How Much Extra Credit Should We Give a Young Lefty Starter?

Edwin Escobar, who came from the Giants in the Jake Peavy deal, debuted with the Red Socks on Tuesday. One inning of work is not enough to know much other than perhaps velocity, but that point alone started a discussion. He sat just under 92 mph, and once you correct for his appearance coming out of the bullpen, you might say he had average velocity. I even said this, on twitter. And Mike Newman responded:

I’ve heard this before. I’ve thought it maybe untrue, for whatever reason. So I decided to check out a few splits among starters this year.

Limiting the pool to only those that had started in more than half their appearances, I took a look at some simple splits:

Right-handed starters: 91.9 mph average fastball, 19.5% K, 7.0% BB
Left-handed starters: 90.6 mph average fastball, 20.3% K, 7.1% BB

So if Escobar could sit over 91 mph as a starter, his velocity would indeed be above-average for a lefty starter. Of course it’s an issue of supply, as only 30% of the sample was left-handed. That means fewer 95 mph guys, fewer Danny Salazars, fewer chances at big velocity.

But the higher strikeout rate for lefties gives us a little pause, no? Over the course of thousands of plate appearances, you might actually see an effect from a .8% difference in strikeout rate. It’s a decent sample, too, with 71 pitchers in the lefty side of the ledger. And although the effect was much smaller last year (.2%), lefties still struck out more batters than righties.

Even if the strikeout rate last year was the same for lefties and righties, you’d think that lefties should strike out fewer batters for a few reasons. For one, they cede the platoon advantage to more than 2/3 of the batting population, since switch hitters add to right-handers to form a formidable obstacle. And for two, if they have less velocity, they should have fewer strikeouts — we’ve related velocity to strikeouts here before.

It could be survivor bias. To make it as a lefty starter, maybe you have to do things to overcome the platoon advantage the batters have against you, and those things lead to more strikeouts than you might expect, given their velocities.

Could pitching mix be a key to the survival of the lower-velocity lefties? Looks like it. Here’s what lefty and righty starters are throwing this year:

RHP 35.0% 30.0% 14.4% 10.1% 10.5%
LHP 37.6% 23.5% 12.3% 10.1% 16.5%

Guess which pitches have the highest traditional platoon splits? Yeah, sinkers and sliders. Changeups have the best reverse platoon splits, which is definitely important to a lefty starter. Changeups also have the second-highest average swinging strike rate among regularly-thrown pitches (and four-seamers get more whiffs than two-seamers).

What does any of this mean? For one, you do have to give a lefty with similar velocity as a righty some extra credit for his handedness. There are fewer lefties, and we can see that there are therefore fewer high-velocity lefties.

And for another, you want to think about the platoon splits of a lefty’s pitching mix a bit more than you might with a righty. That lefty will face an opposite-handed batter more often than his right-handed counterpart, and in the past, at least, that’s meant it’s been important for the lefty to have a changeup. After all, since Wade Miley has been throwing his changeup more, not a single lefty remains on this list of two-pitch starters.

Print This Post

Graphs: Baseball, Roto, Beer, brats (OK, no graphs for that...yet), repeat. Follow him on Twitter @enosarris.

26 Responses to “How Much Extra Credit Should We Give a Young Lefty Starter?”

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

    One thing that’s been frequently cited for left-handed hitters struggling against southpaws is that they have far less chances to see good left-handed pitchers as an amateur. Possibly the same applies to right-handed hitters too to a certain extent.

    Vote -1 Vote +1

    • Jim Price says:

      I think this is true for younger players but by upper minors they must be seeing many more good LH pitchers, and guys in the majors for a while are definitely seeing an average of maybe 30% lefties. Hard to say why the K% is higher but its a thing.

      Vote -1 Vote +1

  2. Clayton Kershaw says:

    Maybe you should hand out my share of extra credit to Felix and Verlander. They need it more than I do.

    Vote -1 Vote +1

  3. Bill says:

    Loved reading this.

    Vote -1 Vote +1

  4. Stuck in a slump says:

    Looking back from 2010, the FIP-WAR leaderboards show that out of the top 10 SP’s, five are lefties.

    The RA9-WAR leaderboards have four of the top 10 being lefties, with Lester, Sale and Gio just barely being edged out.

    How much of a boost should we be giving these guys when almost half of the truly upper crust of pitching by any analysis are lefties?

    Vote -1 Vote +1

    • Eric R says:

      Career FIP fWAR leaders, active:
      Top 10: 4 are LHP [40%]
      #11-50: 9 [22.5%]

      Maybe just the elite LHP are already seeing a boost?

      Vote -1 Vote +1

    • Eno Sarris says:

      Fiddled with the title a bit, because I see this as more about evaluating prospects’ chances than about current ones.

      Vote -1 Vote +1

      • Stuck in a slump says:

        I totally agreed with the premise of your original title that lefties have a much harder time of things, and that a lefty hitting the upper 90’s is about the same as a righty who can hit 100+. I was just more curious if you’d give them a statistical boost for having to deal with platoon issues and not generally being able to harness the same velocity as their RHP counter parts, or if top pitchers are just top pitchers and the fact that many are left handed is more less noise than to do with them being lefties.

