+

Going Deep: Using Statcast to Identify This Year’s Busts

Dan Richards predicts three players will be the biggest busts of 2019 drafts.

We did it. We survived football season. Patriots fans have now transformed back into Red Sox fans and New York Giants and Jets fans have emerged from hibernation, ready for redemption on the baseball field. That means it’s time for some fantasy baseball analysis.

Let’s get down to brass tacks. No one wants to read my musings anyway. This one’s about three purported “power hitters” I’ll be avoiding come draft day. They’re all expensive, have name value, run very little and have provided excellent power in the past. Fantasy players and analysts alike are banking on that power to repeat, and, using Statcast metrics from BaseballSavant, I’ll explain why I think it won’t.

 

George Springer (OF, Houston Astros)

 

In 82 NFBC drafts since January 1, George Springer has gone about 60th overall. In 620 PAs last season, Springer hit only 22 HRs with a .265 AVG and 6 SBs. As he approaches 30 and hasn’t stolen more than 9 bases since 2015, I imagine those steals remain low in 2019. I would be remiss, however, if I didn’t mention that Springer derives much of his value from scoring runs atop the Astros lineup, which he did 102 times last season. But with 71 RBI, he doesn’t move the needle in any other category. And yet, he’s a late fifth-round pick in most drafts. I think people are expecting a bounce-back based on name value alone, but aren’t looking deep enough to catch the warning signs.

Springer does have some things going for him. His 2018 batting average was the same as his career mark (.265), and his .303 BABIP was in line with his career BABIP. I think there’s some batting average rebound potential based on his poor Line Drive and Low Drive marks that dragged down his BABIP and were well below his career rates. These stats take over a season to stabilize and are therefore more likely than not to be a matter of bad luck, signaling potential BABIP and batting average positive regression. Plus, he’s shown the ability to run high Line Drive and Low Drive Rates in the past. However, Springer maintained a below-average 29.5 PH% and xStats gave him a .258 xAVG and .290 xBABIP. With these mixed signals in mind, let’s just conservatively assume he hits about .270 next season.

But this article’s more about power hitting. So for those of you discounting the 22 HRs he hit in 2018 and praying on another 34 homer season, let me drop some knowledge on you. Springer’s .169 ISO and .434 SLG were mediocre. It’s not surprising either, given he put the ball on the ground 49.4% of the time and only hit 34.6% fly balls.

Springer’s 33.4 Hard% was actually below league average this season. Not a great sign for a power hitter. However, as Baseball Info Solutions’s Hard% (available on Fangraphs.com) is calculated, in part, qualitatively, it’s somewhat deficient and subject to random variance. In fact, its correlation with home runs was much lower in 2018 than any other season in the past five years. Accordingly, I will begin substituting it for Statcast’s Hard Hit Rate in my articles, which is rightly calculated as the percentage of batted balls hit over 95 mph. Confusing as this is, there is a meaningful difference between Hard% and Hard Hit%. I like the latter’s objectivity and ideally, we as an industry can move away from simply stopping at Hard% on Fangraphs player pages. Springer, therefore, gets a pass in this respect.

But he can’t escape the truth that lies in the Statcast data. Thus, here’s a handy table that I’ll employ for each player to make sense of their true power potential.

Statcast Metric George Springer League Rank
Brl/PA 6.1% 95
Exit Velocity on FB/LD 93.3 mph 125
Hard Hit 37.2% 150
Launch Angle 9.5° 251

What do these numbers tell us? Barrels are what we care about most because they’re quite likely to become home runs. We could already suspect that Springer would have a mediocre barrel rate given that he only hit 22 home runs in a full season of PAs, but why is it so low? Fortunately, I gave you everything you need to know above to understand.

Per Statcast:

During the 2016 regular season, balls assigned the Barreled classification had a batting average of .822 and a 2.386 slugging percentage. To be Barreled, a batted ball requires an exit velocity of at least 98 mph. At that speed, balls struck with a launch angle between 26-30 degrees always garner Barreled classification. For every mph over 98, the range of launch angles expands.

