hBABIP =/= Luck ← FREE KRAUT!

hBABIP =/= Luck 128

This is one of my pet peeves when it comes to using advanced statistics. Ever since BABIP went mainstream, I’ve been reading blurbs on whatever fantasy site I’ve been playing at the time about how I should target XY, because he has low BABIP or how I should sell WZ because he has a high one.

And there is no question that random distribution plays a bigger role in hitting BABIP than it does with K% or BB%, for example. Very logically so, because apart from the hitter’s own skill, the K&BB rates are additionally influenced only by the skills of pitcher, catcher and the umpire. BABIP, on the other hand, adds the skills of all defensive players into the equation, as well as stadium specifics and the random  distribution of the path and velocity of the batted ball.

But hBABIP is a skill, there is no question about it, and one that has a far better signal to noise ratio than pBABIP.

Very easy way to check that is to look at the hBABIP of NL pitchers, which is around .220 (I used last three years of data as a quick example). Position players with > 100 AB in the same time have a hBABIP of almost .300.

Now, the trigger for this post was the following post on OC, written by SamYam. He took over the role of Sabermetrist on the site and has written good stuff so far. He also seems genuinely nice, so consider this a #samyambait to come over and discuss some stats.

In it, he says:

First of all, Reddick’s been getting very unlucky with his batted balls. He has a BABIP (batting average on balls in play) of .251, which is very low. This means that he’s hitting a lot of balls right at fielders, which is pure luck and probably won’t stay that low for the whole year.

Now, this is obviously not quoted to pick on him, but rather because this is an often expressed sentiment, sometimes on this site, as well. As mentioned above, pitchers hit into a BABIP of .220, not because they are unlucky, but because they are not hitting the ball well.  Now, you might say that this is because pitchers don’t hit line drives, which are the hardest balls to turn into outs. And, partially, you’d be right, the group of pitchers above hits line drives 16.7% of the time, whereas the professional hitters do it 19.8% of the time.

But that alone makes hardly any difference. I used a first xBABIP formula I could find and it looks like this:

xBABIP =0.391597252* + (LD% x 0.287709436 ) + ((GB% – (GB% * IFH%) ) x -0.151969035 ) + ((FB% – (FB% x HR/FB%) – (FB% x IFFB%)) x -0.187532776) + ((IFFB% * FB%) x -0.834512464) + ((IFH% * GB%) x 0.4997192 )

*I always have to chuckle when I see 18th decimal points precision in this sort of rough estimators.

Is this a perfect formula? No, certainly not. But it gives some sort of level playing field for hitters and pitchers. I then tweaked it, so that it centers for my control group (hitters). After that, xBABIP of the NL pitchers is .288, just a tad under that for hitters (.297). They under-perform that by more than 60 points.

So, we can not look at the batted ball distribution alone and just claim that random distribution accounts for all (or even the majority) of the observed hBABIP’s discrepancy to league average.

Discussion encouraged.

128 thoughts on “hBABIP =/= Luck

  1. Bed Aug 1,2013 8:02 am

    I’ve never been a fan of saying someone is having a poor season due to bad luck especially when the season is now into August. I think Reddick is having a poor season at the plate because he’s not a very good hitter. He had a great run the first half of last season but since he’s been a hole in the lineup.

    I know batting average is a stat of less importance these days but hitting .214 sucks no matter how you look at it. Add in the fact that he’s got a career OBP of .298 and I think it’s fair to say he’s nothing special at the plate.

    I know you have to take into account how a guy plays defense and runs the bases when looking at overall value but having a corner outfielder who is hitting this poor is a drag on the team and I wish Billy could have added someone to replace him.

    I guess I should add I’ve seen about seven games total all year so I should probably just shut up and go back to posting about Sheena Easton.

    But seriously, folks....
    • elcroata Aug 1,2013 8:05 am || Up

      What’s Sheen Easton’s VORP?

      Because survival is insufficient
      • Bed Aug 1,2013 8:09 am || Up

        I think her 80’s VORP is somewhere bellow Madonna and Whitney Houston…it’s in the Gloria Estefan range.

        But seriously, folks....
        • PDXAthleticsfan Aug 1,2013 8:55 am || Up

          But way over Toni Basil’s. Although that may be small sample size.

          A soliloquy of fresh-sounding ideas which would probably be disastrous.
          • Future Ed Aug 1,2013 5:22 pm || Up

            that is not where toni basil gets her value though, she has it for choreography. I would take her over Estefan

            I have $5. No I don\'t.
            • PDXAthleticsfan Aug 2,2013 8:16 am || Up

              Her career is pretty remarkable reading her Wiki site. Also, she’s 70.

              A soliloquy of fresh-sounding ideas which would probably be disastrous.
            • DFA Aug 2,2013 11:37 am || Up

              I have no idea who she is.

              In play, run(s)! Talk dirty to me gamecast, talk dirty. - Nevermoor
              • PDXAthleticsfan Aug 2,2013 11:49 am || Up

                She’s known mostly for her 1981 No. 1 song “Mickey” and the video in which she dressed as a cheerleader (the song is mostly sung in a cheerleader-y fashion. See here.

                In truth, she’s primarily a dancer and choreographer for videos, concerts, films, etc., who has worked with the Talking Heads, David Bowie, Bette Midler, Tina Turner, and apparently, even M.C. Hammer.

                A soliloquy of fresh-sounding ideas which would probably be disastrous.
                • DFA Aug 2,2013 1:37 pm || Up

                  OH That song ok.

