I become obsessed with things. Sometimes it can be a particular food or an album (this Jimi Hendrix “vault cleaning” as Rolling Stone describes currently has my ear). Other times, as my mind is often tuned to baseball topics, I incessantly think about a concept from the diamond or evolution of a new statistic. Pitch tunneling is the recent topic earning a spot in my head.

This isn’t the first time I’ve gravitated towards pitch tunneling. Last year I wrote a column about Dylan Bundy’s cutter-slider, it’s usage, and why that pitch is one reason I irrationally like his volatile arm. As I’ve rekindled my interest in the concept, it was time for a refresh after Baseball Prospectus’ recent update. My motive was simple: combining what we know about a pitcher and what we can learn from tunneling might provide us with reasons for optimism.

I’ll admit, this post might get a little bit convoluted, so if you’re not in the mood to try and understand pitch tunneling and determine how much you value it, feel free to hop to one of my last three Razz articles – there’s something for everybody (On Scott Kingery; On ADP discrepancy; On Michael Wacha). Or just skip down to the heading for Patrick Corbin. I’ll try my best to keep things as simple and concise as possible. Teaching a concept is often a great form of learning, so I’ll admit that writing this post, in a way, helps me to understand the topic and its associated statistics better.

Primer

Tunneling as a whole is built around the concept of a tunnel point. This is the point at which the decision to swing must be made. Anything decision made late and your chances of catching up to a pitch are zero. In theory, if a pitch moves substantially after this point, then the chances one makes contact, of any quality, decrease. This occurs because the human eye doesn’t actually track the ball the whole way to the bat when one is hitting – the ball is moving too fast. Therefore, a batter must make a decision to swing at a time when the final point of the ball is unknown, implying there is some prediction of where the ball will be, and thus, where one must swing. More movement after the tunnel point, as far as we know, is a precursor for more whiffs. In layman’s terms, this is essentially “late life” on a pitch.

Stats for pitch tunneling can be broken down at the pitcher level and the individual pitch combination level, each of which have their own merits. Most of the time, you’ll see photos of a pitch visualizer (which we’ll see below) showing all of a pitcher’s pitches and their trajectory relative to one another. This gives you a great idea of how a pitcher tunnels all of his pitches together.

You can also look at two pitches together in sequence and learn how they interact and the results they produce. Ideally, a pitcher wants as little difference as possible at the tunnel point between two pitches that end up at two different locations at home plate. Think about all the gifs you see about devastating changeups; each of those are considered “good” because they look like a fastball until the last second. I like focusing on pitch combinations because most of the time, to one handedness of hitter, a pitcher has a pair that makes up 60-75 percent of what a hitter will see in the aggregate. While I won’t differentiate for sequencing – whether a pitcher goes fastball-changeup or changeup-fastball – we can still gain some information by focusing combination with greater frequency: the offspeed pitch coming after the fastball.

For this dive, the two pitch combinations I looked at were fastball-slider and fastball-changeup. (Curveball tunneling is still a concept I want to refine my knowledge on.)

Here are the tunneling statistics I find most applicable for analyzing pitchers. I’m renaming two of them as I think the current names given confuse more than they inform. These metrics are discussed at length by Baseball Prospectus and the full table of their stats, in all their glory, can be found here.

  • Movement ratio – what percentage of a pitcher’s difference in plate location between two pitches comes from movement after the tunnel point.
    • Average is 11.9. Higher is presumed to be better.
  • Tunnel distance – max distance between back-to-back pitches at the tunnel point.
    • Average is 1.54 inches. Lower is presumed to be better.
  • Release distance – how far apart two pitches are upon release.
    • Average is 2.6 inches. Lower is presumed to be better.

Now, let’s back our way into two interesting arms using our basic knowledge of pitching tunneling. It may seem excessive to detail all this and only cover two pitchers, but I hope to use this column as a reference for further discussion about pitch tunneling and how it relates to pitcher analysis for fantasy baseball.

Patrick Corbin

Corbin is quietly coming off a decent season, posting a .408 FIP over 190 innings in a hitters ballpark. He slotted in as a top-200 player, just falling inside SP4 territory, continuing his ability to neutralize left-handed bats with a 3.18 FIP against and sub-.220 average against. We know he can stymie lefties, but the missing piece, as it is for so many southpaws, is finding a way to beat right-handed hitters.