With that said, Springer’s Barrel% was low for two reasons. First, his average launch angle was low because he wasn’t putting the ball in the air often, which we know from his high GB% and low FB%. It’s not impossible to smack a lot of home runs this way (e.g., Christian Yelich), but it’s far less likely because a low average launch angle means fewer balls hit in the 26-30 degree range, and as a result, fewer opportunities for Barrels in the first place. Broadly speaking, power hitters maintain average launch angles between 12-20%.

Second, he might just not be strong enough to keep up with the best power hitters. As discussed, Barrels require a minimum exit velocity of 98 mph. Yet, Springer hit the ball over 95 mph just 37.2% of the time, a mark bested by 149 other players. When he did put it in the air, it only traveled 93.3 mph on average, which is why he lacked a significant number of Barrels.

For some context, the most prolific power hitters in the league average around 97-98 mph on their fly balls and line drives. They also hit the ball “hard” according to Statcast about 50% of the time. In other words, they, unlike Springer, not only hit their FB/LD just hard enough on average to produce a barrel (when hit at the proper angle), but also hit the ball hard more frequently. For further context, here are some fun names who hit their FB/LD harder than Springer: Marwin GonzalezMitch Garver, Yairo Munoz, and the corpse of Alex Gordon.

Suffice it to say, I’m bearish on a power rebound from Springer unless he was suffering from some unknown injury that sapped his power in 2018. He neither hit the ball hard enough nor elevated it enough to make him stand out as a power hitter. Unsurprisingly, xStats bought the skills decline too, giving him just 23.8 xHR.

If he plays a full season, I’ll project 110 R/24 HR/75 RBI/6 SB/.270 AVG. I don’t want to pay fifth-round value for that, do you?

 

Gleyber Torres (2B/SS, New York Yankees

 

Boy is there helium around Gleyber Torres, who’s priced at 58th overall on NFBC. Granted, he just turned 22 and the sky’s the limit for the kid, just not yet, in my opinion. As a Yankee fan, I love watching him play. In one of the best games I’ve ever been to, he walked it off in style:

But I am a rational fantasy baseball player, and I won’t be buying Torres at his price when I could instead have Jameson TaillonTommy PhamZack Greinke or Lorenzo Cain. In 2018, Torres hit .271 with 54 runs, 24 homers, 77 RBI, and 6 SBs. Let’s start with speed. His 27.1 ft/sec sprint speed was good for 299th overall, so he’s not as quick as many probably expect. At his peak, he swiped 22 bags in a season of A-ball, but that’s not a torrid pace, and it’s possible that, like many others, he won’t run now that he’s in the big leagues (he didn’t much in 2018). Let’s say in a full season he steals 10 bases, as most projection systems have him between nine and 11.

His batting average was weighed down by his 25.2 K%, but buoyed by a .321 BABIP. That K% will likely get worse, given his 14% Swinging Strike Rate portends more like a 28-32 K%. This will certainly drag down Torres’s average next season.

xStats does credit him with a .323 xBABIP, but it’s mostly high due to his elevated 19.7 Low Drive%, as Low Drives almost always result in hits. Like I mentioned with Springer, however, Low Drives do not stabilize for over 2000 PAs, so it’s unclear at this point whether his Low Drive% is a matter of luck or skill. It could easily regress to the mean and bring his BABIP down with it.

In addition, Torres’s Pop-Up% was 6 points above the league average, and he hit 42.7% fly balls and pulled the ball 42.2% of the time. All three of these metrics are bad for BABIP and stabilize quickly, suggesting impending regression. I therefore think it’s reasonable to expect his BABIP to decline at least to a league-average .300, dragging his batting average down to .250. Indeed, he spent the majority of his short minor league career in A-ball. In his most recent large sample from A-ball, he hit .275. Obviously, MLB pitching is better, so I’d say with a natural BABIP and strikeout rate regression, him being more of a .250 hitter makes sense.

Dan! You’ve gotten lost in BABIP and batting average again. Back to the matter at hand. Torres impressively popped 24 homers in two-thirds of a season, with a .209 ISO and .480 SLG. As discussed, he pulls a lot of fly balls, which is great for his power. That’s in part why he maintained a 17.9 HR/FB% and hit so many homers in so few PAs. Nonetheless, let’s take a look under the hood to see if that’s sustainable.