        Vote -1 Vote +1

  5. Jabronies says:

    I always wondered about this. Thanks!

    Vote -1 Vote +1

  6. Catoblepas says:

    It could be survivor bias. To make it as a lefty starter, maybe you have to do things to overcome the platoon advantage the batters have against you, and those things lead to more strikeouts than you might expect, given their velocities.
    Totally minor thing in the context of the piece, but I think this is a misunderstanding of survivor bias, and if it isn’t, I’m misunderstanding it and I’d love some clarification.
    If lefties as a whole are below-average, then the lefties who make it have to be above-average lefties. That doesn’t mean that they’d be above-average for the population as a whole — any pitcher who is good enough should make it, so it shouldn’t take drastic overcoming of the platoon advantage, just enough to be as good as the rest of the population. I don’t think survivor bias would explain the increased strikeouts then, no?

    Vote -1 Vote +1

    • A loaf of bread Mark Buehrle failed to dent says:

      You’re right. Lefties in the population as a whole would prbably have the same distribution of velocity as righties. Because there are more righties there are more extreme cases. As well, because lefty pitchers are more in demand relative to their share of the total population (about 30% of pitchers vs. 10% overall), a larger proportion of lefties make it to the majors. This means that the replacement level righty will be higher up the velocity bell curve than the replacement level lefty.
      Something (novelty, a bigger l vs. l than r vs. r platoon split, or whatever else) allows lefties to have marginally better results than righties with weaker stuff on average. Survivor bias doesn’t explain it, because a lefty is MORE likely to make the show than a righty out of the general population.

      Vote -1 Vote +1

      • vonstott says:

        “Something (novelty, a bigger l vs. l than r vs. r platoon split, or whatever else) allows lefties to have marginally better results than righties with weaker stuff on average.”

        I think there are two major contributors. 1 – Lefties are much less affected by the running game than righties, basically because lefties get to cheat:-) 2 – The percentage of lefties in a bullpen is much higher than in rotations and managers often manage their pen usage to maximize LvL opportunities.

        Vote -1 Vote +1

  7. Basil Ganglia says:

    It might be worth considering that because there are fewer positions for a left-hander to play, most left-handers with a strong arm are tried out as pitchers. So of the pool of players who have the talent to be a MLB pitcher, a greater proportion of the left-handers wind up in that role.

    Vote -1 Vote +1

  8. CardsFan says:

    Are there not a higher % of lefty ‘soft-tossers’ which would affect the mean values to a greater extent than the median?

    Vote -1 Vote +1

  9. JRM says:

    I’ve done a fair amount of work on the use of velocity in pitching projections, and fastball-throwing lefties can survive on significantly less velocity than fastball-throwing righties. I’m convinced the effect is real and robust, and that effect is part of why lefty pitchers are so much more common than the general population.

    Momma, don’t let your babies grow up to be righties.

    Vote -1 Vote +1

    • FIP'n good says:

      that is if you want them to pitch in baseball

      Vote -1 Vote +1

    • jfree says:

      There are more lefty pitchers rel to population than righty pitchers because they are funneled into body-type dependent specialist roles very early on. For a lefty kid, either you are a)the fastest kid on the team and thus CF or b)a big power bat with no glove necessary and thus 1B or COF or c)you are a pitcher or d)you are a future insurance salesman. Righties have any option available on the field so they can dabble at this and that more until their body type itself begins to cut off options. Even until after an MLB team drafts them and decides whether they will be a pitcher or position player.

      Ultimately, the effect is that LHP are far more likely to be normal/average body types (compared to the rest of the population). RHP are far more likely to be power-pitcher body types (tall, long arms/legs, etc). Which means they are non-comparable populations – with different strategies for ‘succeeding’. Most RHP (and the subset of LHP with power-pitcher body types) succeed as pitchers by ramping up the velocity. Everyone else has to succeed via command and/or secondary pitch control. Since everyone kind of knows at puberty what their body type is going to be – and since lefties have far fewer non-pitcher options – the result is that LHP tend to concentrate on command/control/secondary 3-5 years before RHP do.

      The dynamics of what is actually happening in pitching gets distorted/smothered by a different decision that occurs simultaneously – the decision by righties to learn to bat lefty. Righties never have to learn to do anything lefty outside baseball batting – so the overwhelming majority of people perceive all lefty batters as ‘lefties’ not as ‘righties who learn to do one thing lefty’. So the two very very different populations of LHP v LHB and the two non-overlapping populations of LHP v RHP create a giant analysis cluster#$%^ for anyone who is trying to analyze ‘lefty’ v ‘righty’.

      Vote -1 Vote +1

      • vonstott says:

        There are more lefty pitchers proportionally vs. the general population not primarily because they are funneled into anything but because they have built-in advantages over righty pitchers. Granted, there are certainly some RH throwers who could have become RHP but instead end up as hitters, but not nearly enough to explain that lefties are overrepresented by a factor of 3 – especially when one considers how many of the post-draft-decision guys and late switches ARE lefties.

        Vote -1 Vote +1

        • jfree says:

          This is not only irrelevant to the initial question – it is wrong as well.

          The initial question here – ‘why is there a velocity difference between lefties and righties’? – is entirely because of the funneling. There is no difference in throwing velocity if you look at the entire lefty/righty populations. Only in the subset of those who become pitchers.

          And v the majority of batters (righties), righthanded pitchers have the advantage over lefthanded pitchers. The main reason righty batters learn to bat lefty is precisely to overcome that and gain the advantage over righty pitchers. In doing so though they lose the oppositehanded advantage that they previously had over lefthanded pitchers. But that so many players choose to switch tells you that the career benefit greatly outweighs the cost. Thus – and only then – making it necessary for teams to look for lefty pitchers to claw back that advantage that the righty-turned-lefty batters seek to gain.

          If there was some different overall ‘advantage’ (unrelated to the decision by batters to switch from right to left) to lefthanded pitchers, then lefty batters would find it advantageous to bat righty (to take it away) and the universe of MLB players who have done so would be larger than – Rickey Henderson, Cleon Jones, Ryan Ludwick, Cody Ross and a dozen or so pre-deadball era guys.

          If OTOH it were instead just as easy for pitchers to control small muscles and switch pitch as it is for batters to control large muscles and switch hit – and there was some inherent advantage to pitching lefty, then every pitcher would switch pitch for those situations that don’t depend on oppositehanded batter advantage – and pitching gloves would be ambidextrous.

          Vote -1 Vote +1

  10. aariediger says:

    Couldn’t the “lefty strike” have something to do with this? A slider out of a lefty’s hand can be thrown in an almost impossible to hard-hit spot for a lefty batter, and still be called a strike. A righty facing a right handed batter can take that pitch for a ball, while a lefty swinging batter has to offer at it or take a strike. I think the “lefty strike” is why left handed pitchers have better strike out rates and part of why lefty batters have a larger platoon split on average than righties. Also playing into the platoon difference is an example of survivor bias, where righties with big platoon splits can’t hold down a full time job, while a lefty with a big platoon split can, because he only has to face lefties a 1/3 or so of the time.

    Vote -1 Vote +1

    • jfree says:

      IMO, most of the platoon split and K% difference of LHBvLHP compared to RHBvRHP is:

      1. virtually all natural LHB are power hitters (because they are limited to playing 1B or COF) and power hitters have more swing-and-miss.
      2. a large % (certainly more than 50%) of the LHB population is naturally righty. They’ve learned to bat lefty but with any learned handedness behavior there will always remain a brain conflict between the brain’s instinctive/genetic muscle control and the brain’s ‘learned’ muscle control. And it results in a permanent performance difference between the ‘naturally-handed’ and the ‘learned-handed’.

      Vote -1 Vote +1

      • vonstott says:

        The assignment of the labels LHB and RHB are totally arbitrary and functionally probably backwards. The bottom hand does way more than the top hand, so a RH person should (barring some ridiculous level of eye dominance that says otherwise) bat “LH”. The only drawback is that you have to be a good enough athlete to master weight shifts in opposite directions (right foot to left while throwing, left foot to right while hitting). But, for guys with potential ML ability, this shouldn’t be a big deal.

        Think of the LHB who get the “OMG, his swing is so pretty I’m actually aroused” label. Carew, Brett, Boggs, Mauer… You’re really claiming that those guys were fighting an awkward swing their whole career?

        Vote -1 Vote +1

        • jfree says:

          Yes – they had to train their muscles and fight their brain’s instinctive way of controlling their muscles. It obviously can be done. It’s easier to do with the large muscle stuff (batting, operating machinery, driving). Much more difficult to do with small muscle stuff (throwing accurately, writing, shooting a rifle). And no, they weren’t fighting their brain their whole career. They were fighting their brain WHILE they were learning to swing opposite-handed. Since most righties never have to learn to do anything lefty, they don’t have the slightest ability to comprehend that specific training issue. And since batting is not a dangerous activity where ‘instincts’ ever take over, it is difficult to see the internal ‘fight’ with the brain.

          But it’s there. Easier to see with something like driving. Lefties obviously learn how to drive the same cars as righties. And 99.9% of the time there is no ‘instinct’ difference and no one is conscious of anything ‘handed’. But in a panic situation, the lefty brain pulls the left side muscles faster – and the car veers left (into oncoming traffic in the US). Same situation, the righty brain pulls the right side muscles faster – and the car veers right (off the road).

          The ‘performance difference’ (overall driving performance) between lefties and righties results in higher (roughly double) car accident fatalities for lefties v righties in the US. The reverse would be true in Britain but they have a godawful difficult driving test that is both intended to fail half the people and intended to specifically test on recognizing/reducing those ‘panic’ situations.

          Vote -1 Vote +1

      • Payroll says:

        I think there’s probably some merit to #1

        My other thought would be that teams mistakenly matchup RHBs against changeup-heavy lefthanders, when they should instead be playing their LHBs (as the NL East finally did to Ricky Romero)

        Vote -1 Vote +1