Updated: Feb 8, 2018
Good morning and welcome to Pitch Tunneling 101. While "pitch tunneling" seems like it should be easy enough to understand, it can actually be quite complicated, as is the case with many baseball-related concepts. In turn, complicated baseball concepts often lead to common misconceptions. Well, with the conclusion of "Pitching Week" over at Baseball Prospectus (hosted by Jeff Long, Harry Pavlidis, among others), I figured now is the perfect time for a 100-level primer on one of BP's central topics from last week -- pitch tunneling -- in hopes of making the concept easier to understand, overall.
Keeping in mind the existence of both auditory and visual learners, I will provide my usual commentary and analysis on the topic, followed by the trailblazing .gifs created by none other than @cardinalsgifs. Considering my goal of keeping this as a 100-level (okay, maybe 200-level) post, as opposed to the many, already-published 300- to 400-level posts, I will admittedly oversimplify some concepts. Because after all, one must first understand the basics before grasping the complexities.
So...what exactly is pitch tunneling?
The great Warren Spahn is credited in saying, "Hitting is timing. Pitching is upsetting timing." Spahn played from 1942 through 1965, so it is clear that pitch sequencing and tunneling are by no means new topics. However, with numerous technological advancements (namely PitchF/x, TrackMan, Statcast, and high-definition/slow-motion video), the two have jumped to the forefront of baseball analysis.
In the most simplest terms, each pitch, during its flight home, yields a respective “decision-making point” for the batter. When you take into account release point, velocity, location, movement, and batter handedness, the decision-making point varies with each pitch. An up-and-in fastball will present a decision-making point earlier than a down-and-away changeup. Before even taking pitch location into account, this makes perfect sense using velocity alone (as a pitch traveling at 95 MPH is moving faster than one at 85 MPH).
By its very definition, hitters must conclude at the "decision-making point" whether or not they are going to swing the bat. Now, individually, pitch tunneling does not exist. Well, considering the amount of high-definition video that goes into game preparation for hitters (and the fact that these same hitters take that prep. into their at bats), I guess it does exist, to an extent. But within a given plate appearance, two pitches must be thrown before tunneling truly comes into play. Thus, once two pitches are thrown, did pitch two look similar to pitch one, out of the hand? Further along (i.e. at the tunnel point), at the the hitter's "decision-making point," how close did pitch two look to pitch one?
Okay, so "decision-making point" and "tunnel point" ... what exactly are those, you ask? First and foremost, Baseball Prospectus dives way deeper into this. And with BP's latest update, the terms are now used interchangeably. In the most general terms, the decision-making point is a given distance (based on time and the hitter's physical ability to track a pitch with his eyes) between the pitcher's mound and home plate. Admittedly, it's tough to explain, but think of it as the viewing pane of a hitter, as he tracks a pitch out of the pitcher's hand all the way to his bat.
While a tunnel-focused pitcher aims to match up decision-making points on each of his pitches, this is, by no means, an easy task. Yet, when this task is achieved, pitchers can expect success. The best part about tunneling effectively is the pitches' final locations can still be very different despite flying through the same tunnel, possessing nearly identical decision-making points.
For example (since that last sentence was probably closer to 200-level than 100-level), a right-handed-thrown changeup and a right-handed-thrown slider can very easily share a decision-making point and tunnel, and yet, due to opposite horizontal movements (arm-side for changeup, glove-side for changeup), these two pitches will ultimately land in very different zones, and as extreme as opposite sides of the plate.
The "decision-making point," metaphorically-speaking
To better understand the concept of the "decision-making point," visualize yourself sitting in a turn lane on your way home from work -- a route you've taken it thousands of times over, but is still never exactly the same due to various environment factors. There are two lanes of opposing traffic between the turn lane and your neighborhood. In order to make the turn into your neighborhood -- without crashing -- you must make the judgment that neither lane is currently occupied (or will be occupied) by oncoming cars.