Statcast Metric Gleyber Torres MLB Rank
Brl/PA 6.0% 99
Exit Velocity on FB/LD 91.2 mph 229
Hard Hit 36.8% 159
Launch Angle 18.8° 22

Okay, so Torres is different than Springer. He elevated the ball basically twice as often as Springer did, which is a great sign. And, playing most of his games in Yankee Stadium and other small parks across the AL East, one could believe that his launch angle alone is sufficient to hit another 24 homers next season.

Still, I’d argue that optimal launch angle is necessary, but not sufficient, to continue hitting home runs at this clip. As discussed with Springer, the other necessary component is raw power. Torres doesn’t hit the ball as hard as Springer, producing poorer marks in exit velocity on FB/LD and Hard Hit%, while maintaining a similarly mediocre Barrel% (due to his higher launch angle). That’s why in the minor leagues, he never had a HR/FB% close to what he maintained in 2018. In his two largest minor league samples, he had HR/FB rates of 2.5% and 11.4%. I just finished telling you how Springer doesn’t hit the ball hard enough to be an elite source of power, and he hits the ball harder than Torres does and he hits it hard more often too. If I had to pick one of the two to sustain an elevated HR/FB% based on power metrics and track record, I’d pick Springer, even though Torres might hit more homers overall because of his launch angle. That 17.9 HR/FB% is likely coming down.

Also, I derided Springer, so, in fairness, I have to pick on Torres too. Here are some guys who hit their FB/LD harder than Torres: Johnny FieldJason HeywardJace Peterson and 37-year-old Ben Zobrist.

Perhaps Torres can just continue to pull enough fly balls to take advantage of the contours of Yankee Stadium. Or maybe his K% catches up with his SwStr% and his truly middling power tendencies emerge, limiting not only the amount of contact he makes, but also the number of home runs resulting from that contact. Maybe he really was what he did in the second half of 2018, where, in 243 PAs, Torres batted .249 with nine home runs. Let’s convert those nine homers into 12, assuming he would have 325 PAs in a true half-season, double it, and we’ve got 24 homers total. In fairness to my skepticism, xStats gives Torres only 19 xHRs for the 2018 partial-season, as opposed to the 24 he hit. RosterResource also has him batting near the bottom of the Yankee lineup, sapping his run and RBI totals.

Based on the foregoing, here’s my projection for Torres: 70 R/24 HR/85 RBI/10 SB/.250 AVG. With that projection, he pretty clearly will not return fifth-round value.

 

Miguel Andujar (3B, New York Yankees)

 

Fellow Yankees fans will likely be seething by now, but I have to call it as I see it. Don’t worry, my next sleeper article will highlight one Yankee I love for the same reasons that I hate these three players at their current prices.

But I digress. Going about 71st overall in NFBC drafts, Miguel Andujar undoubtedly had an excellent rookie season, hitting 27 home runs in 606 PAs with 83 Runs, 92 RBI, 2 SBs and an impressive .297 AVG. Don’t let the slash line fool you though.

I actually buy Andujar’s batting average and .316 BABIP. Maybe his 16 K% increases a little due to a 9.6 SwStr%, but he’s always had good plate discipline in the minor leagues, so I could see his SwStr% declining too. Unlike Torres, Andujar’s BABIP isn’t supported by an elevated Low Drive Rate and he doesn’t have a concerning Pop-Up Rate. Sure, his PH% was very high, but so was his VH%, and he hits a fair amount of line drives and ground balls, which should sustain a high BABIP. For these reasons, I’m not surprised xStats gave him a .326 xBABIP. I expect Andujar to hit between .285 and .300 next season, right in line with his projections.

What’s really cause for concern is his power indicators. Andujar hit 44% ground balls and 35.8% fly balls and his Pull Rate was 47.5%. A decent recipe for home runs no doubt, though I’d like to see more fly balls. In any event, his HR/FB Rate was 15.7%, which begs the question: is he really the type of hitter that can muscle fly balls out of the park at that rate?