                  In play, run(s)! Talk dirty to me gamecast, talk dirty. - Nevermoor
        • 5Aces Aug 1,2013 8:59 am || Up

          You are wrong and that is an absurd statement. She worked with Prince, therefore she is great.

          Camelot sure fell apart, didn't it? -Steve McCatty
          • Bed Aug 1,2013 9:16 am || Up

            So she worked with Prince and Don Johnson…maybe she should be more highly rated by me.

            But seriously, folks....
    • 5Aces Aug 1,2013 9:08 am || Up

      I will say a season may be an “unlucky” one, but the key is to me is remembering that this doesn’t mean they will suddenly get incredibly “lucky”. The trap that we fall into is that “the luck evens out”. I guess if you looked at a large enough population over a long enough period of time maybe, but just because I hit a screaming line drive right at someone my first AB doesnt mean that I will get a bleeder past the infield next time up.

      And the other key is we view things from beginning to end and often in the way we want to see them. We want to believe Reddick (or any of our guys-hell CY25) as hitting in bad luck because 1)we want them to do good and 2)they performed better before. But as you said Bed, it is just as possible that first half of 2012 was the “good luck” for Reddick. Now in his specific case I would put qualifiers because it was clear he wasnt right between the trips to the DL and that influences his currnt #s. But its just always a bad idea to look at the best 2 months someone ever had and say “that is how they typically play”.

      Camelot sure fell apart, didn't it? -Steve McCatty
      • Dial C for Concupiscence Aug 1,2013 3:34 pm || Up

        I think it is fair to say that Reddick is capable of playing to the level he showed in those few months, and exploring why he isn’t playing to that level now though. You don’t have to assume those good months are his true talent level to wonder why he put up those numbers then and isn’t now. If you look at those numbers (with 1H ’12 as a rough estimate for his “good period”) and compare it to 2013, there are some standouts.

                GB/FB   LD%     GB%	FB%      IFFB%   HR/FB   IFH%
        1H '12: 0.59   22.6 %  28.9 %   48.5 %   12.9 %  17.2 %  8.7 %
        
        2013:   0.91   21.9 %  37.1 %   41.0 %   12.8 %   5.8 %  5.1 %	
        
        
                 BB%     K%     BB/K     AVG     OBP     SLG     OPS     ISO    BABIP
        1H '12: 10.6 %  21.6 %  0.49    .268    .348    .532    .880    .264    .292	
        
        2013:   10.1 %  18.5 %  0.55    .214    .293    .344    .636    .130    .248
        

        BABIP is one of the things that jumps out, but it’s not the only thing. Reddick is hitting way more grounders this year and his power has disappeared. So BABIP looks like it could be playing at least small role (such as the drop in IFH%), but the big drop in power is much more concerning. Who knows if that’s because of injury, luck, swing mechanics or whatnot.

        His walks and K’s have actually gotten back to 1H ’12 levels after falling apart in 2H ’12, so that’s not the explanation. In fact, he’s improved his year-to-year BB% and K% every year up to this point (depending on what happens in these next few months). Those tend to be positive signs, which makes his offensive collapse more confusing.

        1H ’12 is not (likely) his “typical level,” but it’s certainly a level he’s capable of playing to. He might need “good luck” to reach it, but there’s no denying that he has reached it. It may very well be the case that luck has played a bigger role in him putting up good numbers than it has in his failures, but like we shouldn’t chalk up his struggles solely to luck, we shouldn’t chalk up his successful period solely to luck either.

  2. nevermoor Aug 1,2013 9:26 am

    I’m not sure I buy the pitchers point as disproving that BABIP is largely luck. I just think that pitchers (who make the major leagues based ~0% upon their ability to hit) are a separate category.

    Isn’t the better way to determine whether BABIP is a skill to look for period-to-period correlations?

    "There's never enough time to do all the nothing you want"
    • elcroata Aug 1,2013 9:33 am || Up

      Looking at extremes makes it easier to understand the concept. And the concept is that the shitty hitters underperform their expected BABIP.

      Because survival is insufficient
      • nevermoor Aug 1,2013 9:44 am || Up

        But pitchers aren’t just shitty hitters. They’re non-hitters.

        Even if position players pitched more frequently, I don’t think you’d be able to learn anything about professional pitchers from how those position players perform.

        "There's never enough time to do all the nothing you want"
        • elcroata Aug 1,2013 9:54 am || Up

          They’re non hitters in the sense of their job-description, perhaps. But as soon as they step in the box, they are hitters, just like you and me would be, no matter how overmatched.

          The fact they are not good at hitting, and by a big margin, enables us to have such a strong signal that it overrides the noise.

          Because survival is insufficient
          • nevermoor Aug 1,2013 10:10 am || Up

            Sure. But doesn’t this go back to selecting appropriate priors?

            "There's never enough time to do all the nothing you want"
            • elcroata Aug 1,2013 10:14 am || Up

              I am not using them as a prior, but rather as an indicator that hBABIP is a skill. You are absolutely right that I cannot use their performance to quantify anything wrt real hitters, but more as a confirmation that there is a signal in there. Deciding how much signal and how much noise is there in our observation is best done by other techniques as you suggest, like p2p correlations…

              Because survival is insufficient
              • nevermoor Aug 1,2013 11:42 am || Up

                Oh, ok. I thought you were saying more than you are.

