post

Revisiting xBABIP Outliers: 1st Half 2019 (Fantasy Baseball)

Today marks the much anticipated Opening Day for Major League Baseball. Unfortunately, there is no baseball. This is a sad day, BUT, it has allowed for more time to go back and research topics I otherwise would not have had time to revisit. Let’s first go back to the original piece I wrote at the midpoint of 2019 covering BABIP outliers.

Using xBABIP to Find Outliers – Players to Buy/Sell for the 2nd Half




In that piece, I covered both over and under-performers. I’ll cover the under-performers next week, but first, let’s recap the players with the largest discrepancy between xBABIP and BABIP through the first half of 2019.

2019 1st Half xBABIP Outliers - Over-performers

PlayerBABIP 1HxBABIP 1HxBABIP-BABIP
Rhys Hoskins0.3080.242-0.066
Omar Narvaez0.3240.249-0.075
Charlie Blackmon0.3490.285-0.064
Brandon Lowe0.3970.314-0.083
Nolan Arenado0.3170.261-0.056
Eduardo Escobar0.3070.250-0.057
David Peralta0.3500.297-0.053
David Dahl0.4100.367-0.043
Miguel Cabrera0.3610.312-0.049
Trevor Story0.3610.307-0.054
Christian Vazquez0.3210.272-0.049
Gleyber Torres0.3190.266-0.053
Eric Sogard0.3190.272-0.047
Corey Seager0.3220.273-0.049
Elvis Andrus0.3490.294-0.055
Christian Yelich0.3280.295-0.033
Brian Goodwin0.3550.313-0.042
Marcus Semien0.2920.266-0.026
Austin Meadows0.3680.332-0.036
Tim Anderson0.3720.329-0.043
Jorge Polanco0.3490.320-0.029
Jeff McNeil0.3800.340-0.040
Adalberto Mondesi0.3520.322-0.030
Xander Bogaerts0.3280.301-0.027
Juan Soto0.3650.323-0.042
Joey Votto0.3260.296-0.030
Difference0.3430.296-0.047

Most of us were likely able to identify many of these players as BABIP regression candidates for the second half. As a whole, these outliers had an average BABIP of .343 through June 26th, 2019, nearly 45 points above the league-average. In fact, xBABIP pegged the group as essentially league-average in terms of BABIP based on the batted ball data per Baseball Savant. The table below tracks how each player fared after June 26th. We can expect regression, but how much?


2nd Half Performance: 2019 1H xBABIP Outliers - Over-performers

PlayerBABIP 1HBABIP 2HBABIP 2H - BABIP 1H
Rhys Hoskins0.3080.228-0.080
Omar Narvaez0.3240.286-0.038
Charlie Blackmon0.3490.321-0.028
Brandon Lowe0.3970.278-0.119
Nolan Arenado0.3170.306-0.011
Eduardo Escobar0.3070.260-0.047
David Peralta0.3500.274-0.076
David Dahl0.4100.324-0.086
Miguel Cabrera0.3610.311-0.050
Trevor Story0.3610.3620.001
Christian Vazquez0.3210.291-0.030
Gleyber Torres0.3190.273-0.046
Eric Sogard0.3190.313-0.006
Corey Seager0.3220.286-0.036
Elvis Andrus0.3490.266-0.083
Christian Yelich0.3280.3900.062
Brian Goodwin0.3550.314-0.041
Marcus Semien0.2920.2990.007
Austin Meadows0.3680.300-0.068
Tim Anderson0.3720.4300.058
Jorge Polanco0.3490.306-0.043
Jeff McNeil0.3800.289-0.091
Adalberto Mondesi0.3520.3680.016
Xander Bogaerts0.3280.3480.020
Juan Soto0.3650.266-0.099
Joey Votto0.3260.308-0.018
Difference0.3430.308-0.036

19 of the 25 outliers regressed in the second half of 2019. Some of them saw heavy regression. Trevor Story essentially broke even, so basically, 80% of the over-performers finished with a lower BABIP in the second half. The average drop in BABIP from the group was 36 points. When compared to the average xBABIP-BABIP differential in the original table, the group collectively regressed about 77%. If we exclude the players who actually improved their BABIP in the second half, the differential between 1H BABIP and 2H BABIP is a whopping .052! We’ve got a small sample of outliers but it’s very telling that the first half xBABIP was a much better predictor of second-half BABIP. At least for this group of outliers. Let’s dive into the analysis on each player with some tidbits for 2020.