One of the talents Corbin has is his ability to create late movement when he sequences his four-seamer and slider to right-handed hitters. This specific combination has the second highest movement ratio (definition above) among pitchers with 75 or more sequences of a four-seamer and slider (17.6). He sits slightly above pitchers like Justin Verlander and Jacob deGrom on the metric, but what I noticed is that while this fastball-slider combo is his dominant sequence to right-handed hitters (174 instances), he opts for a sinker-slider combination a decent amount of the time (125 instances). This stood out because the disclosed tunneling metrics – including this movement ratio – are all better when he sequences his slider with his four-seamer rather than his sinker. Given this slider is clearly Corbin’s best pitch, intuition would suggest he would want to set up that pitch to make it even better.

Corbin’s used his sinker more than any pitch against right-handed hitters last season, yet this sequencing of his slider with his four-seamer seems to suggest there is some understanding that it’s better to set his slider up with his four-seamer. While I’m not a pitching coach willing to suggest Corbin should use his four-seamer more to right-handed hitters, the application of this to fantasy would be keeping an eye on Corbin early and seeing if there is any increased usage of his four-seamer to righties. If so, I wouldn’t be stunned to see his slider improve, his strikeouts increase to above 9 K/9 and for his one weakness – right-handed hitters – to be relatively neutralized. Given his four-seamer has been more successful than his sinker against right-handed hitters and it seems to set up his slider better, there may be a path to him turning around his struggles.

Corbin is currently going 227th overall in NFBC leagues, the 87th pitcher off the bard (relievers included). We at Razzball have him right in that same window, but I’m tempted to jump on him slightly earlier to fill out my staff given what I’m seeing here. Here’s a decent look at his Corbin’s slider, with a suggestion from fellow fantasy writer Van Lee (Fantrax) related to the one factor I didn’t touch on: Corbin’s slider was better outside of Chase Field and we all know about the humidor.

Jon Lester

If you’re using Rudy’s WAR room – and you should be – then you’ll see Lester popping up as a good value in leagues around 100 overall. We at Razzball have him ranked right in the ninth round, 17 spots ahead of where NFBC leagues are seeing him leave boards. Despite an underwhelming performance last season his peripherals look pretty stable and if you can get past his age, there is some value to be had. This is especially true if you believe in his ability to tunnel pitches and repeat his delivery exceptionally well.

Lester stands out in tunneling metrics when you look at his aggregate pitch tunneling among all his pitches to hitters (as opposed to what we looked at with one combination for Corbin above). Among pitchers with more than 1,000 pitches, Lester has the lowest difference in release distance compared to 161 other pitchers (see above) at only 1.54 inches. Said frankly, everything Lester throws comes out of his hand from a similar spot more than any other pitcher in baseball. This nearly eliminates the ability for hitters to pick up what Lester might be throwing upon his release, leaving them at the mercy of other queues to pick up on in a shorter amount of time.

Below is a screenshot of Lester’s pitches only to right-handed hitters. A really good visual of tunneling is looking at Lester’s ability to make multiple pitches come out of a centralized point (follow this link to play around with the tool I am using to generate these shots). While you might suggest it’s easy to pick out which pitch is which  based on its intended direction, understand that you’re doing so with a still frame. The time between separation of those pitches and the decision to swing is extremely small. Lester does an amazing job of taking away any advantage a hitter might have before this point.

Four-seamer (red), Cutter (brown), Curveball (blue), Sinker (orange), Changeup (green)

Here is another picture of the above shot from a different angle.

This angle gives you a better idea of what things look like out of Lester’s hand from the eye level of a hitter as opposed to putting a GoPro in a catcher’s mitt, like the image above. While Lester’s curveball stands out, it’s the meshing of his other four pitches which seamlessly blend together.

To give you an idea of what an average pitcher looks like from this perspective, let’s actually look at Corbin’s pitches in the aggregate, who as a whole – removing his fastball-slider tunneling – is an average “tunneler.”

Comparing the two pictures, it’s easy to see how each pitch’s average path is distinguishable, whereas with Lester, the deviation happens much later.

One thing we know about Lester is he’s one of a small crop of starters projected for under a 4.00 ERA by Steamer, even with the context of his age. I also want to point out that in the age of believing innings are everything, Lester has thrown 180 or more innings every year since 2008 – ten years in a row. The polish Lester has on his mechanics is better than any pitcher in baseball, remaining an elusive reason for this prolonged durability. That repeatability translates into tunneling, creating what I believe to be a fantastic floor, even for a 34-year-old. If you’re considering drafting Madison Bumgarner at his current price tag of 27th overall, I would suggest a long, hard look at waiting roughly 100 picks and grabbing Lester instead. Tunneling is just one reason

 

Tunnel your way to Twitter and follow me.