                "There's never enough time to do all the nothing you want"
    • elcroata Aug 1,2013 9:40 am || Up

      But speaking of p2p correlations, of course BABIP has a lot of noise, but the r=.7 point (820 BIP) is not that far away from LD% (600 BIP), yet the former is always presented as luck and the latter as skill. Considering all the additional noise (fielders, park effects) it actually speaks that BABIP is probably more of a skill than LD%…

      Because survival is insufficient
  3. SamYam Aug 1,2013 10:13 am

    Consider the bait taken!

    I agree that the BABIP = luck argument is waaay overstated. It’s much more complicated than that. That being said, I stand by my argument that Reddick’s low BABIP specifically is mostly due to noise based on a few reasons.

    1. Reddick has a a LD% of 25% or 21% depending on whether you trust bref or FG. The league average is about 20%, which means that Reddick is hitting more than an average number of line drives. Typically, line drives fall for hits at a much higher rate than other batted ball types. This isn’t Reddick making weak contact. To be fair, he does have a much higher GB% than last year by a huge margin, but it’s not coming at the cost of line drives; it’s coming at the cost of fly balls. This explains the power outage, but not the low batting average.

    2. Using the xBABIP calculator on Baseball Prospectus, Reddick’s got an xBABIP of .311. I recognize that noise is part of the game, but based on his batted ball profiles, you would expect a much, much higher BABIP than he’s had.

    3. His BB% is a career-high, and his O-swing% is down quite a bit, which is indicative of better discipline in his at-bats. This doesn’t necessarily make him a better average hitter, but swinging at fewer balls out of the zone certainly helps.

    Note- this is my first post on FK, and am excited to join the discussion here!

    • nevermoor Aug 1,2013 3:29 pm || Up

      Testing to see whether a reply allows me to approve.

      "There's never enough time to do all the nothing you want"
      • nevermoor Aug 1,2013 3:32 pm || Up

        It does. Welcome SamYam.

        All your comments should post immediately going forward. And I’m working to get things moving smoothly on the new server, so this ultra-slow BS isn’t permanent.

        "There's never enough time to do all the nothing you want"
    • dmoas Aug 1,2013 3:31 pm || Up

      Wooo!! New blood!

    • ptbnl Aug 1,2013 3:40 pm || Up

      Welcome!

      If this is His will, He's a son of a bitch.
      • SamYam Aug 1,2013 6:15 pm || Up

        Thanks guys! I see so many names that I’ve missed over at OC (??? that’s what elcroata referred to it as up top, so I’m rolling with it) from when I was a lurker there. Nice to see that all of you guys are still around and talking A’s. Will definitely be posting here more.

        • nevermoor Aug 1,2013 6:26 pm || Up

          Old Country. You’ll also see it referred to as **.

          "There's never enough time to do all the nothing you want"
    • Dial C for Concupiscence Aug 1,2013 3:44 pm || Up

      Welcome!

      1. It doesn’t explain the power outage though. Less total fly balls equaling less total homers makes sense, but his HR/FB% has completely tanked. Not only has he traded fly balls for grounders; the fly balls he is hitting haven’t been nearly as powerful. Fly balls have the worst BABIP of the big three batted ball types, so trading fly balls for grounders wouldn’t logically lead to a lower BABIP.

      • SamYam Aug 1,2013 5:52 pm || Up

        Agreed about the HR/FB% point. It’s less than 5% currently. But even if it was normal (say, 10%), he’d still have fewer homers than in the past due to a decreased total number of fly balls with which to work. Also agree about the BABIP point- my point was more that the increased grounders haven’t sapped the line drives.

        • Dial C for Concupiscence Aug 2,2013 9:43 am || Up

          Yes, his LD% has held steady, which is not a bad thing. My point is pretty much what ec is saying in that Reddick’s GB/FB and HR/FB ratios are more indicative of something than his LD% and BABIP. The root causes behind those numbers is beyond me though.

      • 5Aces Aug 1,2013 11:18 pm || Up

        How horrible would it be to cherry pick out some if his numbers? The big problem I have whenever I consider his numbers for this year is that period between DL trips. It was clear he wasnt right, but he kept going, driving everything in the wrong direction. And I still wonder if that is the problem with CY25. It would not shock me to see some blurb this December about a “lingering issue” that he “played through for the team” smfh…

        Camelot sure fell apart, didn't it? -Steve McCatty
        • Dial C for Concupiscence Aug 2,2013 9:52 am || Up

          Wrist injuries do sap power, so that is a very viable possible explanation for Reddick’s plummeting HR/FB ratio. CY25 has definitely not performed the same since his shoulder injury last year. An offseason of rest didn’t appear to help him unfortunately. Hopefully the same isn’t true of Reddick.

          • DFA Aug 2,2013 11:46 am || Up

            I think being wary of shoulder injuries is really important. Look what happened to Barton.

            With regard to Reddick, the man is a dumbass and his tendency to play hard, run into walls, and dive makes it highly likely that he will continue to injure himself. Because of him being a dumbass I fully expect him to play rather than getting treatment for a myriad of maladies and for it to regularly sap his production. I think whishcasting periods of hurt induced suckage away to predict a higher true talent is a faulty notion because he is likely to repeat more hurt induced suckage in the future.

            In play, run(s)! Talk dirty to me gamecast, talk dirty. - Nevermoor
            • Kay Aug 2,2013 3:22 pm || Up

              Good points.

