I’m sure you’ve all heard the title of this column uttered by curious hapless on-air personalities at some point or another. But what does it mean? At the base level, it means a particular batter hits well when the pitcher throws fastballs.  As a corollary to this, I’ll say that ALL major leaguers are good fastball hitters.  It’s kind of in the job description, and you likely can’t make it all the way to the bigs without knowing how to handle the hard-and-straight pitches! How can we separate who is better at this skill however?  And how can we figure out if someone is a good change-up hitter? A good curveball hitter? I’ll tell you!

Something I’ve only recently begun to pay attention to are pitch values.  More precisely, they are known as pitch type linear weights, but pitch values sounds a lot better and is more intuitive.

The basics are as follows:

1) We can classify each pitch type thrown to a batter
2) We can identify the outcome of that pitch (strike, ball, out, hit, etc.)
3) We can assign a value to each of those outcomes
4) We can sum the outcomes to arrive at a metric!

Seems easy right?  Well, using the RE24 framework, as well as some more math that breaks values down to the per-pitch basis, rather than the per-at-bat basis, it is quite easy!

Regarding #1 above, I will note that both Baseball Info Solutions (BIS) and PITCHf/x classify pitch types.  I will be using the BIS classifications.

Think about it this way; if you’re a batter who never swings at sliders in the dirt, but a league-average hitter does swing at sliders in the dirt, you are more skilled against sliders (assuming similar outcomes when you make contact). Makes sense, right?  Our patient hitter will accumulate value when the pitcher throws sliders because he’s getting himself into a 1-0 count, or a 2-0 count.  When you are ahead in the count, you are more productive.  It’s one of those immutable facts of baseball.  Every hitter in history is better hitting 3-0 than 0-2.  So, shouldn’t you get credit for getting yourself into good counts?  I think you should!

The above example is a simple way to think about how these numbers can be interpreted.  As with many useful metrics, pitch values are scaled to a league-average value, which happens to be zer0.  Then, the metric can be scaled to a per-100-pitch basis to eliminate the need to say Player A has faced 100 fastballs while Player B has faced only 80 fastballs.  What we arrive at are our pitch weights.  The metric is written as “wFB/c.”  The “wFB” portion shows you are using the linear weights on fastballs (FB). You could also have “wCH” for a change-up, for example. The second portion with the “/c” means “per 100 pitches.”  Don’t ask me why they used the letter C, because I do not know!  Perhaps they are reliving their high school Latin class…

So, if someone has a wFB/c of 1.50, that means this player produced 1.5 more runs than a league-average fastball hitter per 100 fastballs seen.

I know you are now just dying to see how our Reds rank, right?  I hope so, or this column has been a spectacular failure!

Let us look at fastballs to begin:


The generic term “fastball” includes both the 4-seam and 2-seam variety.  I also included cutters here, since they are a type of fastball.

What we see is sort of interesting.  No Red is really producing a lot more than average, or a lot less than average.  This sort of goes back to my statement that just about everyone hits fastballs well.  Because of this, the spread from the best hitter in the league to the worst hitter in the league is smaller with fastballs than with any other pitch.

Cutters are a different story.  Although classified as fastballs, they require a different type of skill to hit, apparently.  Also, cutters that move back over the plate from the inside or outside (back door or front door, depending on handedness) are commonly taken for strikes by some batters, thus lowering their overall wCT/c.  Conversely, if someone is very good at taking cutters for balls, their wCT/c might be higher than expected.  Because of this interaction, we see a larger range.

We’ll next look at breaking balls:


If given a guess, I doubt many of us would have identified Jay Bruce as the best slider hitter on the team.  Again, it would be more accurate to say Bruce has been the most valuable against sliders, not necessarily the best hitter.  If Bruce is chasing fewer sliders in the dirt or outside this year, it makes sense that he’s doing well against the pitch.  To find out these types on things, we can look on sites like Brooks Baseball, which will show us swing rates against any type of pitch, as well as contact rates and slugging.  If anyone desires to dig deeper, I recommend starting at Brooks Baseball and trying to find out why Bruce has been so good against sliders this year.  I think you’ll find the experience rewarding!

Regarding curveballs, Joey Votto is a man among boys.  His 9.4 runs above average (RAA) figure for wCB/c is #1 in the National League by almost 50%.  The guy in 2nd is Aledmys Diaz of the Cardinals and he’s in the mid-6s.

When Votto puts a curveball in play, he’s 11 for 17 (.647 BA) with 2 homers.  Also, we can safely assume Votto chases fewer curves out of the zone than just about anyone else because, well, he swings at fewer pitches of all types out of the zone than just about anyone.  Combine good results on contact with good discipline, and NL pitchers would be wise not to throw Votto many curveballs.

Finally, we’ll look at off-speed pitches.  This section will consist of only change-ups because splitters have weird results caused by he fact that BIS doesn’t classify splitters very well (in my opinion). Also, it is a little-used pitch giving us smaller samples.


