Darren O’Day: Lefty Killer

If you’re reading FanGraphs, you probably like baseball. If you like baseball, you’ve probably watched an Orioles game this year. If you’ve watched an Orioles game this year, you’ve probably seen Darren O’Day pitch. Even as relievers go, he pitches a lot, as he appeared in 69 regular season games, the wild card play-in game, and has now pitched in all four games of the ALDS. And, if you’ve seen O’Day pitch this year, then you’ve probably noticed that he throws from a really low arm slot, but thanks to our friends over at SBNation, we’re going to show you O’Day striking out Alex Rodriguez last night, just so you can appreciate the funkiness in all it’s glory.

Watching that, you almost feel bad for Rodriguez. How is he supposed to hit that? What is any right-hander supposed to do with a slider from that arm angle that starts on the outer half of the plate and ends up a foot outside? And, if that was the match-up option, why not send Raul Ibanez up to pinch-hit for Rodriguez? Certainly, a lefty would have far more success against O’Day’s drop-down delivery, right?

You would think. And, with pretty much every other low slot/sidearm/submarine guy on the planet, you’d be right. Chad Bradford had insanely large platoon splits. Same deal with Mike Myers. Justin Masterson‘s low arm slot is the main reason he struggles against left-handers. Guys who throw the ball from that kind of arm angle are usually brutal on same-handed hitters and ineffective against guys from the opposite side of the plate.

And yet, O’Day ran through the entire Yankees line-up last night, including six guys hitting from the left side: Nick Swisher, Curtis Granderson (okay, not much of an accomplishment right now), Raul Ibanez, Ichiro Suzuki, Mark Teixiera, and Robinson Cano. That’s a pretty decent collection of left-handed bats, and O’Day just mowed them down. And perhaps the most amazing part of the entire thing is that we probably shouldn’t have been surprised.

Left-handed batters hit just .200/.237/.420 (.279 wOBA) against O’Day this year, and even for his career, they’re only hitting .232/.285/.403 (.293 wOBA) against him. To put that in comparison, lefties hit .225/.267/.402 (.288 wOBA) against CC Sabathia this year. Sabathia, of course, is left-handed. And really good.

So, how does a low-slot right-hander hold lefties to the same offensive performance as a power lefty?

Pitch backwards. Last night, O’Day threw a first pitch slider to four of the six left-handed batters he faced, and three times it got him ahead in the count. Nick Swisher fouled off his first pitch slider. Raul Ibanez took it for strike one, as did Robinson Cano. Most hitters go up looking first pitch fastball, and O’Day simply declined to give them that pitch, which is a pretty good idea for a guy who sits 85-87 and releases the ball from his hip. By starting lefties off with the slider and usually getting a free strike, O’Day was able to expand the zone with his fastball.

Here’s BrooksBaseball’s PITCHF/x plot of O’Day’s pitches to left-handers last night.

Focus on the location of the fourseam fastballs — represented in green — he threw to left-handers. Up and almost always away, but mostly up. O’Day elevates his four seam fastball extremely well, and despite only throwing it in the strike zone to a left-handed batter once, he got two swinging strikes, plus a called strike on the one he grooved down the middle.

Sliders early, high fastballs late. From a side-armer whose Twitter profile claims he enjoys “throwing like a girl”. It’s an unusual skillset to the say the least, but it works, and it’s worked for his entire Major League career. O’Day is the rare drop-down pitcher who doesn’t really have much of a platoon split, and that’s because he doesn’t really try to get groundballs. If his fastball was down, lefties would crush him. Up out of the zone, though, he can get swinging strikes just based on location, and his willingness to use his slider to get ahead in the count keeps hitters off balance.

Most fastball/slider pitchers use the heater to get ahead and the slider as the put away pitch. O’Day does it backwards, and it works really well. When broadcasters talk about how pitching is disrupting a batter’s timing, this is what they’re talking about. O’Day has pedestrian stuff and delivers it from an angle that should make him prone to left-handers, but the way he mixes his pitches and the locations he puts them in allow him to be a relief ace and not just a match-up righty.

Even after throwing 30 pitches last night, Buck Showalter won’t be afraid to use O’Day tonight, as Game 5 is an “all hands on deck” situation. He’ll almost certainly go with Brian Matusz and Troy Patton in critical situations against left-handers, but he shouldn’t be afraid to turn to O’Day either. O’Day’s funky delivery notwithstanding, he’s more capable than most at getting left-handed hitters out.

Print This Post

Dave is a co-founder of USSMariner.com and contributes to the Wall Street Journal.