              \"Weren\'t you already aware the Kay is already writing everyone\'s story? We\'re all just characters who believe we are real. Things make more sense now, don\'t they. Be honest.\"- DMOAS
              • 5Aces Aug 2,2013 3:27 pm || Up

                Agreed. When I commented I was thinking of forecasting at evil empire. Often we will have a day where “shit happens”. It isnt repeatable and it offers no help in a future forecast, so in our models we will normalize or exclude. But when we think about players-and Reddick in particular, the points about him doing the agressive thing in the field if it makes sense or not makes it much more likely to happen again, and he also seems to be one that would always try to tough it out to his and our detrement.

                Camelot sure fell apart, didn't it? -Steve McCatty
    • Kay Aug 1,2013 5:47 pm || Up

      Welcome to FK. Did we meet at the Blog Press Conference in January?

      \"Weren\'t you already aware the Kay is already writing everyone\'s story? We\'re all just characters who believe we are real. Things make more sense now, don\'t they. Be honest.\"- DMOAS
      • SamYam Aug 1,2013 5:59 pm || Up

        I only started writing for the blog in May, so I doubt it! Maybe next year?

    • elcroata Aug 1,2013 9:33 pm || Up

      Back off, everybody. I brought him, and the 10% of his harvested keystrokes are MINE!

      Because survival is insufficient
      • nevermoor Aug 1,2013 10:00 pm || Up

        I’ll tell you what, if we haven’t scared him away I’ll double your stake.

        "There's never enough time to do all the nothing you want"
        • elcroata Aug 1,2013 10:11 pm || Up

          I’ll need it. The site is so slow now, that I’ll be late for both my elctroshocks and work.

          Because survival is insufficient
    • elcroata Aug 1,2013 10:02 pm || Up

      Hey Sam,
      nice that you found your way over.

      I think you are way overvaluing the importance of LD%.

      First, this is the stat that has a lots of noise coming in (it is a stringer’s personal decision to call something a line drive or a fly ball, es evident by the discrepancy between the two numbers you mentioned Reddick has).

      Second, there is almost equally high distribution in the results of LD% as it is in BABIP. LD% reaches the r=0.7 point at ~ 600 BIP (balls in play). That is the point where we, said in a very simplified manner, can attribute half of observed talent level to true talent level, and half to random distribution. Reddick’s improvement to his career numbers (non-existent if you trust FG, ~3 percentage points if you trust BR) is based on 200 BIP and is far, far from a meaningful sample. So, I wouldn’t say that he is hitting line drives at a career high level, based on that evidence.

      And finally, and most importantly, LD% tells us something about the angle of the batted ball, not about its velocity. If you hit the ball at the same angle but weaker, you will have worse chance of such ball being a hit. This is exactly why I used NL Pitchers xBABIP to show that. They also hit their fair share of line drives (obviously fewer than professional hitters, ~17% vs ~20%) but they fail to reach their expected BABIP by far (~.060 to ~ .080). This is why Reddick’s xBABIP is not necessarily a point to regress to. If we think that he is not hitting the ball hard, he will not reach his xBABIP goals.

      And we know two things about Reddick this year with much, much higher certainty than that he is hitting line drives. We know that he is hitting a high amount of ground balls (that stat reaches r= 0.7 at 80 BIP) and we know he is not hitting those fly balls he does manage to hit very far (HR/FB reaches r = 0.7 at 50 FB). Both are very indicative of the skill set of a person not hitting the ball very hard.

      As a summary, here are the main thoughts I’d like you to think of:

      1. Be reluctant to use LD% as a yearly stat. Only a few batted balls classified “fly ball” instead “line drive” (or vice versa) by an inexperienced and sometimes overmatched stringer can cause a swing of several percentage points
      2. LD% only tells us something about the angle of the bat towards the ball. On every 100 balls put in play, Miguel Cabrera will only have four more line drives than an average NL pitcher
      3. If a player also displays changes in his GB% and his HR/FB profile, look there first to try to find guidance for changing true talent level

      Because survival is insufficient
      • SamYam Aug 1,2013 10:57 pm || Up

        Interesting points. I’d like to hear danmerqury’s thoughts on batted ball classification because he’s now working at BIS. I think it’s fair to question the LD% data, but no matter which way you slice it, one thing is (relatively) certain: his LD% certainly hasn’t gotten worse, which is, I think, a good indication. Often, you’ll see a struggling hitter’s LD% dip. The other good sign, IMO, is that based on the eye-test (which will have to do unless I can somehow get access to Hit f/x) is that he seems to be hitting his liners very hard.

        The GB%/FB% and HR/FB% are clearly very concerning though. It’s clearly sapped a ton of his power, which is where he derives a huge part of his offensive value. I don’t think I’m ready to argue that Josh Reddick’s true talent level is a good hitter, especially based on these concerns. But based even on these skewed batted ball profiles, his xBABIP is 60-70 points above his BABIP. In my post at OC, I specifically avoiding saying that his true talent level is an above-average hitter. I wrote “Reddick (probably, hopefully) won’t hit worse than Willie Bloomquist the rest of the year”, which I think is a fair guess even if his BABIP regresses to even something like .275, which is around his career average. But even there I wasn’t so certain!

        When I wrote my original post, I actually didn’t even plan to get into BABIP stuff at all- my overall point was more that Reddick is a valuable contributor even with a replacement-level bat based on his defense, and that his true talent level is probably somewhere above replacement level. I stand by that- I don’t think Reddick is Yuniesky Betancourt at the plate. I do think it’s pretty clear he’s been unlucky, but just how unlucky is questionable.

        • elcroata Aug 2,2013 1:25 am || Up

          Yeah, perhaps Dan can shed some light on hit classification.