Statcast Metric Miguel Andujar MLB Rank
Brl/PA 5% 146
Exit Velocity on FB/LD 92.7 mph 146
Hard Hit 36.5% 168
Launch Angle 12.3° 175

Hopefully, at this point, you already know what I’m going to say. Andujar’s Barrel% is bad, the worst of my three subjects, and it’s the result of his poor exit velocity on FB/LD and low Hard Hit Rate. He’s elevating fine, but he doesn’t hit FB/LD very hard, and he doesn’t hit the ball hard often enough. Here are some names of players who hit their FB/LD harder than Andujar: Matt JoyceManuel MargotBrian Anderson and Kevan Smith (who?).

You could bet on Andujar and Torres simply repeating their impressive home run totals because they play in Yankee Stadium, but I’m not much of a betting man, and even if I were, I’d rather not just assume that two not-so-powerful hitters will overperform in power two seasons in a row.

Plus, the Yankees might wisen up. Despite nearly winning AL Rookie of the Year, with a full season of PAs, Andujar could only muster a 2.7 fWAR. That feels a lot like his ceiling, given his low .328 OBP and historically awful defense. The Yankees know batting average is no longer the name of the game, so if his power diminishes as I expect it might, and he doesn’t improve on defense, that may be enough to keep him out of the lineup. Indeed, the Yankee lineup in particular in flush with talent and there are too many talented hitters to let Andujar fill the DH slot.

Assuming he plays a full season, I project the following: 75 R/21 HR/80 RBI/2 SB/.290 AVG. He’s batting sixth according to RosterResource, so I have to give him a few more runs than Torres, but with a lower barrel rate than the other guys, I just can’t give him as many HRs and RBI. And to me, that return isn’t worth the late sixth-round price.

Photo by: Juan DeLeon/Icon Sportswire

Dan Richards

Dan is a lifelong New York Yankees and Giants fan. A practicing attorney, Dan is better known for aggressively bothering his leaguemates about trades. You can follow him on Twitter @Fantasy_Esquire or by clicking the button above.

11 responses to “Going Deep: Using Statcast to Identify This Year’s Busts”

  1. theKraken says:

    It will be interesting to check back in after the season and see how these trends went. Its an interesting sample that you chose as the two Yankees are in for regression following fortunate rookie seasons and an aging Springer – who is the most interesting choice to me. Personally, I don’t think that xData is sticky or menaingful enough to tell us much about the future but I am certainly interested in following it. I don’t even understand why we use average launch angle at all – median is the only measure of center that makes sense. So a bunch of IFFB help? Its people like you, putting this stuff in writing that makes it possible to analyze the results, so thanks for that. Most talk about barrels and angles feels pretty anecdotal, but this is consistent and thoughtfully discussed. Isn’t it kind of interesting that people don’t admit to scouting the stat line, but scouting the stats of the batted balls is next level? I am not sure it is, but there is only one way to find out. Thanks for the work!

  2. Turp says:

    This is awesome! Uhhh….can you do another one of these? :)

    • Dan Richards says:

      Thanks! I’ll be doing one on hitters I like (instead of just dumping all over players I don’t, for once) next week. I also post interesting Statcast notes frequently on Twitter so give me a follow there for more @drichardsPL

  3. Travis Brown says:

    Now that’s what I like to see! I’m investing in Voit, and hoping some spots open up for Lemahieu that isn’t 1st base haha.

  4. Eli says:

    I have Andujar at $7 in a 12 team keeper league. I’m buying your analysis, but do you think I should be shopping/cutting him even at that price?

    • Dan Richards says:

      On a $260 budget, he’s like a true talent $5-10 player in my mind, but he might go for $15ish. I’m obviously lower on him than the market. If perceptions are higher in your league, perhaps try shopping him?

  5. Ben Pernick says:

    I’ll beat this drum again, but what about Nolan Arenado? His Barrel% relative to the league has declined every year. It ranked #39 in 2015… Thanks to lower launch angle, last year it was down to #135. and Statcast listed his closest comp as Asdrubal Cabrera. Yet he went at #5 before J.D. Martinez in my last mock draft.

    • Dan Richards says:

      This is interesting. I’ve noticed this with other Rockies hitters as well, like Blackmon. They definitely overperform their expected barrel and HR rates based on their exit velocity and launch angle data.

  6. RaymondTraug says:

    Natural Stress Solutions Full Spectrum CBD Tincture Archives: http://tropaadet.dk/pills76312

Leave a Reply

Your email address will not be published. Required fields are marked *

Account / Login