39 Responses to “Darren O’Day: Lefty Killer”

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

    Sergio Romo says hi.

    Vote -1 Vote +1

  2. payroll says:

    Immediately brings to mind Chad Bradford/Moneyball. Its almost cheating what these guys are doing. I foresee a day where all non-changeup reliant pitchers are taught to drop down against opposite handed batters.

    Vote -1 Vote +1

    • Dave Cameron says:

      But this isn’t Chad Bradford at all. Lefties hit .315/.404/.451 against Bradford. He’s exactly why you don’t want to drop down against opposite handed hitters.

      The point is that O’Day is doing something that almost never works, but for him, it’s working. It’s not that sidearming against lefties is a good strategy.

      Vote -1 Vote +1

      • reillocity says:

        I followed a minor league lefty reliever this year who throws sidearm against lefties but will throw over the top and sidearmed also against righties. I’m surprised that we don’t see more of that kind of thing from sidearmers when facing opposite handed bats.

        Vote -1 Vote +1

  3. Krog says:

    I wonder if it has to do with the velocity on the slider. Did Randy Johnson have success against lefties despite throwing from a low angle? Is O’Day throwing hard as well?

    Vote -1 Vote +1

    • TX Ball Scout says:

      You mean righties?

      Vote -1 Vote +1

    • Dennis says:

      Randy Johnson was lefty, and he had success against his opposite hand (righties). But I’m pretty sure that the fact that he’s 6’11″ and had a 5 foot head start because of his stride length, and the fact that he threw 99 mph well into his 120th pitch had something to do with that.

      I think we’re talking about pitchers with fringey stuff, not people with ace stuff that happen to drop down.

      Byung-Hyun Kim probably had the best stuff of any softball throwers (his fastball got into the low-mid 90s) and he got smacked around by opposite hand hitters for most of his career, 2003 notwithstanding.

      Vote -1 Vote +1

  4. Tim says:

    Great article Dave. While Jim Johnson is the high velocity closer who gets a lot of the credit. I feel like O’Day is the best reliever in the accomplished Orioles bullpen. The guy is lethal against righties combined with his great approach against lefties, which that allows him to succeed in a non conventional way. I personally think he’s one of the top relief pitchers in baseball, and perhaps the best non closer.

    Vote -1 Vote +1

  5. ShaveNaHC says:

    O’Day has to be the Orioles MVP of this series. He’s been just brutal on hitters all four games.

    Vote -1 Vote +1

  6. chuckb says:

    I wonder if O’Day pitches righthanders the same way. Does he still throw the fastball up in the zone against them? It seems to me that would be a prescription for disaster. One reason that people like Bradford were so successful against righties is that they were able to keep the ball down against them. My guess is that he pitches righties quite differently from the way he pitches lefties and, if the lefties are smart, they’ll start looking for that first pitch slider where they can hit it.

    Vote -1 Vote +1

  7. Baltar says:

    I han’t seen O’Day during the season, and I was absolutely delighted to watch him in this series.
    To see more of him is by itself enough reason to root for the O’s tonight.

    Vote -1 Vote +1

  8. GonzoFlyBall says:

    Any thoughts of revisiting the Uhera trade that brought back Chris Davis, Pedro Strop and O’Day? Would be interesting to analyze the trade. To me, this represents the most significant move made by Baltimore’s front office, in terms of return, to this magical season.

    Vote -1 Vote +1

    • Derp says:

      Those were all different moves. The Uehara trade netted the Orioles Davis and Tommy Hunter. Strop was for Mike Gonzalez, and O’Day was a waiver pickup in the offseason

      Vote -1 Vote +1

  9. jose says:

    I still think Gerardi did wrong by not pinch hitting either Raul or Chavez in that critical stage of the game:
    bottom of the 8th with one out and men on 3rd and 2nd base. Go ahead run 90 feet away and then three outs for the Series win. Gerardi blew it.

    Vote -1 Vote +1

  10. FRBR1966 says:

    Just let me add that O’Day has only given up *one* hit in his 7 post-season innings pitched. That and a walk to give him a .029 WHIP and .048 BAA!

    Vote -1 Vote +1

  11. Sean says:

    23 batters faced in the ALDS. He’s allowed 2 to reach base – 1 hit, 1 walk.

    Vote -1 Vote +1

  12. During the Giants-Reds Game 5 yesterday, the announcers kept saying Sergio Romo is tougher on righties than lefties, which would make sense since the slider is his best pitch. However, he actually held lefties to a lower OPS than righties this year. I’d be interested to see how Romo has success against lefties, too.

