xBA on Fastballs vs. Breaking Balls, Part 2
As mentioned in a previous post, now is the time to look at some possibly overlooked players headed into 2020, based on their consistency in hitting both fastballs and breaking balls well.
As a reminder, here is what the scatter plot looks like. You can look up specific players using the "highlighter" feature on the Tableau Viz.
Green represents consistency on both types of pitches, or guys that kill fastballs so much that the gap in expected average on breaking balls is not something to be concerned with. Red represents a player that is bad at hitting fastballs, or has a wide gap in expected batting average between fastballs and breaking balls. Yellow represents a concerning gap between the two. Once we look at bucketing players into these different groups, we can then take the next step of seeing if there is a discernible trend in pitch mix. For example, as noted in last week's post, Justin Upton saw fastballs seven percentage points less in 2019 than 2018, and that could have contributed to his decline when he was on the field. Now, let's take a look at some players in the green bucket that may be getting overlooked.
J.D. Davis (NYM - OF)
Truthfully, this is a guy who probably isn't getting overlooked. Everyone is pining for the Mets to get him regular playing time with the universal DH coming to an empty stadium near you. Everyone knows that he is a Statcast darling, but his consistency on each type of pitch is astounding.
DJ Lemahieu was the only other hitter to have xBA's above .300 in every type of pitch. That's some decent company. Now, I would expect JD's pitch mix to include more breakers and offspeed pitches going forward, given that barely anybody sees 60%+ fastballs anymore. At an ADP of 165, Davis is worth the 12-13th round pick with the universal DH. While there are reports that Davis may lose out on playing time even with the DH, and he still strikes out too much, expect him to play 90%+ of games in some capacity. When in doubt, pick the player with the best skills. Right now, that's Davis.
Alex Dickerson (OF - SFG)
Dickerson doesn't pummel fastballs (.286 xBA), but his minor dip on breaking balls (.256 xBA) means that he isn't likely to ever get into an extended slump. Dickerson's whiff rate on breaking balls is a way-too-high 33%, but his launch angle (12 degrees) and average exit velocity (90 MPH) make up for it. Launch angle is a function of getting out in front of the ball (Christian Yelich has a great explanation of it), so it's not a surprise that Dickerson whiffs a lot by trying to get out in front of the ball (and thus is early on breaking balls when he is expecting a fastball). To confirm this, he pulled 16 of his 33 balls in play against breaking balls - a limited sample, to be sure. Given that he is 29, he may not improve on pitch recognition. But, he could be a waiver wire pick up if he starts out hot and these trends hold. At an ADP outside of 600, there's no risk here.
Garrett Cooper (OF/1B - MIA)
If Cooper could just stay healthy, more people would know about his skills. Cooper's xBA on fastballs and breaking balls is an identical .273. However, he gets there in VERY different ways. While he doesn't whiff much on fastballs, his launch angle is 1 degree! He gets to this .273 through spraying the ball to all fields:
While those spray percentages are on all pitches, it's easy to see that Cooper's all-fields approach is a good batting average approach. As for breaking balls, his launch angle is more optimal (12 degrees), but whiffs at a 29% rate. Further, while that launch angle is better, his exit velocity is two miles lower on breaking balls than fastballs. Like Dickerson, expect Cooper's pitch mix to even out a bit more going forward. At an ADP of 413, you could get nearly 20 homers and a decent batting average out of him over a full season. If he plays most of the 60-game season, he will surely accumulate surplus value.
Tommy La Stella (3B - LAA)
Cubs fans shriek a little inside every time something good about La Stella pops up. His .314 xBA on fastballs and .300 xBA on breaking balls makes me do just that. His xBA's are fueled by tremendous plate discipline, whiffing less than 12 percent on both fastballs and breaking balls. His exit velocities and barrels are nowhere close to impressive, but he optimizes his launch angle to have a good batting average.