Rhys Hoskins’ regression was obvious given his profile. Slow-footed hitters with 50% fly ball rates and high pull percentages rarely produce near league-average BABIP, let alone above-league average. Not only did he regress, but he also fell below his xBABIP from the first half. Despite a great eye at the plate, we can expect Hoskins to continue to carry a BABIP around .250 going forward.


Regression came but not as hard as xBABIP predicted for Omar Narvaez. He’s shown strong bat to ball skills and a tight launch angle variance which has helped him outperform his metrics over the last two seasons. It’s no surprise that he once again managed a league-average BABIP. He may continue to outperform his expected metrics going forward but I’m not betting on a .300+ BABIP. Coors Field is largely at play for Charlie Blackmon. Look no further than his home/road splits: .376 BABIP at home vs .296 BABIP on the road. Simply put, he’s a .325 hitter at home and a .275 hitter on the road.

This was an easy win with Brandon Lowe. A .397 BABIP is not sustainable (unless you ask Yoan Moncada, Tim Anderson, and Fernando Tatis Jr.). My concern for Lowe is that his true talent is a .300-.320 BABIP hitter. We need a larger sample but if that’s the case, he’s going to hit .230. Once again, Coors Field is to blame for Nolan Arenado. No need to dig deeper. I’d expect him to hit .275 if he’s traded.

Man, I really expected a major collapse from Eduardo Escobar in the second half. While his BABIP almost completed regressed, his power did not. Besides, the BABIP dropping, his power remained strong in the second half despite extremely poor power metrics. Alas, his power sustained as he hit 17 home runs in the second half after clubbing 18 in the first half. Shrug emoji. Although he’s a major candidate for regression based on my eHR metric in 2020, he’s still a safe .260 hitter with low-to-mid-20s pop.

Injuries certainly played a role here, but David Peralta’s batted ball profile did not portend to anywhere near a .350 BABIP as his speed continues to diminish. I’m not one to project a resurgence to the 2018 version of Peralta but stranger things are happening at the moment. I really love that David Dahl was carrying such a lofty xBABIP through the first half of the season. It all came crashing down to a still solid .324 in the second half. His career BABIP is .369 and I think that’s close to his skill level given his batted ball profile, speed, and Coors Field. He was unlucky based on eHR, so health is really the only thing holding him back. A healthy Dahl could be a major breakout and a top-50 fantasy asset.


Almost nailed it! Miggy is a shell of himself but despite being 36 and one of the slowest players in the majors, he’s still posted better than average BABIP. Even xBABIP thinks so. But I digress, there’s no value here. He’s turning into empty batting average much like Joe Mauer circa 2015. Trevor Story put together a hell of a season. He outperformed his BABIP in the first half but managed to match his xBABIP in the second half. Despite posting back to back seasons with a batting average over .290, the projection systems and his xBABIP peg him as a 275 hitter. What do you think?

I’m not sure Christian Vazquez will maintain a .300+ BABIP again but it’s fun to look at 2019 as an outlier. Gleyber Torres only hit 5.7% of hit ground balls to the opposite field yet managed an above-average BABIP on balls hit on the ground. He was shifted on 33% of his plate appearances. I expect that to rise while his BABIP on ground balls plummets. Projections have his BABIP over .300 which I think is a mistake, especially if he continues to hit pop-ups at an above-average clip. What happens if Gleyber is a .250 hitter?