@LanceBrozdow

BigThreeSports is where my personal fantasy rankings will be housed… but really just listen to Rudy’s…

   
  1. Scrapski says:
    (link)

    Interesting take Lance, thanks for the info!
    Would think Lester’s tunneling pattern looks different when throwing to first base!

    • Lance

      Lance says:
      (link)

      @Scrapski:
      Thanks!

      I bet that’s the only tunneling Lester is bad at haha

  2. Ian says:
    (link)

    This is fantastic. Can’t wait for more in the series. You made it so a dumdum like me understands.

    • Lance

      Lance says:
      (link)

      @Ian:
      Thanks, Ian.

      Hope this opens up a new world for people.

      I’m sure I’ll be talking about it more this season.

  3. nick the smooth dick says:
    (link)

    I love when you talk confusing to me

    • Grey

      Grey says:
      (link)

      Haha

    • Lance

      Lance says:
      (link)

      @nick the smooth dick:
      You got an audible laugh out of me here, well done.

      • nick the smooth dick says:
        (link)

        @Lance: Ha interesting stuff here tho buddy. Nice article.

  4. OaktownSteve says:
    (link)

    Interesting. The visualization tool is pretty cool. This puts a different lens on the whole concept of pitch sequencing, which is usually thought of in terms more like game theory where the end result is to throw a pitch the hitter isn’t anticipating before the delivery even begins. This raises the possibility that a certain two pitch sequence may be more successful due to the way these pitches are cognitively processed during flight even if that sequence is more likely to be anticipated. Of course it would seem there would be a tipping point where if a particular sequence was used so frequently that the tunneling benefits would be negated or inverted by the predictability.

    • Jbona3 says:
      (link)

      @OaktownSteve: Oaktown, you’re right on about how the pitches are processed because to Lance’s point above, the batter is basically swinging where they expect the ball to be. The idea about over-use leading to predictability is a good one, but I wonder if a pitcher goes FB/Slider so much it becomes predictable, does reversing it and going Slider/FB negate that? So, as long as the pitcher is playing off that expectation, it could negate it.

      • OaktownSteve says:
        (link)

        @Jbona3:

        Game theory in a nutshell. I wonder if any MLB teams utilize game theory math vis sequencing. I’ve always though great game calling catchers are like natural poker players. But if you brought more math a la Jesus Ferguson maybe you could gain an edge.

        • Lance

          Lance says:
          (link)

          @OaktownSteve:
          Another great point!

          I want you to comment on all my columns!!! hahaha

          On game calling, I would love to pick a ton of brains in and around the game about what makes a good game caller. I would be very interested in any underlying themes, points that come up, below the surface-level jargon.

      • Lance

        Lance says:
        (link)

        @Jbona3:
        This would be a fun point to look at. Going to put it in the back of my mind and see if there is any way to test/look at it.

    • Lance

      Lance says:
      (link)

      @OaktownSteve:
      I’ve really enjoyed the visualization tool, it’s a ton of fun to play around with.

      I think you actually make a fantastic point here when you say, “This raises the possibility that a certain two pitch sequence may be more successful due to the way these pitches are cognitively processed during flight even if that sequence is more likely to be anticipated.”

      This is the bridge between nerding out with tunneling (guilty) and talking more to the general fan about how to “fool” a hitter.

      Often, I feel like we hear that when pitchers are their best, the hitter can know what’s coming and still not hit it. This ties back, in some ways I think, to your above points.

      You would think the latter as well in your last point, but just playing devil’s advocate, I noticed a lot of relievers with heavy usage of two pitch sequences. I wonder if this is more a matter of not seeing a guy enough time to get a sample as a hitter and be able to adapt.

      Even if you’re a Giants’ bat, you may only see Jansen a handful of times.

      Thanks for reading!

  5. bobby says:
    (link)

    Yikes I drafted bumgarner as my ace of my staff, I’m doomed right? My SP’s are

    SP: bumgarner
    SP: keuchel
    SP: Yu Darvish
    SP: Jose berrios
    SP: lance mccullers
    SP: Jon gray

    Bench I have Ohtani(SP/DH), Jameson taillon

    Available SP’s: Blake snell, tanner Roark, pomeranz, Lucas giolito, Patrick Corbin , Sean mannea.

    I plan on dropping an extra bat for one of those

    • Lance

      Lance says:
      (link)

      @bobby:
      Not doomed! I just think he’s a little bit overvalued. Very easily I could be wrong or the rest of your team can be so good it won’t matter.