If La Stella reverts to his 2018 self, he is somewhat close to being Hanser Alberto. They have similar strikeout and walk rates, plus similar max exit velocities. La Stella barrels the ball a bit more, but its far from impressive. His 278 ADP is lower risk, but someone like Alberto is still on the board over over five rounds later.
Yandy Diaz (3B - TBR)
Yandy isn't the next Yelich quite yet, but his bat path suggests he has the capability to do just that. Specifically to this exercise, his high exit velocities on both fastballs and sliders yields high xBA's. His whiff rate is about league average on both pitches, and the only thing holding him back is getting more in front of the ball, resulting in a more optimal launch angle. Diaz will likely see less fastballs this coming year, but given that he hits all pitches so well, this shouldn't be an issue. If he can stay healthy, he should exceed that price.
There is certainly more to explore here, which starts with xwOBA on different types of pitches. As the season picks back up, tracking how the featured players in these pieces perform will be interesting. It takes a bit of time for expected stats to stabilize, and we may not be able to declare winners and losers of this exercise until after the season. Running some analytics on "xISO," or xBA-xSLG, is also intriguing. This would be similar to xwOBA, but with more of a focus on power.
The main next step, of course, is actual baseball to be played!
xBA on Fastballs vs. Breaking Balls, Part 1
In previous posts, there has been a focus on the difference in expected batting average between fastballs and offspeed pitches. Let's turn our attention to the hammers, the Uncle Charlie's, the slurve's. Rather than create tables, here's a scatter plot as an introduction:
The y-axis is the expected batting average on fastballs, and the x-axis compares the xBA difference between fastballs and breaking balls. This scatter plot is interactive here with options to search by player, league, and even position. One key item of note: amongst relevant fantasy players), the average xBA on fastballs was .276 in 2019, and the average xBA on breaking balls was .222.
The most consistent hitters have a difference within 40 percentage points - guys like DJ LeMahieu, Cody Bellinger, and Mookie Betts. These are the green circles. The yellow circles represent players that don't fare well against breakers, but absolutely pummel fastballs, more than making up for breaking deficiencies. Nick Castellanos, Nelson Cruz, and Marcell Ozuna are the representatives in this graph. There are also a handful of players that hit fastballs and breaking balls about the same, but don't do either particularly well. These few are also captured in yellow. Finally, the red circles the players who a wide gap in their xBAs. We talked about Ian Happ before (his weakness is fastballs, so he shows up in both offspeed and breaking ball analysis), so let's extend the analysis further.
There are plenty of players to investigate, but let's start with a few oft-injured players:
Aaron Hicks (NYY - OF)
2019: xBA FB: .241; xBA BB: .116
It's worth getting excited about Hicks being able to play mostly a full season in 2020, given that he was a 5.0 WAR player in 2018, which was his first relatively healthy season in the bigs. But even in that season, he struggled with breaking balls, as he had a .202 xBA on them. Hicks whiffed on breakers over 40% of the time in 2019, compared to under 20% on fastballs.
Hicks saw 55% fastballs last season, right around the average for fantasy-relevant hitters. That was a slight decrease from 2018. More importantly, he saw an uptick in breaking balls in 2019, as pitchers realized he struggles with them. Bet on the increase in breaking ball mix to continue, and do not expecting him to hit above .245 moving forward.
Justin Upton (LAA - OF)
2019: xBA FB: .246; xBA BB: .190
Is Upton starting to show cracks in his armor? After a few injuries in 2019 which limited him to 63 games, Upton had trouble with the bat (92 wRC+). Upton swung and missed at breaking balls nearly 46% of the time, compared to just under 22% on fastballs. Dating back to 2018, this gap was smaller (38.3% and 27.3%, respectively). Upton has always struck out above the league average, which should continue. However, his xBA on breakers back in 2018 was .257, which was above the league average.