The second half metrics were strong for Corey Seager but xBABIP isn’t buying it. If he never fully develops into a 30-homer hitter, he could be another boring .280-20-HR type player that does very little for me. Elvis Andrus was dealing with an injury but even still, he was never going to maintain a BABIP near .350. He’s on the wrong side of 30 and his sprint speed is scary low for a player with 30 swipes in 2019. We may be looking at the beginning of the end for Andrus.

Christian Yelich: The un-regressionable candidate: Ideal launch angle for batting average, elite hard contact, great foot speed, the list goes on. He’s the s%$t. Hey, look! I nailed this one – Thanks for making me look good Brian Goodwin. Marcus Semien just keeps getting better. He bounced back spitting in the face of his first-half xBABIP crushing it in the second half. I think we saw the peak Semien season in 2019 but he should be a solid fantasy player going forward. Just not at his current price.

Austin Meadows xBABIP was a solid .332 in the first half and he came all the way down to .300 in the second half. Do we have enough data on Meadows to know what kind of hitter he is? I’m not so sure. For those expecting batting average as one of Meadows’ major assets could be disappointed in 2020. I see him hitting anywhere from .250 to .290. Hi Tim Anderson! Major shrug emoji here. He did hit the ball harder, at lower launch angles, plus he’s got great speed. Even still, Anderson is likely to hit .270 next year and that’s just fine given his power/speed combo.

Yeah, we didn’t believe you either Jorge Polanco. He is like a poor man’s Jeff McNeil. There’s value here but also no need to reach at all. What type of fantasy player is Jeff McNeil if he has a .289 BABIP? Well, he hit .276 in the second half. His power did jump up, but I don’t believe it’s fully sustainable. The good news is, I actually believe he can carry a .330 BABIP going forward based on the data from a majority of two seasons but expecting 23 homers again is a fool’s errand.

We have to accept that Adalberto Mondesi is always going to outperform his xBABIP. It’s likely due to his batting average on ground balls. His batting average minus expected batting average (BA-xBA) on ground balls was .035. I don’t think Statcast fully takes into account the elite speed aspect of his game. He will always outperform his xBA on grounders. However, he was fortunate on line drives by about 100 points, so expecting a BABIP of .350 again is not wise.

For Xander Bogaerts, here’s my explanation. His continued overperformance is a little bit of luck and a little bit due to his home park, Fenway. His BA-xBA on balls in play was .012. So, a little lucky, but nothing crazy. However, if we isolate his balls in play in Fenway Park, his BA-xBA is .059! We should anticipate another BABIP north of .310 from Bogaerts but with neutral luck, we are looking at something close to .320.

Overcorrection much? Juan Soto may have been lucky on his BABIP in the first half but it came all the way back and then some in the second half. I know Soto is a lefty but he sprays balls all over the field and rarely pops up. He’ll continue to carry a .300+ BABIP while smashing 30+ homers. He’s still just 21. I think before he’s 26, we will see a .325-40-120 season from Childish Bambino. One can dream. Joey Votto is kind of in the same camp for me as Miguel Cabrera. After an extended period of greatness, their time has passed. Stay tuned for the underperforming list next week.


Follow me @FreezeStats. Check out my work at FantasyPros and Pitcher List.




(Photo credit: Andy Marlin, Andy Marlin-USA TODAY Sports)

Shortstop – The Choice Is Yours

The first article I did for ADP value picks was for outfielders with speed. This one is going to be similar but for shortstops in that most of these options at short will have some speed. As a reminder, I’m using ZIPS projections for 2018 and NFBC ADP for all the players. So how this works is, I display ZIPS projections for four or five similar mystery players at the same position and we figure out whether you should (to quote Black Sheep – The Choice is Yours) “get with this or you can get with that.” Yes, old school rap is the inspiration for this segment. Ok, here’s are the mystery players.