    Vote -1 Vote +1

    • ElJimador says:

      Giants broadcasters (presumably parroting what they hear from management) say that Romo’s development of a reliable 2 seam fastball is the reason that he’s had more success vs. lefties this year and why Bochy was confident installing him as closer late in the year after Casilla developed blister problems and faltered a bit. What this leaves out is A) that Romo has always been just as good against lefties (career wOBA against = .237 by LHB vs. .241 by RHB), and B) that even having assumed the closer role now, Romo was still used overwhelmingly vs RHB this year and in fact even moreso than usual (71% of batters faced this year with a platoon advantage vs. 66% career).

      So while it’s nice to hear that the Giants are figuring out that Romo can get lefties out too, it would be even nicer to see evidence over a full season that they actually believe it. As it is now, I fear that 1 more Eric Hinske type shot by a lefty in this year’s playoffs could just as easily move him back to Bochy’s ROOGY doghouse forever.

      Vote -1 Vote +1

  13. MGL says:

    “Left-handed batters hit just .200/.237/.420 (.279 wOBA) against O’Day this year, and even for his career, they’re only hitting .232/.285/.403 (.293 wOBA) against him. ”

    O’day is a really good reliever, that’s why lefties have only hit .293 against him in his career.

    His career wOBA against, which I don’t think you mentioned, is a lot better, .264.

    His career platoon ratio is 1.11. The average lefty pitcher has a platoon ratio of 1.13, so he is right around what we would expect from a typical lefty.

    But, we have to expect that a side-arming lefty who throws mostly sliders and fastballs (the pitches with the greatest platoon differentials) would have a true platoon split of more than 1.13.

    So we would regress his 1.11 actual platoon ratio to something maybe like 1.20 or so. After all he has only faced 364 lefty batters in his entire career! That is the equivalent of around 85 innings of pitching to lefties only! So you really think that there is not going to be enormous fluctuation in that .293? There will be.

    And he has only faced 1,004 batters in his entire career. That is the equivalent of one season for a full time work horse starter. Have you ever seen the random variation in platoon splits for starters from year to year?

    Sabathia himself had no platoon split in 2012 and an enormous one in 2011 and a reverse one in 2010!

    How about we make up some narratives in each of those years as to why he had those splits. First, for this year we have to make up a narrative as to why he was such a righty killer (.291 wOBA against) and why lefties hit him equally as well (.288)

    Then last year, we have to make up a narrative about why he was such a lefty killer (.248).

    Then in 2010 we have to make up a narrative as to why lefties hit him better than righties!

    That’s why, folks, we have to regress observed platoon splits so much. They have very little stability in small samples and since they are based on two samples per season (versus LH and RH batters), one season for a reliever is a ridiculously small sample. For example, O’day faced 94 LHB in 2012 and only 31 in 2011! Do you really think platoon splits in those years are going to be even remotely reliable? They won’t be.

    For lefty pitcher, we regress splits 50% toward the mean when they have faced 1000 LHB. O’day for his career has faced 293. That means we regress his observed splits 77% toward the mean! That means that simply using a population mean (lefty relievers who throw side-armed or underhanded with primarily a fastball and slider) is much better than using observed splits. IOW, we might just as well ignore the observed splits.

    In any case, if we regress his observed 1.11 toward a population mean of say 1.20, 77%, we get an estimated true platoon split of 1.17.

    All this other stuff is nonsense, in my opinion.

    P.S. Dave, sorry about picking on you lately, but I think you deserve it…

    Vote -1 Vote +1

    • Sivart says:

      Fortunately O’day is in fact a lefty reliever.

      Vote -1 Vote +1

    • PM says:

      The meds wear off early today? The author was merely pointing out that O’Day has had more success vs. LHBs than one might have thought, and postulating as to why this might be the case. The title was written partly tongue in cheek (which would be apparent to any sane reader). I don’t personally know either of the nerds in question here (nor do I care to), but it’s pretty obvious one of them is venting lifelong frustrations.

      Vote -1 Vote +1

  14. MGL says:

    Of course O’Day is a RHB. So substitute RHP for LHP.

    It doesn’t change anything. It actually makes my criticism stronger. The average platoon split for a RHP is 1.08. So O’Day’s observed 1.11 is actually higher than the average RHP likely due to his arm angle and pitch repertoire. So David’s whole thesis makes no sense. A “lefty killer” means as compared to how a pitcher does against righties I would think. Otherwise almost all good pitchers are “righty killers” AND “lefty-killers.”