          Often, you’ll see a struggling hitter’s LD% dip

          For the fun of it, I looked at all the hitters who had at least 100 PA in 2011 and in 2012. I completely arbitrarily defined “a struggling hitter” as someone within +/- 5 wRC+ of Reddick’s current delta to the yesteryear, or in other words, all the players whose wRC+ dipped between 25 and 35 points. What I was left with was a sample of 35 hitters.
          Eighteen of those hitters had worse LD% than in yesteryear, one had the same and 16 had better. Seven of them had their LD% increase by at lest two percentage points.
          As a group, their LD% dipped by one percentage point. So, I guess we can observe what you mentioned, but probably not to the extent one would intuitively expect.

          And, I know what you said, and I agree with it. Reddick has value even as a below average hitting corner outfielder. One of the most annoying aspects of mass wisdom on ** was the prevalence of the “ZOMG CAN’T WIN W/O HR FROM CORNER OUTFIELD!!1!”. And yes, just by looking at his career numbers he is likely to regress some. But I don’t think he is necessarily being particularly unlucky now (I have to admit, I have watched only a handful of games this year, so I am not aware of what most of hits looked like, for better or for worse).

          The reason that I focused on this small subset of what you wrote is that I prefer general discussions, because people can better keep their emotions and preconceived opinions in check, and because they have much higher value, as they apply to the game as a whole.

          Because survival is insufficient
          • SamYam Aug 2,2013 8:02 am || Up

            Sounds to me like we’re mostly in agreement. I actually have to generally take the opposite approach to what you’re saying when I’m over there… if I write a post about a concept rather than about a specific player (or, say, the “A’s pitching staff), generally, nobody reads it. In the end, I’ve felt I’ve had to tailor what I write to that audience. But personally I get a lot more value out of the broader discussions as well. In all, it seems to me like I have to spend more time here, where perhaps people will be a bit more conversant/actually willing to engage math.

            Also, can you send me a link to that study of r values and the number of batted balls it takes to get there? Or is that something you’ve found on your own?

            • elcroata Aug 2,2013 9:35 am || Up

              Pretty much.

              I do a lot of such research on my own, but in this case I was referring to Pizza Cutter’s work that can be fund on FG.

              Because survival is insufficient
          • lenscrafters Aug 2,2013 11:18 am || Up

            Plate discipline can also figure into this calculation. From what I remember, you (or someone else) showed that different zones of the strike zone resulted in different BABIPs. We all know Reddick has a tendency to swing at anything close and borderline around the zone so I don’t think it’s unreasonable to think that his mediocre BABIPs could be a result of poor contact stemming from that tendency.

          • DFA Aug 2,2013 11:56 am || Up

            As a group, their LD% dipped by one percentage point. So, I guess we can observe what you mentioned, but probably not to the extent one would intuitively expect.

            Is that a requisite sample size to conclude that? I mean it seems like a pretty small difference.

            In play, run(s)! Talk dirty to me gamecast, talk dirty. - Nevermoor
        • danmerqury Aug 2,2013 8:09 am || Up

          Hey there.

          So as far as the LD stringer thing, BIS changed its classification method in 2009. Now, we time every batted ball from bat to glove (or ground or wall or anything), with definitions based on hang time and location. We classify every batted ball as a liner, a fliner liner, a fliner fly, or a fly. Now here’s the thing: I don’t know if Fangraphs uses our data or the rough eyeballed Gameday stringer data. I also don’t know how they group up the fliners. But over the last three years, batted ball stats have been much, much more objective.

          • ptbnl Aug 2,2013 8:25 am || Up

            I’d just like to chime in (server permitting) and say that this conversation is FKing cool – thanks all.

            If this is His will, He's a son of a bitch.
          • SamYam Aug 2,2013 8:53 am || Up

            according to the Fangraphs glossary, they use BIS data. I wonder if bref uses the gameday data

          • elcroata Aug 2,2013 9:36 am || Up

            We meet again, baseball researcher.

            Thanks. Also, Fliner Fly should be somebody’s name.

            Because survival is insufficient
            • PDXAthleticsfan Aug 2,2013 9:56 am || Up

              Fliner McFly is Marty McFly’s (back to the) future grandson. After driving a hybrid electric/diesel road bike into a lightning storm at 88 mph, he will travel to 2001 and put a trip wire 10 feet in front of home plate to make Jeremy Giambi slide.

              A soliloquy of fresh-sounding ideas which would probably be disastrous.
            • danmerqury Aug 2,2013 8:50 pm || Up

              Fliner Fly was the album Curtis Mayfield was working on when he died.

    • Future Ed Aug 2,2013 7:14 am || Up

      welcome. come by more often

      I have $5. No I don\'t.
      • Future Ed Aug 2,2013 7:16 am || Up

        holy shit. my fisrt comment approved in 24 hours.

        i made some hot hot jokes that got eaten by the server, too. too bad for you people

        I have $5. No I don\'t.
        • Bed Aug 2,2013 8:59 am || Up

          I know…I finally finished my epic work about VORP but it wouldn’t post…oh well, I guess you folks will have to wait another five years to read it.

          But seriously, folks....
          • elcroata Aug 2,2013 9:38 am || Up

            Actually, I think that’s what brought the site down.