SS     NFBC
PlayerAVGHRRRBISBADP
Player A0.2731380522577
Player B0.28811586024131
Player C0.2737574727206
Player D0.25119676111230
Player E0.2599615120236

Based on the projections, Player A looks to be the most productive in terms of run production, has good speed and a little bit of power. The average is fine, but I’ll be honest, I don’t understand why this player’s ADP is up at 77 based on the projections. The 80 runs and 52 RBI tells me this player hits near the top of the lineup. That’s true, this player is slotted to hit second in a solid AL lineup with a couple of notorious veteran run producers hitting behind him. Yup, this is Mean Jean Segura. I actually agree with the power and speed production but can’t understand the low runs and RBI numbers given a full season of at-bats, especially with Cano and Cruz behind him and Dee Gordon in front. The .273 average appears a alittle low as well, Jean has hit .309 the last two years combined and will be 28 this year. His BABIP is high but matches his good speed, high GB% profile. Despite the low projections by ZIPS, I’d expect a .290 AVG with 90+ runs and around 60 RBI along with the power and speed. He’s a clone of Elvis Andrus 20 spots later, GET WITH THIS.

Player B looks a lot like Segura but with very low run production. Interesting, a player like this should hit near the top of the lineup and not at the bottom. The issue appears to be his almost non-existent walk rate (3.7% in 2017) and utility-type role. I may have given him away but if I didn’t he’s also only eligible at shortstop if your league requires 15 or fewer games played at a position. Yup, this veteran is Eduardo Nunez. I would love Nunez more if he wasn’t on Boston. I know that sounds odd because Boston has a stout AF lineup. Normally that’s great for production but Nunez will hit 8th or 9th when he plays. He’ll start at 2nd while Dustin Pedroia is out then fill in as a utility role player. So, 500 PA is going to be tough to get. The other issue is the fact that Boston doesn’t run much. If he was in San Diego for instance, I’d love him because he would start every day and hit first or second in the lineup while running wild. While I’d take a shot on him because Pedroia is going to have trouble staying healthy, he needs another position player to get injured for him to see nearly a full season of ABs. DON’T GET WITH THIS. Unless he slips into the 160-170 range.

Player C
Wow, more steals and even LESS run production than Nunez! He still provides a solid batting average but isn’t taken until after pick 200. This player’s SHIT-uation doesn’t look great. He looks to be slotted near the bottom of the lineup and could possibly be on the strong side of a platoon. However, this 23-year-old NL player clearly has some speed considering the playing time issue. Player C is Jose Peraza, everyone’s favorite speedster sleeper from 2017. Now, everyone is hating even though he’s been handed a starting gig. I understand the risk especially with the news that Nick Senzel is taking reps over at short. While I love Senzel, he isn’t a shortstop and I believe it will take him the first half of the season in the minors to get comfortable with the new position. At that point, Peraza would be a utility player at 2B, SS, and OF. He’s still a good bet to get 500 PA. The high contact rate and speed should give him a .270-ish batting average but you won’t get much power. So if you need speed around pick 200 GET WITH THIS. I don’t love him before pick 200 though.

Player D
Player D is my beau. If you haven’t read a lot of my stuff or follow me on Twitter (plug) you may not know who this guy is. Clearly he the most power out of this group but the least amount of speed. His run production is second to Segura but I actually think it’s a joke. My projections for this AL LEAD-OFF hitter are 22 HR, 14 SB with 80+ runs and 60 RBI. Player D is Marcus Semien. How ZIPS projects 67 runs from a lead-off hitter, I have no idea. Also, Semien stole 12 bases (yes it was a career high) last year but in less than 90 games and is now projected for only 11? The power is legit as well, his full-season projection for home runs last year was 19 BUT he missed significant time with a wrist injury which sapped his power upon his return in July and August. It’s all good though because he clubbed five HR in September to prove the wrist is no longer an issue. