When you think about a good fastball, it is tempting to think of it as simply a hard fastball. 100 MPH is better than 93 MPH right? That is partially true and most certainly, mistakes are harder to punish when you have that fraction of a second less to identify it for what it is. However, baseball’s simplest pitch actually contains some unseen layers that can make or break it in terms of quality. Tools like Rapsodo and Trackman allow us to both appreciate the hidden nuances of a fastball and understand the emerging concept of pitch design. For simplicity, I will focus most on the standard four-seamer.
Shape
When we talk about fastball shape, we’re really talking about how it moves as it makes its way towards the catcher’s glove. In MLB, there is this concept of the “dead zone” which numerically is a fastball that has about 13.5 inches of both vertical and horizontal break. Of course, it is more formidable when thrown harder than when thrown soft, but 99 MPH in the dead zone” plays less in today’s MLB than 94 MPH with break (be it horizontal or vertical) that deviates significantly from this baseline. It matters less whether the fastball is of a rising variety or if it has extreme sink. What truly constitutes great fastball shape is it not behaving like a “normal” fastball and therefore fooling hitters.
An even more extreme example of how fastball shape can help compensate for less velocity is Marco Estrada during his heyday in Toronto. With a below average 88 MPH heater, he survived because his fastball moved in ways that fooled hitters and put them behind in counts, putting them at the mercy of his elite changeup. The significantly above vertical break and armside run induced both whiffs and weak quality contact, courtesy weak fly balls thanks to the nice carry on the pitch. You can see this pictorially thanks to this great primer from Fangraphs on the concept of shape. In the pre-analytics era, Sandy Koufax was known for his fastball carry that made it seem like it was rising. This combined with his hammer curve and overhand arm-slot that seemed to make the fastball carry defy physics made him into one of the greats. Tyler Rogers survives with mid 80’s stuff today like Estrada did but through making the fastball sink like nobody else does in today’s MLB. It makes sense given his delivery. Conversely, Sixto Sanchez throws upper 90’s but because his fastball follows a predictable path, hitters' eyes see it as ‘flat’ and if they can time it up, they can square it up. No to mention the pitch appears slower. It is worth noting that Max Scherzer these days has a fairly ‘flat’ fastball, but his superior command, ability to plan for each hitter and quality secondary pitches have mitigated this.
Vertical Approach Angle
Vertical Approach Angle (VAA) is a concept that’s becoming increasingly prevalent in modern discourse around developing pitchers. Some of the best have noticeably lower VAA ratings. What is VAA? Well, it is meant to quantify at what angle the ball approaches home plate vertically speaking. It is better explained in pictures so here’s a good video to help visualize the concept. I want to point out that VAA is not especially illuminating in isolation. Context is what makes a new favorite tool of pitching coaches and scouts. Context involves things like pitch type, area of the strike zone in question etc. Since this is an article about fastballs we’ll zero in on these. Typically, guys with lower VAA tend to see noticeably more success with fastballs, especially when throwing it up in the zone. Why? It correlates with better carry, or induced vertical break. (the perception of the fastball rising) It is even better with guys whose deliveries would make you think that the ball would sink more. A flat VAA can even make up for a lack of movement when the fastball is well placed. 2022 Alek Manoah was a great example of this because in spite of his size, he achieved a 92nd percentile VAA of roughly 4.3 degrees which would be unexpected given his size. Harder throwers like Josh Hader who also generate flat VAA are even nastier, although Hader’s arm slot makes this more expectable. Another guy whose flat VAA is a very powerful tool, especially paired with his nice fastball movement in general, is Yoshinobu Yamamoto. Apart from the easiest 95-97 MPH you’ll ever see, he has a well below average release height (even for his size) which induces a VAA of around 4.1 degrees on average. Same goes for fellow Japanese Shota Imanaga who rode that to the best stuff+ out of all fastballs at the WBC.
Spin Rate And Spin Axis
Spin rate became a major topic of conversation a few years ago when the question of sticky substances came up with the subsequent airport style checks umpires perform these days. We have learned that higher spin rates tend to correlate very strongly with higher velocity and that this plays particularly well when throwing up in the zone. Rather than simply working on the idea spin rate though, we want to better understand the relationship between spin rate and velocity. When you divide spin rate by velocity you get a ratio referred to as a Bauer Unit. (yes, named after Trevor) A BU value of below 22.5 would be low spin relative to velocity, 24 is considered average and anything above 25.5 would be fastballs that were “acting like a bunch of Coles”, to put it in Bauer’s words. In his brilliant 2020 campaign, Trevor Bauer registered an average BU rating of 29.7 which is extremely high. As expected, he put up not just the highest spin rate, but also the highest spin rate relative to velocity. Naturally, the carry induced by such a high RPM made Bauer extremely effective up in the zone, giving him tremendous whiff numbers when working on the upper half of the plate, even though his 93 MPH average fastball is not exactly overpowering relative to an Aroldis Chapman or a Jacob DeGrom. Going back and looking at video from that year, you can see that Bauer’s fastball had great shape, making it a huge weapon that led to a Cy Young Award.
That being said though, a low BU rating is not a death sentence. Kyle Hendricks routinely ranks near the bottom percentile in fastball spin rate and BU rating, ie: his fastball spins little even relative to his velocity which is far inferior to a Bauer fastball. Even so, he has managed to have success by working effectively on the lower half of the zone, by mixing his varied arsenal and through great control. Fastball spin behavior on its own does not predict success, but it does hint at what the best plan of attack is for different pitchers. I will also add a note that BUs are far less useful for analyzing breaking balls, given that they vary so much and that different pitchers throw different types of breaking balls.