      What’d you pay for him ADP-wise?

      SP depth seems insane to me if you have those guys on your waiver, you’re stacked, is this a 8-10 teamer or something?

      • bobby says:
        (link)

        @Lance:

        I got mad bum round 4 at 30th overall. I was looking at sale but he went to spots before mad bum. 8 team league with bunch of buddies

        • Lance

          Lance says:
          (link)

          @bobby:
          Yeah I have Bum around 50th overall.

          But I wouldn’t be too worried. In an 8-teamer there is so much depth, I’m fine with trying to take shots on a guy like that rebounding because there will be a ton of the waivers.

  6. Jbona3 says:
    (link)

    Lance, are you at all worried about Lester losing a MPH on his pitches last year? I think he represents great value, but the drop in velo does scare me.

    • Lance

      Lance says:
      (link)

      @Jbona3:
      I don’t love it, but I think it’s natural deterioration.

      If there is anybody who can buck this, it’s Lester. He was never high velo to start, just has to rework his arsenal a bit.

      When guys lose velo, it’s often most concerning, to me, because it affects his OTHER pitches. I think we often go crazy over a FB dropping 1-2 mph, but I always look first to how that will affect a pitcher’s changeup. If it’s concerning enough, then you’re effectively eliminating a pitch, and sometimes that 3rd pitch is the only reason a guy can neutralize his natural platoon split (LHP v RHH, RHP v LHH).

      Think about if you take away Severino’s changeup to LHH? He MIGHT be ok, but I’d be thoroughly concerned.

      Bit of a rant there, but hopefully you get my point.

      • jbona3 says:
        (link)

        @Lance: I do and it makes sense…my bigger concern seems aligned with yours, in that his velo is important with respect to the gap between his FB and off speed stuff. Because Lester was never a pure power pitcher.

        • Lance

          Lance says:
          (link)

          @jbona3:
          Yeah, but I don’t expect his velo to drop again honestly, and he survived with 91mph instead of 92mph.

          So overall I’d say I’m not as worried on this particular point with Lester.

  7. Justin says:
    (link)

    I guess I’d like to know how tunneling has predicted success in the past. is this a valid statistic to use to see if a pitcher is under/overvalued?

    • Lance

      Lance says:
      (link)

      @Justin:
      I think it’s a great addition to analyzing pitchers.

      I’m not endorsing the use of this as an end-all-be-all statistic.

      Always look at things in context of other things!

    • OaktownSteve says:
      (link)

      @Justin:

      I thought the same thing. Spin rate came to mind as a new measure that proved predictive. McHugh is the poster child.

      The idea of tunneling is analogous to spin rate in that it describes the characteristics of an individual pitch. The idea of correlating sequencing to tunneling might be just as easily applied to whether pitches with different spin rates might be optimally sequenced. Really there could be lots of inputs to a game theory driven calc.

      • OaktownSteve says:
        (link)

        @OaktownSteve: Also if one could demonstrate and reasonable level of predictive power for tunnel sequence it would be a career maker in this business. Eno Saris got some traction with release point analysis. He was out in front of launch angle a bit as well.

        • Lance

          Lance says:
          (link)

          @OaktownSteve:
          Echo everything Oaktown is saying too.

      • Lance

        Lance says:
        (link)

        @OaktownSteve:
        Realizing now that I didn’t really answer Justin’s question haha

        From my understanding, and what we know, there is a correlation between two pitches that are tunneled well (similar at tunnel point, large difference at plate), and the swinging-strike rate.

        The below article is pretty solid, although it is an old study and I’m sure there is something more recent that gives you a better nature.

        In terms of predictiveness, I’m not particularly sure unfortunately, but I would anticipate it helping with some statistics, particularly on the strikeout quality side of things.

        https://www.fangraphs.com/tht/the-effects-of-pitch-sequencing/

  8. Jose says:
    (link)

    Lance can I have your advice? I have 8 SP and I want to drop 2 of this 3… Which ones?

    Jameson taillon
    Michael wacha
    Miles mikolas

    • Lance

      Lance says:
      (link)

      @Jose:
      Mikolas easily.

      I prefer Wacha over Taillon, closer there.

      • Jose says:
        (link)

        @Lance: what about Charlie Morton?

        • Lance

          Lance says:
          (link)

          @Jose:
          I have them ranked Wacha-Morton-Taillon. All in the same tier.

  9. True and Correct says:
    (link)

    Go Miles Go !

Comments are closed.