            Because survival is insufficient
        • Dial C for Concupiscence Aug 2,2013 9:38 am || Up

          Ed’s hot hot jokes? Noooooooooo

    • andeux Aug 2,2013 9:57 am || Up

      Chiming in a little late here:

      Saying he has been unlucky because his LD% or his xBABIP are decent is kind of begging the question. When people first started looking at this stuff a few years back, LD% was used by itself as a predictor of BABIP. Presumably whatever formula BPro is using is a little more sophisticated than that. But either way, before dismissing deviation from that predictor as luck, we need to know how good a predictor it is in general. One way to do this would be look at the variance in (BABIP – xBABIP) over a whole population, and compare it to the expected sample variance. The difference reflects some variance in skill that is not captured by the predictor, and would also tell you how much to regress an individual’s observed value toward the predicted value.

      I don’t know what those numbers would be in this case, but my general sense based on what I’ve read on this subject in the past is that the predictors are not all that great, and there is a lot of skill that they do capture. ec has already covered some of the reasons for this in his comments: lots of noise in the measurement of batted ball types themselves, and in the proxies used for other factors that effect BABIP like how hard someone hits the ball and how fast they run.

      TINSTAAFK
      • andeux Aug 2,2013 10:12 am || Up

        Oh, and on your larger point about Reddick being valuable even with mediocre defense, agreed.

        TINSTAAFK
  4. Soaker Aug 1,2013 1:53 pm

    I see your comment, SamYam, but FK is in a pretty severe Callapso mode right now and my attempts to approve it aren’t sticking. We’ll get you on-board soon though.

    What I discovered Blew. My. Mind. -- Pat Boone
  5. nevermoor Aug 1,2013 3:47 pm

    Jeff reads FK. Is nowhere near as dumb as Grant.

    "There's never enough time to do all the nothing you want"
    • elcroata Aug 1,2013 10:21 pm || Up

      Did you regress those numbers?

      Because survival is insufficient
    • andeux Aug 2,2013 10:11 am || Up

      Among contemporary players, we find Dan Johnson near the bottom of the big-league BABIP list, at .244.

      Dan. Johnson.
      Besides being a running gag here, Dan Johnson really is a good example of BABIP as skill. After a decent rookie season in 2005 (.275/.355/.451 including a BABIP of .278), he fell off the next two years, with BABIP of .245 and .246. I remember the A’s manager or hitting coach commenting some time during those two years that Johnson was having trouble with pitch recognition, and was consistently late on fastballs and early on breaking pitches.

      And if you look at his spray charts for that time (this used to be available on the official site, but I can’t find it now), you’ll find a ton of fly balls to the opposite field (late on the fastballs) and pulled ground balls (rolling over on the breaking pitches). IOW, lots of weakish contact that was easy to field. Plus the predictability made the infield shift especially effective, further reducing the number of balls that could get through for hits even when he did hit them a little harder.

      TINSTAAFK
      • lenscrafters Aug 2,2013 11:14 am || Up

        Daric Barton post 2010: decent line drive rates, but crappy BABIPs due to, from what I can figure, the outfielders never having to play more than 30 feet behind the infield.

        • andeux Aug 2,2013 11:24 am || Up

          Also Kendall, Suzuki, Weeks, though in all cases their LD% did go down a little also as their hitting declined.

          TINSTAAFK
        • mikeA Aug 2,2013 11:56 am || Up

          That would be lack of power much more than positioning I think. All three batted ball types hit weakly.

      • elcroata Aug 2,2013 11:23 am || Up

        Defensive shifting is a great keyword. There has been an increasing use of shifts lately.

        Over the last few years, the use of The Shift Defense has increased dramatically in baseball. Here are the Major League Baseball totals of the number of shifts in baseball, as tracked by Baseball Info Solutions (BIS).

        2010  	 2011  	 2012  	 2013 Projected
        2,465	 2,358	 4,577	 7,586

        Being predictable as a hitter will cost you in BABIP, especially if you are a lefty. Matt Swartz debunked theory that David Ortiz (and ilk) was clutch, because he came uo with big hits with men in scoring position. The simple explanation was that his BABIP (but not other numbers) was higher then, because defenses were not shifting.

        Because survival is insufficient
        • SamYam Aug 2,2013 4:22 pm || Up

          This is a great point. I might take a future post to look into the A’s use of the shift.

        • elcroata Aug 3,2013 11:48 am || Up

          Also Dan, can you chime in on how BIS defines the shift. Stringer putting in the data or already using Field f/x?

          Because survival is insufficient
          • danmerqury Aug 3,2013 3:21 pm || Up

            We don’t use FIELDf/x, and you asked this just at the right time, because I just got back home from a SABR convention panel which touched on it briefly. It’s not out yet. They’re still in sort of a pilot alpha mode, so it’s only in a handful of parks. San Francisco, Boston, and two or three others. So no one has access to FIELDf/x in any sort of meaningful analytical sense other than testing.

            No, we just do it the old-fashioned way. Wait for an overhead shot or the ball-in-play camera angle, and look at infielder placement.

    • dmoas Aug 2,2013 1:47 pm || Up

      is the unread comments widget running slowly (along with page loading) or is it just me?

      • PDXAthleticsfan Aug 2,2013 1:48 pm || Up

        It is running slowly for me, as well.

        A soliloquy of fresh-sounding ideas which would probably be disastrous.
      • Dial C for Concupiscence Aug 2,2013 2:21 pm || Up

        Safe to assume that everything is running slowly and it’s all related to the transition issues NM is dealing with.

        • dmoas Aug 2,2013 2:39 pm || Up

          As long as it’s not just me. Everything on this machine is running slow for some reason as it is.

      • elcroata Aug 2,2013 2:44 pm || Up

        One thing that I found that works a bit better is if I open all the posts in the new tabs, and keep the front page on its separate tab.

