .250, especially when they typically fall within the normal range, that they have been “unlucky” — and that the guy hitting .350 has been “lucky”. But even if we have enough data to say that that batter is hitting more long line drives than he usually hits on pitches of a certain speed in a certain area, can we be sure that his performance is lucky — i.e., attributable to factors beyond his control or that he cannot navigate? No way. Maybe the major cause is that new swing. Or leg kick. Or that he finally made peace, even in the short term, with his sister-in-law with whom he has been feuding. Or he’s facing a hurt Justin Verlander, who is throwing just as hard but his control is not exacly the same — but only enough different that his catcher (not Foxtrack) can tell. Perhaps he has a new-found concentration based on yoga and meditation. Or perhaps he is slightly injured but we will not find about it for weeks — perhaps not ever.

    We can regress and regress and regress and try to control for these factors, but we simply do not have enough data to say, at any point, that a hitter has been lucky. Because we do not know WHY he has been succesful. Perhaps, Fernando Rodney finally put it all together last year because of where he moved on the rubber — but for all we know, it is something comforting that Joe Madden whispered into his ear after each successful save opportunity. Perhaps Chris Davis just saw enough curveballs that he learned how to either hit them or lay off — or perhaps he is wearing different underwear that makes him more comfortable and feel better about himself. Maybe it is that leg kick that explains Jose Bautista, after all — or maybe he is eating more carrots than before. We do not know, because we hear about a few things from Buster Olney and the journalists, but in reality we hear about so little.

    So it’s anothert thing entirely to call something unsustainable. To call something unsustainable means that, even if there is some skill or change that is the cause of a short term performance spike, long term trends suggest that that performance spike will disappear — not because luck will change, but because those causes will disappear. Josh Hamilton can switch churches, and his newfound inner peace will allow him to take a pitch every once in a while. But we know Josh, and the data tell us that he will probablt revert to his free-swinging ways.

    Sustainiability is not an insult. It’s a prediction about future perfomance. “Great job! However, even if we attribute your fantastic recent performance largely to your hard work, we expect based on past results that it will not continue.” Luck, by contrast, is an insult: “Sorry, your hard work is not the real cause of your great performance.”

    That’s why, Jeff, I agree with you so heartily.

    Vote -1 Vote +1

    • Joe L says:

      Ugh, I should have put this one through spell and grammar check before posting…sorry!

      Vote -1 Vote +1

    • Aaron Franklin says:

      Your intuitions about what people mean when they say someone is lucky seem off to me; it’s not that “luck” is some metaphysical force which acts on every under-determined event in the universe, it seems more of a way of describing the randomness inherent to unsustainable results.

      Even if we had the epistemic certainty with regards to the causal profile of every single event in the universe, we might still say someone is “lucky” because the causal factors which produced the result he/she was looking for came from something other than their own agency.

      Vote -1 Vote +1

      • Joe L says:

        Actually, plenty of folks do use the word “luck” to mean a metaphysical force.

        For example: http://www.merriam-webster.com/dictionary/luck

        I think your use of the word “a way of describing the randomness inherent to unsustainable results” is inconsistent with how most people use the word. Lots of people think of themselves as “lucky” — and they don’t think of it a unsustainable. Sure, most of these people are idiots, but it says something about what the word means in common parlance.

        Luck is a bad term in this context because it tends to suggest that some cause other than the hitter’s ability was the dominant/prime/main reason for the outcome. This is laziness because of course we have no idea what caused any particular outcome based on statistics.

        In fact, “unsustainable” is in fact a second-order conclusion while luck is a first-order one. Luck simply implies that something good happened that was outside of the control of the lucky person. Unsustainable means that it is probable that this luck will continue.

        “Unsustainable” is not that much harder to say than “luck.” And it’s more accurate.

        Vote -1 Vote +1

    • Lucky says:

      Agree wholeheartedly with this comment. Except I’d rather use the word luck. The word was invented for a reason. It’s used to explain outcomes that aren’t linked to skill. The Cardinals overall batting ‘skill’ is NOT luck. However, unless you tell me that the Cardinals bat differently with the bases empty vs with men on base, the distribution of that batting skill is luck.

      There are SO many articles on this site which says being ‘clutch’ is not a real skill. Well, isn’t having RC+ of 140 in effect being clutch. I’d rather think of everything as skill or as luck. And if being ‘clutch’ is not a skill, I’ll take the performance of the Cardinals’ hitting with men on base SO FAR THIS SEASON as luck.

      Vote -1 Vote +1

    • LK says:

      It seems to me that if you can’t call getting 75 heads in 100 coin flips “lucky,” you’ve defined the word to be meaningless.

      Vote -1 Vote +1

      • Joe L says:

        Then you did not understand my comment. If someone flips 75 heads out of 100, we tend to assume it was lucky. But we don’t know simply based on the stats. Perhaps, with a coin, it’s a good assumption because there is simply no way to learn to flip a coin so it tends to lie on one side or the other. I have no idea the physics of that one.