    Plus it takes 2000 PA to regress a RHP platoon split 50% toward league average. That makes O’Day’s observed split even more unreliable.

    Vote -1 Vote +1

    • Nadingo says:

      Reading Comprehension issue:

      “A “lefty killer” means as compared to how a pitcher does against righties I would think.”

      That might be a reasonable assumption in isolation, but not in the context of this article. Try this for the actual focus of the article:

      “Guys who throw the ball from that kind of arm angle are usually brutal on same-handed hitters and ineffective against guys from the opposite side of the plate.

      And yet, O’Day ran through the entire Yankees line-up last night, including six guys hitting from the left side:”

      Guys who pitch like O’Day usually have larger platoon splits than most pitchers (so your 1.08 number is irrelevant for this conversation), but O’Day’s performance defies those expectations.

      Vote -1 Vote +1

  15. Nox says:

    Dave Cameron is pretty bad at stats.

    There’s really nothing left to add after MGL’s crushing of him here and on The Book blog but holy crap I thought we could expect a little bit more than THIS from the site’s lead writer.

    Vote -1 Vote +1

    • Justin says:

      Bad at stats, crushing of him?

      Really? Even sabermetric writers have fanbois and anti-fans now? Hilarious.

      Vote -1 Vote +1

      • Nox says:

        “Even sabermetric writers have fanbois and anti-fans now? ”

        No, this has nothing about being a fan or anti-fan. MGL is a complete dick but he also happens to be a complete dick who’s 100% right here. What Dave is presenting as “evidence” here is amateurish at best.

        From reading the back and forth on The Book blog it’s become pretty clear that Cameron doesn’t actually understand regression to the mean beyond the general idea that you should do it when making projections. Interpretations will obviously vary, but I find that to be a damning discovery.

        Vote -1 Vote +1

      • PM says:

        Right, Nox. MGL has a real handle on reality. The same guy who went bipolar bc someone else read into platoon splits a tad too much, promptly proceeded to make the same mistake…twice — and then deleted the multiple comments calling him out:

        On Robertson:
        “David Robertson, according to my numbers is by far and away the best reliever on the Yankees and he has reverse platoon split (almost even) so he can pitch effectively (great, actually) against RH and LH batters.”

        On Rapada:
        “Rapada, you cannot bring him in if the opposing manager may pinch hit with a righty! Even a terrible RHB is going to hit like Pujols against Rapada.”

        Whatever credibility that guy had is quickly approaching zero.

        Vote -1 Vote +1

    • Really? He seems to primarily focus on insulting him personally over at The Book, which makes me think it’s all some personal beef and makes me gloss over the rest of the crap.

      “I also think you have a lot to learn about sabermetrics and logical and critical thinking in general. Maybe that will come with maturity (Dave is young, at least from my perspective).

      I doubt it because guys that are as stubborn, dogmatic, and defensive as you rarely learn much, because they never think they are wrong. You may say the same thing about me, but I am a lot older and I have learned a lot of these lessons. Plus, I have already forgot more about sabermetrics than you know…”


      Vote -1 Vote +1

      • PM says:

        Ha. Translation: “People who already think they know it all never learn new things. Except me. I’ve already learned everything.”

        That guy is so angry it’s sad.

        Vote -1 Vote +1

      • blox says:

        MGL’s a dick at times, but he’s absolutely right on the money with his criticisms of Dave here.

        Vote -1 Vote +1

  16. jc says:

    The only thing mgl’s posts, both here and at the book blog, have done are 1) demonstrated that he is being a total dick and 2) reinforced the dismissive and arrogant stereotype that saber people have been trying to lose. Both mgl and Dave are talented…. but mgl needs to stick to the baseball.

    Vote -1 Vote +1

    • Guy says:

      And Dave should probably stick with things he understands.

      As far as I can tell that list looks like:

      1) Being a Mariners homer

      Someone who doesn’t understand the fundamentals like Cameron has no place trying to educate the hordes.

      -6 Vote -1 Vote +1

  17. PL says:

    So O’Day does something that rarely works, but it works for him now, so why is this piece on fangraphs when it has zero to do with anything other than a freak of nature?

    Are you guys going to do a piece on all the results of balls hitting birds next?

    Vote -1 Vote +1

  18. austin says:

    Why all the hate for this article? So, a right-handed reliver has a small platoon split, about in line with other rightys. However, he is a sidearmer, which means he should have a huge platton split.
    He doesn’t. That was about the jist of dave’s article, and I don’t see anything wrong with any of those claims.

    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 ye@r *