Before making the decision to turn (again, with the end goal being not crashing), you must first adequately evaluate the type of car (pitch type), projected speed of said car (pitch velocity), which lane the car is currently in (pitch movement), and the possibility that this car will change lanes (pitch movement). At some point, you will have to decide to begin your turn because it's rush hour -- duh, you're coming home from work -- and it may be minutes before you get another opportunity.
Unfortunately, our decision-making in driving isn't always perfect. Sometimes, cars appear to be traveling slower than they actually are. Some drivers forget to turn their lights on at night. Or, there exists a scenario in which an anxious driver -- driving a small car -- speeds around a large car after you already began your turn, leading to an accident. Think of this situation as a perfect execution of "pitch tunneling." When you made the decision to turn, all appeared clear in the path to your neighborhood, but after committing, the small car -- which was not visible to you at your decision-making point -- ended up getting in the way. While this is by no means a perfect metaphor, if it helps just one person grasp the concept, I'll consider it a success.
Addressing some misconceptions
1) It's all about release point.
While it certainly helps to possess consistent release points across all pitches, movement, velocity, and gravity make it (nearly) impossible for identical release points, particularly with breaking balls. Essentially, as long as release points are not obviously distinguishable by the hitter, the pitch's location at the "decision-making point" is considerably more important.
2) Tunneling is easy, important, and necessary
Not all pitchers can effectively pitch tunnel, and in some cases, that's okay! For instance, sinkers generally don't tunnel well with other pitches. Given the amount of vertical movement, 12-6 curveballs don't tunnel all that well, either.
As pointed out in Pegasus' Viva El Birdos post on the topic -- which is an absolute must-read, by the way -- Adam Wainwright is not good at pitch tunneling. Yet, Wainwright will still go down as one of the best pitchers to ever don the Birds on the Bat. If he is bad at tunneling, how is this the case?
Frankly, while tunneling is "all the rage" right now, you don't have to tunnel in order to be successful. While this data has just now become available (and there is so much to learn, still), Waino was likely never particularly good at pitch tunneling, largely because he simply does not have a repertoire compatible with consistent tunneling.
Peak Wainwright painted corners with his fastball and struck out hitters with his curveball. While fastballs and curveballs rarely fly through the same tunnel, they still complement each other quite well due to a virtually identical spin profile from the hitter's point of view. Waino serves as just one example -- of almost certainly, many -- that tunneling isn't necessarily a must when it comes to success.
Sequencing with a perfect release point
As I have already documented in previous posts (part 1, part 2), new Cardinals reliever Dominic Leone is terrific at replicating his release points. Yet, despite an identical -- yes, identical -- release point, these two pitches don't tunnel particularly well, as you can see in the .gif above and the image below (via Baseball Prospectus). Sure, Leone was able to induce whiffs on both pitches, but this is due to his successful elevation of fastballs above Davidson's swing path, and not necessarily pitch tunneling.
Meeting up at the decision-making point
Unlike the Leone sequence, Martinez's release points are not identical on the two pitches. And that's okay because as stated above, they are still "close enough" to not present a distinguishable difference to Josh Bell. The end result is a nearly perfect example of pitch tunneling. As you will see below (again, via Baseball Prospectus), when Bell has to make his decision on whether or not to swing at these pitches, the decision-making point is nearly identical (look at the black shapes, circled in green, on along the dotted path prior to their diverging).
And in .gif form, with the yellow trail representing the overlap between the two pitches:
Finally, one of my all-time favorite tunnel sequences, as thrown by Jack Flaherty:
Look at just how close the decision-making points (again, look at the black shapes, circled in green, along the respective paths) are on pitches one and two. While pitch three produces a distinguishably different decision-making point, Flaherty sped Anthony Rizzo up with the up-and-in fastball, leading to late, jammed-shot contact.
I hope this post can serve as a foundation for those looking to learn more about pitch tunneling. As the season goes along, I fully intend on taking a deeper dive into this data. But prior to doing so, I felt the need to produce a primer, for my reference and yours. I simply cannot wait to play around with these new toys made available by Baseball Prospectus. Here's to hoping it can take my pitch analysis posts to the next level.
Questions? Comments? Concerns? Please leave them below, and I will be more than happy to discuss!