Ouch.  Only 2 Reds are above-average at handling change-ups.  Seeing things like this make it no wonder that guys like Jeremy Hellickson (on Opening Day) do so well against our Reds.

I really like seeing Adam Duvall doing well in this category, because with the type of player he is, as long as he can differentiate between fastball and change-up (seems like he can), he always has the chance to learn how to lay off sliders and curves.  That’s a solid blueprint for a successful hitter.

I know there was no big reveal and nothing ground-breaking here, but I just wanted to introduce everyone to a perhaps new metric that I’ve been studying lately.  It gives us yet another way to dig deep and find out what is happening with players who may be streaking or slumping.  Perhaps a good change-up hitter has faced 5 change-up pitchers in the last week, which might be artificially inflating his stats, for example.  Regardless, I hope you had a decent time reading!

Pitch value figures are courtesy of BIS via FanGraphs.

13 Responses

  1. Hotto4Votto

    With Duvall hitting curveballs well, in fact he’s only below average in cut-fastballs. I don’t know how a batter would go about identifying cut-fastballs vs other fastballs. But he seems to hit everything else above average. Same for Bruce, consistently above average other than curveballs.

    Nice article. Thanks for info. It’s fun.

  2. mdhabel

    Patrick – How much of this type of information do you think goes into game planning for hitters? Obviously each at bat is different and more complex than just saying “dont throw Votto curveballs”. At least I would assume?

    • Patrick Jeter

      I’d actually say very little of this goes into game planning. You’re right. There’s a lot more to it. If you have a ton of data about a certain player, maybe it weighs in some?

      However, according to the description article on FG, the year-to-year correlation of pitch weights is only 0.25… so just because you were good at hitting curves this year doesn’t mean you’ll be good at hitting curves next year. Most of it is likely tied to the pitcher. For example, much harder to hit Kershaw’s curve than Moscot’s curve (assuming he throws one…)

      I’m not quite sure what to make of it all, but I think it is a fun topic to start thinking about.

      • lwblogger2

        Also when talking about a particular number of a certain type of pitches, I’m guessing the sample size is pretty small over a single season. Perhaps not for fastballs but for every other pitch type I’d think.

  3. John Gattermeyer

    I find this fascinating. I had the same initial thought about using it for game planning. However, I think as a pitcher, you can use this to a certain extent. There was a recent FG article on Mike Trout and how he’s been demolishing curveballs this year. That article identified Joey as the master of mashing the bender. I think as a pitcher that relies on a curveball, knowing Votto either won’t swing at a ball in the dirt or will crush a hanger, you try to avoid that. Otherwise, pitchers probably don’t have enough variance in their arsenals to eliminate a whole pitch against one batter. Thanks for the article on this though, very interesting!

    • Patrick Jeter

      Good point, John.

      Unless you are a Johnny Cueto-type (who can vary delivery, timing, etc), you are probably right that eliminating 1/3 to 1/2 of your arsenal would probably yield even worse results.

  4. cfd3000

    Interesting stuff Patrick, thanks. I suspect the year to year variation is a sample size issue much like the (limited) usefulness of batter vs. pitcher histories. Not all sliders, curves, change ups or even fastballs are the same. Nevertheless when you see large positive or negative values backed by decent data sets then you have something to plan with. You can beat Votto by changing speeds, and to a lesser extent locations, but he can clearly identify curve balls in ways that no one else can. On an unrelated note I can’t help wondering how this data might affect the thinking of guess hitters vs. see and react hitters. Thanks!

  5. msanmoore

    Very good stuff. Easy to understand but, as others have noted, there are many other factors in play and that vary in game. Bigger samples over time will tell just how useful this could be.

  6. Playtowin

    Good stuff. Does anyone know why Votto strikes out so much? With a good eye and apparently good bat control he sure swings and misses a lot.

    • Patrick Jeter

      “Bat control” is sort of a misnomer. Votto has about average contact on pitches in the zone. He swings at every few pitches on the edges of the zone, and umps are ringing him up more often this year. So, I think a combination of a different zone being called than the last few years, and perhaps a bit of variation.

      • vegastypo

        Increasingly, Votto seems to be guessing more at what pitch he is going to get, and appears less able to flick away a foul ball to keep the plate appearance going when he guesses wrong.

        Maybe it’s my memory, but I don’t recall from him in the past.

  7. Eli J


    Good stuff once again! Thanks for the thoughtful approach.

    I read an article on ESPN recently on the same topic–various batters’ abilities against specific pitches.

    Rather than the RE24 framework, the article uses OPS. ESPN is not as generous as you in giving the hitter credit for getting to a more favorable count.

    It’d be interesting to unpack this a bit more. Basically: how much value comes from working the count, and how much value comes from the outcome of the plate appearance? Are there any hitters who are atypically bad at using a pitch to get to a better count, but atypically good at using that pitch to get on base (or vice versa)?

    I bet there’d be very few surprising results here, but any major discrepancies could be put to tactical use in deciding how to pitch to a batter.