Now onto the idea of spin axis. Imagine a clock with your standard x-y axis superimposed. Now imagine a spot denoting where the ball was caught. That is called spin axis, also known as spin direction or tilt, should you choose to measure instead in degrees. Another term I’ve seen thrown around in this space is ‘gyro effect’. There are handy tables these days also to convert from spin direction to tilt and back as needed. Spin axis, put simply, serves to quantify observations like “he has late life on his fastball” or “his fastball has a little arm-side run”. I will not go into breaking balls, that’s beyond the scope of this article. With fastballs, you can expect the spin axis for right-handers to generally be from the 12-3 o’clock range, depending on arm slot, grip, wrist angle, seam orientation etc. For lefties, it would most likely be in the 9-12 o’clock range.
I will spare you the physics of this whole concept as much as I can without confusing things, but if you want to delve into things like spin efficiency for example, you may wish to brush up on some trigonometry. A good understanding on how to work with vectors would also help if you want to dive deeper into the physics of the matter as we need to consider perspectives, how the ball moves along the x-z and x-y planes etc. If you are interested in the physics of pitching, or baseball in general, here is a link to get you started.
What matters from an evaluation perspective is how we manipulate spin direction to give hitters what they do not expect. Well, a typical fastball will generally hit in the 1 o’clock region for RHP and the 11 o’clock region for LHP. What coaches would look to do is tinker with the usual things like grip, arm slot and seam orientation to try and generate different behavior from the norm. Smart pitchers will also know how. Again using Bauer as an example, his fastball has generally been around the 1:30 region, which is well above average arm-side run. This combined with that great induced vertical break that makes it ‘rise’ makes it so effective up in the zone when he’s at his best. Yu Darvish poses an interesting challenge with his four-seamer also. His spin direction can vary very widely, ranging from 12:30-1:30 as he tinkers with grips and so on mid-game. As a result, his four seamer is really different depending on what he wants against a given hitter. He can press harder or softer on the seams depending on how much run and IVB (induced vertical break) he wants at a given moment.
Induced Vertical Break
That term I just used is another one that’s come into vogue recently. IVB is meant to quantify how much a pitch breaks upward from what is ‘normal’ or expected. Guys who generate a high IVB are the guys who have a ‘rising’ fastball or one that carries well. These types of pitchers are the ones like Bauer or Houston’s Cristian Javier who have generally done very well pitching up in the zone, generating high whiff and high K rates on fastballs on the upper half of the zone. They are not necessarily overpowering either. Javier generally operates in the 93-95 MPH range which is roughly league average these days but he routinely ranks near the top in IVB and in 2022, his fastball was in the 97th percentile in terms of run value, meaning only 3% of pitchers had a better fastball, even those with “better stuff”. He still had above average IVB in 2023, but it was less than in 2022 and that along with a slight drop in velocity made him noticeably more hittable, especially after the All-Star break. Another pitcher who’s got great ride on his fastball is Yoshinobu Yamamoto. His 17-18 inch IVB in NPB is no joke and combined with his low arm slot and VAA (courtesy in part of his smaller stature) it is a nightmare for hitters in spite of it not being the triple digit weapon of Roki Sasaki or Shohei Ohtani. He also threw it for strikes around 68% of the time last year in NPB.
Now, a very good question is how the same IVB would play with a 93 MPH fastball versus a 98 MPH fastball. That is where we get IVB’s answer to the Bauer Unit. As noted in this great Eli Ben Porat article, it matters how much time a fastball has to rise. This leads to a ratio of IVB/second. Courtesy his work with the data, we find that there’s a certain threshold that we want to hit here and that is around the 44 IVB/s range. Anything above that is considered a fairly good fastball by this metric. Anything below is not that great, even if it is coming at 100 MPH, as Ben Porat notes using the example of Andres Munoz. It all goes back to the idea of fastball shape that we discussed at the top of this piece. In the article, he also notes that the 2022 leader in this metric was Pete Fairbanks at 53.7 IVB/s. Anything above 50 is considered in the elite range. Going on Fairbanks’ Baseball Savant page, we find that in 2022, his four seamer had a run value of 6 that year, or 2.9 /100 pitches thrown. The latter figure ranked 2nd in baseball (minimum 50 PA) among four seamers. Viewed in context of the elite IVB/s, we should not be surprised that Fairbanks generated a 38.1% whiff rate on the four seamer with a .147 xwOBA in spite of his often erratic command and injury issues.
Summary
Analytics serves less to challenge the old wisdoms of baseball than it does to understand the ‘why’ behind them and to contextualize them. These ideas give us tools to understand why certain players are great, and when players fall into ruts, give a better diagnosis of the issues and fix them. Analytics also allows us to see the many variables and nuances that lie beneath the surface of something so seemingly simple as a good ol’ country fastball. With these new tools and ideas, we can go back and understand why Marco Estrada was so good during the 2015-16 run of success in Toronto or why certain fireballers tend to get hit around and squared up more than what you would expect. Are analytics changing our ways of evaluating talent, yes, to some degree, but they are also confirming what we always knew, that baseball is a game of highly complex simplicities with further complexities hidden deep within.
Comments