        But with baseball, I’m pretty sure that people can work on their skills and become better hitters or not. So the assumption is less warranted, and can be wrong. So it is ridiculous for people to suggest that a high BABIP over time is based on luck. If the guy is actually hitting the ball harder, farther, whatever, maybe it’s skill.

        And if you want to say that by definition, something that’s a “skill” must be sustainable — then fine. But what do you call it when I figure out, for a short time, exactly how to do something right? The cause, for that short time, was my ability to figure it out. The actual cause. It just turned out that I don’t have the ability to keep it all together.

        So perhaps the problem is with the word skill, as well.

        If I figure out how to fold a complicated origami, and do it right, is that luck? Even if it is so complicated that I soon forget how to do it and can never get it again? Or was it, for a brief shining moment, the highlight of my skill? A skill so fleeting I could not retain in for more than one origami. I believe it is ridiculous to call this luck, but if you want to call it something other than skill I will buy that. I can see the usefulness of another distinction.

        Vote -1 Vote +1

    • NY Expat says:

      Occam’s Razor would like a word with you.

      Vote -1 Vote +1

      • Joe L says:

        You seem to be forgetting the middle (dependent) clause of the razor.

        The gist of the razor is that “simpler explanations are, other things being equal, generally better than more complex ones.” Meaning: The more simple explanation is better than the more complex one IF both explanations have EQUAL informative value. But if the more complex explanation does a superior job of explaining, the razor has no bearing.

        “Luck” is not a good explanation for anything. It’s not an actual cause of anything. Each time I dribble a grounder to the second baseman rather than hit a line drive into the outfield, there are many actual causes and luck is not one of them. Even aside from differences among pitchers, the list of actual explanations goes on and on: I did not swing the bat early enough, the sun got in my eyes, my swing is slowed by age, I am injured, I was distracted, I got something in my eye, my swing mechanics are bad today. These are actual causes.

        We do not have enough information to know the actual cause of our failure to hit the ball. Attributing bad at bats to luck is simply a lazy way of dealing with this information failure.

        Vote -1 Vote +1

    • CONSTERNATIONNATION says:

      “For starters, there is in fact no such thing as luck. It’s not a real “thing.” Every event has an actual cause”

      Wow. I’ll have what she’s having. And by that, I mean the drugs.

      In other news, I remain overjoyed that I got my first comment in right at the top. Stick to GIF compiling, Sullivan. Not really, though, (but kinda really).

      Vote -1 Vote +1

  14. noseeum says:

    Agreed 100% with this post. So why use FIP for WAR again?

    Vote -1 Vote +1

    • Kevin says:

      The motivation for removing batted ball data from pitcher’s WAR calculations is NOT due to a lack of predictability, but to an inherent imprecision in assigning credit for known results.

      When a batter makes an out or gets a hit, we can more or less definitively say that he himself caused the result. In other words, none of his teammates sitting on the bench or standing on the basepaths significantly influenced the play.

      Pitchers, on the other hand, are dependent on the 8 guys around them for every ball not grounded back up the middle. On these plays, we can’t properly assess how much credit to give the pitcher or how much credit to give the fielders.

      In summation, pitcher WAR uses FIP, not to eliminate “luck”, but to ensure the integrity of the reported statistic.

      Vote -1 Vote +1

      • Urban Shocker says:

        When a batter makes an out or gets a hit, we can more or less definitively say that he himself caused the result…Pitchers, on the other hand, are dependent on the 8 guys around them for every ball not grounded back up the middle.

        Your comment is exactly why this post was written-there is practically a daily post about hitter/pitcher BABIP, and whether this is, or is not, ‘luck’.

        Vote -1 Vote +1

  15. Hank G. says:

    I think you’re trying to create a distinction that doesn’t exist.

    Vote -1 Vote +1

  16. Bryan says:

    I think the term luck encourages lazy thinking. I would ban the word. Instead of saying a baseball event is lucky, explain in concrete terms why the situation is random, and where one might expect regression.

    Want evidence of lazy thinking? Go back to pages on this site from October. Apparently the Giants won the world series because they were “lucky” at rolling dice. Luck is an imprecise idea that masks sloppy analysis.

    Vote -1 Vote +1

    • Joe L says:

      Thanks Bryan. You achieved what I was trying to say (and what Jeff said) in many fewer words. Occam would be proud.

      Vote -1 Vote +1

  17. Gyre says:

    “Normal” is what happened yesterday.
    Ty Cobb ‘normal’ no longer holds sway. It and many other ‘normals’ have been kicked down the road to make room for the latest new thing. The game changes and moves on, the Cards could be seeing the shifts made due to runners and simply be able to put it where they ain’t better than ‘normal’.

    The amazing thing is that after 100 years, not that it’s running at all, but that it’s gathering steam. If that takes luck, I’m all for it.

    Vote -1 Vote +1

  18. Sean says:

    Luck is a very useful term. It’s the word that describes when you have an emotional investment in something, and that something is following a path that you can’t rationally explain. If you happen to LIKE the thing and it’s doing better than expected, it’s called “good luck”, and conversely bad luck.