Upton saw an eye-popping 61% fastballs in 2018, well above the league average. That trend goes back to at least 2016. In that small 2019 sample size, Upton saw 54.5% fastballs. Upton has posted a .365-ish xwOBA since 2016 on fastballs. Yes, he has hit breaking balls well too, with the exception of last year. With the downward trend in fastball usage by pitchers league-wide, and this concerning trend with J-Up, Upton may not make a full comeback. Assuming Upton is fully healthy, he should hit right around .245.
Gregory Polanco (PIT - OF)
2019: xBA FB: .263; xBA BB: .131
Polanco has severely struggled with left shoulder issues, and played just 42 games in 2019 - a small sample. However, his 2018 stats are not much better - he had a .283 xBA on fastballs and a .180 xBA on breakers. The common theme is that the gap on whiffs between fastballs and breaking balls is 10%. Historically, Polanco has struck out a below league-average rate, but his strikeout rate will likely be above league average as he shakes off the rust.
Much like Upton, Polanco has seen 60%+ fastballs in the past, including last year. His breaking ball mix should see an increase, given his sub-.200 xBA on these pitches the last three seasons. Based on this trend, a career-low in batting average is possible - think low .240s.
Let's shift to some healthy dudes.
Maikel Franco (KCR - 3B)
2019: xBA FB: .264; xBA BB: .207
Franco has shown so much promise, yet so much disappointment in his four-year career. Franco is the antagonist of the three true-outcome story (career 15.3% strikeout rate, 6.8% walk rate). But, it is safe to say that when does strike out, it's on a "secondary" offering:
Franco has always struggled with making hard contact, which is the reason his BABIP has been low, resulting in a low batting average. This is most evident on breaking balls, which Franco sees a ton of, relative to most hitters (35% since 2016). One slight adjustment could result in a breakout season, but that's been a theory for a few years now. The realistic conclusion is that these xBA's aren't all that bad, given the league averages stated above. He should hit a league-average .250 average with 20 homers over a full season - something available from a ton of players.
Jonathan Schoop (DET - 2B)
2019: xBA FB: .266; xBA BB: .207
Schoop is the mirror image of Franco. Consider this fun fact:
Schoop is the only player I've seen to have a higher exit velocity on non-fastballs! He did this in 2018 too. His lack of hard contact also results in low batting averages; however, when he does make hard contact, he barrels the ball much more often than Franco does. So despite striking out at a significantly higher rate than Franco, we can expect a slightly higher average - somewhere between .255-.260. Schoop sees anywhere between 56% and 59% fastballs in a given year, and based on past seasons, that trend should continue. Much like Franco, Schoop's xBA's are just below the league average.
The next post will focus on players who surprisingly have green circles, and may be overlooked heading into 2020!
xBA on Fastballs vs. Offspeed Pitches, Part 2
In Part 1 of Expected Batting Average: Fastballs vs. Offspeed Pitches, I took a look at some hitters who we could reasonably expect to regress. We found that Francisco Mejia and Sam Hilliard were guys who we could expect to negatively regress, based on them struggling against fastballs, and have the potential to see additional fastballs thrown to them as a result.
Today, we'll begin by taking look at hitters who are pretty consistent against fastballs and offspeed pitches, and whether that is good or bad. Let's start with this graph:
The key takeaways:
- Aristedes Aquino is due for some major regression! He struggles against both fastballs and offspeed pitches. His 195 ADP as of February 1 is a bit too rich for my blood
- Jeff McNeil, Carlos Santana, and Nolan Arenado are who we thought they were
- Hanser Alberto gets mocked for his low average exit velocity and hard hit rate, but his lack of whiffs carries his fantasy relevance. When I do a separate piece on xSLG by type of pitch, that's when we will see the flaws in his game.
- Baez has consistently proven that he outperforms his xBA, so seeing him have good-not-great xBAs on both fastballs and offspeed pitches is not surprising.
The one guy that is (rightfully) getting helium this draft season is JD. Davis. He does well on both fastballs and offspeed pitches. However, He clearly gets thrown a ton of fastballs, and I will be interested to see how he adjusts as pitching staffs realize that he needs to see more secondary offerings.