        Because survival is insufficient
  6. lenscrafters Aug 2,2013 11:10 am

    I’m curious as to how his minor league numbers figure into this discussion.

    We can’t look too much into his minor league BABIPs given that there is even more noise in those than major league BABIPs, but I’m wondering if his minor league HR/FB data can be also be used to ascertain his true talent HR/FB. One reason I (and others) were cautious about Reddick’s power surge in the first half last year was that he never really demonstrated this skill in the minors (or rather, much offensive ability at all). My feeling is that Reddick is a guy who’s never going to hit for high or even average BABIPs so he’s going to live and die on his HR/FB rates.

    • SamYam Aug 2,2013 4:15 pm || Up

      He’s had low BABIPs even throughout the minors, except at low levels. I think you’re right. But his current BABIP is significantly lower than even that, so I can’t imagine it staying down there forever

  7. vignette17 Aug 2,2013 11:12 am

    Hitter BABIP is overused, especially with prospects. I don’t recall seeing data on this topic, but whenever someone points to low hitter BABIP for a minor leaguer and expect it to regress to league mean, I get a little wary.

  8. aardvark Aug 2,2013 12:44 pm

    Wait, what’s the difference between hBABIP and pBABIP.

    • elcroata Aug 2,2013 2:05 pm || Up

      I just use that to distinguish whether we are talking about hitters’ or pitchers’ BABIP, as they have different causes and association with skill level. I don’t think anybody else does.

      Because survival is insufficient
  9. elcroata Aug 2,2013 3:21 pm

    On a tangent. Some of the comments from that thread:

    It’s closely related to AN’s penchant for not understanding how math works.

    by Ben Folds’ Nuts on Aug 1, 2013 | 10:05 PM up reply

    Flagged.

    I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her “Hotties,” and isn’t afraid to show it. -IM4Oakgal
    by Nico on Aug 1, 2013 | 10:09 PM up reply

    Really? Out of all the personal attacks that get made on AN, this is the one you choose to publicly flag?
    It’s almost like you have a specific personal issue with me.

    by Ben Folds’ Nuts on Aug 1, 2013 | 10:26 PM up reply

    There’s nothing personal.
    Email me offline if you want clarification. (Except I’m going to bed so I won’t reply until tomorrow.) G’nite.

    I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her “Hotties,” and isn’t afraid to show it. -IM4Oakgal
    by Nico on Aug 1, 2013 | 10:31 PM up reply

    Because survival is insufficient
    • andeux Aug 2,2013 3:25 pm || Up

      Plus ça change…

      TINSTAAFK
      • ptbnl Aug 2,2013 3:43 pm || Up

        plus c’est la même chose.

        If this is His will, He's a son of a bitch.
        • elcroata Aug 3,2013 11:55 am || Up

          Flagged for thinking that everybody else who didn’t write things in French is an idiot and beneath you. Just because you are a rocket scientist, it doesn’t mean that you can come here and tell us all we are idiots. This is like a sports bar, man. Just chill, we are all on the same side.

          Go A’s!

          Because survival is insufficient
          • dmoas Aug 3,2013 12:06 pm || Up

            Flagged for flagging. Coincidentally, had I would have flagged Tounces’ comment for baiting and meta if I were still willing to go there.

            • 5Aces Aug 3,2013 12:08 pm || Up

              First rule of **, you cant talk about **.

              Camelot sure fell apart, didn't it? -Steve McCatty
              • Dial C for Concupiscence Aug 20,2013 9:57 am || Up

                First rule of **, you cant talk about **.

    • elcroata Aug 2,2013 3:25 pm || Up

      SHRUBBERY!

      Because survival is insufficient
    • 5Aces Aug 2,2013 3:29 pm || Up

      Has it been 18-20 months already? Man how time flies…

      Camelot sure fell apart, didn't it? -Steve McCatty
    • PDXAthleticsfan Aug 2,2013 3:31 pm || Up

      Looks like a good time to recruit new FKers.

      A soliloquy of fresh-sounding ideas which would probably be disastrous.
      • elcroata Aug 5,2013 12:26 am || Up

        I’ve scanned front page postings a bit and somehow, I’m just not that sure there is so much to recruit. There was this guy who wrote 8000 words, basically repeating this single thought (and that is only a first part of the saga, part 2 is still to come up):

        Of course, that doesn’t change the fact that Smith has been a far inferior hitter in an A’s uniform than he was in a Colorado one

        One million numbers, but he somehow omits these four:

        Smith’s wRC+ in 2010 with Colorado: 95
        Smith’s wRC+ in 2011 with Colorado: 110
        Smith’s wRC+ in 2012 with Oakland: 107
        Smith’s wRC+ in 2013 with Oakland: 94

        Adjusting for park and league – it really does exist.

        Because survival is insufficient
        • nevermoor Aug 5,2013 10:47 am || Up

          4 points of wRC+: CLEARLY INFERIOR HITTER!

          "There's never enough time to do all the nothing you want"
          • 5Aces Aug 5,2013 10:52 am || Up

            Nor as good as he was in COL.

            BUT WHAT ABOUT HIS SMILE?!? Sorry, instant reaction…

            Camelot sure fell apart, didn't it? -Steve McCatty
      • elcroata Aug 5,2013 2:37 am || Up

        And then, there’s another guy, who I think writes a lot about hitting techniques and wrote a huge post about correlations between batted balls and different stats. He looked at 650+ hitters in last year and compared their batted ball profiles with their statistical outcomes.