    It’s not at all surprising that so many folks on a numerically-oriented site are so resistant to the idea of luck, since it requires the notion of their emotional investment in an outcome, and statistical analysis at its most dogmatic is an attempt to whitewash away the mysterious emotional elements of the game.

    You can pull out the 1984 playbook and ban all these words you don’t like, but it doesn’t mean that when the team you love to hate gets more good breaks than bad breaks, you won’t be thinking, “those lucky f*cks”.

    Vote -1 Vote +1

    • lukey says:

      I don’t see why we should believe that emotional investment is a prerequisite for viewing events as lucky or unlucky. I can recognize that a series of dice rolls are lucky even if I’m not emotionally invested one iota in the game of Candyland I’m currently playing. So your corollary, that “numerically-oriented site are so resistant to the idea of luck, since it requires the notion of their emotional investment” is not really persuasive.

      Vote -1 Vote +1

  19. hookstrapped says:

    As an Orioles fan, this critique of the use of the term “luck” is pretty gratifying. Last season, as the Orioles 1-run game record and extra-inning game record defied logic and their W-L record defied pythagorean expectation, we read a lot about how lucky they were.

    It’s been suggested here that “unsustainable” or “random” be used instead of luck when events diverge from expectation based on known predictors, but I think neither term gets to the root of the issue. For example, the Orioles did sustain their 1-run and extra inning records through the whole season. And it didn’t seem random for the very reason it was sustained.

    I think what’s going on in these situations is variance from the norm for reasons we don’t fully understand / metrics we don’t have. My snide summation is stathead dismissal of the Orioles last year as lucky is that luck is the term sabremetricians use for events they can’t explain. I think that remains an underlying reason for resorting to that word, and it is inherently lazy and regressive (in the sense of developing ideas, tools, and knowledge).

    Vote -1 Vote +1

    • Jason B says:

      “For example, the Orioles did sustain their 1-run and extra inning records through the whole season. And it didn’t seem random for the very reason it was sustained.”

      As I mentioned earlier, we as fans like to think our teams in particular did something to influence things and didn’t just happen to get lucky. No one likes to be told their kid isn’t special.

      Vote -1 Vote +1

    • Kevin says:

      “Statheads” can be guilty of lazy analysis, i.e. not assigning credit where credit is due, but this is no less greivous an error than your average fanboy projecting all-star production from a mediocre player on a hotstreak, i.e. assigning credit where it is NOT due.

      The Orioles success in close games last year should not be seen as merely luck, or as entirely skill, but rather as an acknowledgement that baseball is unpredictable, AS WELL AS an invitation to examine certain skills more thoroughly, such as Buck Showalter’s bullpen usage, or the tendencies of certain pitchers to excel in certain situations.

      Vote -1 Vote +1

    • CONSTERNATIONNATION says:

      Really, though, fuck the Orioles.

      Vote -1 Vote +1

  20. LTG says:

    Am I the only one confused by the ‘decelerate’ example? Is there an authority that claims it isn’t a word? I know of a couple of authorities that claim it is. Moreover, the word is in common use with a common meaning. So, even if it weren’t in a dictionary (it is), it would be a mistake to claim it isn’t a word.

    Vote -1 Vote +1

  21. Lucky says:

    Definition of luck (according to the internet):

    “Success or failure apparently brought by chance rather than through one’s own actions.”

    If the Cardinals do something differently with men on base to produce better results, I’ll believe it’s not luck. Otherwise the distributions of their outcomes are luck. You argue that the results are likely not sustainable. That is true. Because they’ve been lucky. It’s not that the overall results are lucky. It’s the distribution of the hits that is. You could easily flip performance in the two situations and argue that they’ve been unlucky. I’d buy that too.

    It’s all semantics and I’d rather call it luck. It’s easier than saying variance from expected performance or sustainability or whatever.

    Vote -1 Vote +1

  22. Nick says:

    I don’t think you needed to defend Buster Olney here as I’m sure his reasoning is nothing more than, “The Cardinals are super clutch!”

    Vote -1 Vote +1

    • Anon21 says:

      Fangraphs writers seem to be constantly seeking a rapprochement with dimbulb mainstream writers with big audiences. Fool’s errand, so far as I’m concerned.

      Vote -1 Vote +1

    • lukey says:

      I agree, though it’s a little off topic. But yes, the whole article could have been written, without any loss of persuasion or insight, had there been no mention of Buster Olney. But still, a really good post.

      Vote -1 Vote +1

  23. Mr Scout says:

    Stat geeks gave us “luck”, how ironic stat geeks now decry it.

    Vote -1 Vote +1

  24. Mr Scout says:

    Score this a Win for scouts.

    Vote -1 Vote +1

  25. brian says:

    Rob Neyer needs to read this piece, he’s one of the more egregious users of the term “luck”

    Vote -1 Vote +1

  26. Zigs says:

    Good article! Too many times we misuse the term luck when we look at advanced statistics, especially when someone is doing very poorly. A lot of the time hitters really are just hitting extremely poorly when their stats are low.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current day month ye@r *