Finally, here's a preview of who fares well on fastballs, compared to offspeed pitches:
This list shouldn't surprise many, given that some of the best hitters in the game reside here. However, there a couple key takeaways:
- We know that Joey Gallo has trouble hitting for average, but actually seeing that he hit 100 basis points better on fastballs than offspeed pitches, when his xBA was only .254 to begin with, is just crazy. There is no way that Gallo repeats his .253 average in 2020. That, combined with the injury concerns I have (I'm expecting around 575 plate appearances), I won't be drafting Gallo this year. You can get Nelson Cruz in a similar spot in your draft for the same home run production and a better average.
- The same idea applies to Rhys Hoskins, but without the injury concerns. Hoskins is going 20-25 picks later than Gallo. Even with Hoskins' additional 85 projected plate appearances, he is projected to have less homers, similar counting stats, and only a slight better average. In other words, I don't think I'm taking either of these guys.
- From 2018 to 2019, Joey Votto saw a 4% decrease in fastballs, 1% inc. in breakers, 3% inc. in offspeed pitches. Let's see how that played out for him.
I bet he keeps seeing less FBs, and his average ends up around .260. I would love to see a bounceback, but I don't see how that is going to happen.
- For Miguel Sano, just check this tweet:
Next, we'll dive deeper into breaking balls. Enjoy the ride!
xBA on Fastballs vs. Offspeed Pitches, Part 1
As a fantasy community, we love talking about xStats. I decided to take a fresh approach in analyzing xStats and what that may mean for 2020. One thing that has not been analyzed enough in the industry is the difference between xBA (expected batting average) on the three types of pitches: fastballs, offspeed, and breaking ball pitches. If I'm a major league front office (I'm not), I would imagine that it would be easy to identify which hitters fare well (or should fare well) on certain types of pitches, and that certain hitters are seeing too much of those pitches (and vice versa). If a player fares well on fastballs relative to the other types of pitches, and sees a ton of fastballs (55%+), we should assume with reasonable certainty that he will see less of those pitches the next year.
To analyze the graph above, let's start with the most glaring example, Ian Happ. Happ had a paltry xBA of .189, .176 lower than his xBA on offspeed pitches. He crushes offspeed pitches relative to fastballs. This also applies to 2018 (.080 better on offspeed pitches). So, we know a trend is forming with Happ. Given that he is seeing 49% fastballs already, there isn't much room for him to go up. A deeper dive shows that Happ hits 36% of fastballs he puts into play to the opposite field, and another 36% to straightaway center. Simply put, he is habitually late on fastballs. If I'm the rest of the league, I'm upping that percentage to 55%, until he can prove that he can handle fastballs. Given that we know this data, we shouldn't expect the Player of the Week from the last week of September to be all that relevant in fantasy.
On a similar end, we have Francisco Mejia, who crushes offspeed pitches. But, he is only seeing 40% fastballs! We know that number is going to increase next year, and if he starts getting hyped up based on his prospect pedigree, you'll know better. Make him prove that he can hit fastballs before diving in.
Conversely, we have the opposite - guys who smack around fastballs, but have trouble with the change:
Judging by this graph, it is crazy that someone would throw Starling Marte a fastball 59% of the time. I would expect the league to adjust, and throw less fastballs going forward. Now, Marte is a veteran with an above-average track record, and is rightfully taken within the top 60 picks on an annual basis. However, looking at a younger guy like Rhys Hoskins with less of a track record, we are led to believe that he should be thrown less fastballs and more secondary pitches. Given that he is currently going ahead of Eloy JImenez in 2020 mock drafts, this type of analysis should shine light on the fact that he is overrated. You'll know better than to draft Hoskins at his ADP (or at all - I'm still salty that I took him in the fifth round in The Great Fantasy Baseball Invitational).
I plan to expand this analysis to breaking balls and xSLG as the offseason progresses. Let me know your questions and any particular players you are interested in!