        His conclusion is:

        A couple things to note here. Homerun-to-flyball ratio is a GREAT predictor of OPS. Everything else pales in comparison.

        This is exactly how you should not use correlation, and exactly the statement that is completely unfounded. You can not look in sample for correlations and then say that they have predictive value.

        Not surprisingly, he finds out even better correlation between HR/FB and HR%. So, in other words if you have hit a high percentage of fly balls out of the park, you have hit a high percentage of home runs. No shit.

        If you want to make any statement about predictive value of a correlation, you have to look out of the sample.

        It’s a pity that there are people with obvious energy and enthusiasm there, yet nobody there has a basic understanding to push them in the right direction.

        Because survival is insufficient
    • dmoas Aug 2,2013 3:54 pm || Up

      Ha. That’s funny. Someone shoot that poor guy an email to here.

    • SamYam Aug 2,2013 4:12 pm || Up

      I’ve been trying to figure out a way to subtly direct him here without pissing of the “powers that be”. His email isn’t listen in his sbnation profile.

      • SamYam Aug 2,2013 4:13 pm || Up

        Typo. It’s up to you to figure out whether I really meant “pissing off” or “pissing on”

        • Future Ed Aug 2,2013 9:48 pm || Up

          don
          t editors have back stage powers? Are you an editor? have gigs do it.

          FYI: nobody here cares if you post both places.

          I have $5. No I don\'t.
          • elcroata Aug 3,2013 11:51 am || Up

            This
            I think you play a valuable role in educating the readers over there. There are lot of people who are curious and willing to learn if somebody takes the time to explain things to them. While the OC might not be the best place in the world to have a discussion if you don’t fit into the prevailing mold, that doesn’t mean that you can’t both reach an audience and educate them.

            Because survival is insufficient
            • dmoas Aug 3,2013 11:53 am || Up

              Plus… um… we kind of a need a recruiter…

              • elcroata Aug 3,2013 11:58 am || Up

                Yeah, but since I brought him here, we’ll split 10% we get for selling new users’ cookies.

                Because survival is insufficient
                • dmoas Aug 3,2013 12:07 pm || Up

                  That’s entirely up to the two of you, however, if he were to take that position, given our acquisition rate, I’d think we need to offer 20% that could be split 10% a piece or 15/5.

                • Future Ed Aug 3,2013 5:27 pm || Up

                  cookies?

                  I have $5. No I don\'t.
              • SamYam Aug 3,2013 9:57 pm || Up

                Ha. Been trying to figure out how to do it subtly

          • 5Aces Aug 3,2013 12:02 pm || Up

            Nobody cares here. But fair warning, if you head there and brings this site up much you will hear something about it.

            Camelot sure fell apart, didn't it? -Steve McCatty
      • AV Aug 3,2013 3:18 pm || Up

        he doesn’t believe in magic.

        so he should come over just so we can teach him to spell it right.

        *i’m* AV. alex vause. put this loon in psych before she hurts someone.
    • lenscrafters Aug 2,2013 4:15 pm || Up

      They’re still operating under the pretense that there’s a moderation policy?

    • Jennifer Aug 3,2013 12:04 pm || Up

      Email me.

    • Bed Aug 3,2013 2:11 pm || Up

      I didn’t think anyone said anything interesting enough on AN to get flagged anymore.

      But seriously, folks....
      • nevermoor Aug 3,2013 5:47 pm || Up

        Doesn’t sound like he did.

        "There's never enough time to do all the nothing you want"
    • Dial C for Concupiscence Aug 5,2013 10:20 am || Up

      Ah, the ol’ punish you publicly and run away 1-2 punch. I guess it’s hard to break habits, even on a sinking ship.

      And now I’m imagining Cap’n Napo yelling about dirty dishes on the Titanic.

      • Future Ed Aug 5,2013 10:25 am || Up

        serious question: is Cindi more sexist, or homophobic?

        I have $5. No I don\'t.
        • Dial C for Concupiscence Aug 5,2013 10:46 am || Up

          Super sexist. Cindi essentially amounts to “OMG, I’m a girl, what’s a baseball? lolz!”

          • elcroata Aug 5,2013 10:50 am || Up

            That was my take, too. What’s the argument for homophobic?

            Because survival is insufficient
            • Future Ed Aug 5,2013 10:56 am || Up

              cindi talk(s)(ed) about hotties. A proxy for a Napo to enter the ass conversations that were prevalent in the height of harden crosby mulder **.

              I have $5. No I don\'t.
          • 5Aces Aug 5,2013 10:53 am || Up

            Garza told her to STFU.

            Camelot sure fell apart, didn't it? -Steve McCatty
            • Future Ed Aug 5,2013 10:57 am || Up

              my alternate personality would never speak out of turn

              I have $5. No I don\'t.
              • dmoas Aug 5,2013 11:04 am || Up

                Mine would. She’s a total bitch. Unrelated, she might be Black, but most likely wanna-be-Black so she turns up the sass and attitude to an 11. Come to think of it, she might be slightly racist. That worries me. I think I might need to kill her and come up with a new one. Is that still murder?

                • Future Ed Aug 5,2013 11:25 am || Up

                  Not only did I laugh out loud, I sustained it while reading further. well played.

                  I have $5. No I don\'t.
          • DFA Aug 5,2013 6:12 pm || Up

            Its mostly just really creepy. Like really creepy.

            In play, run(s)! Talk dirty to me gamecast, talk dirty. - Nevermoor
    • Glorious Mundy Aug 19,2013 5:42 pm || Up

      Hahahahaha

Leave a Reply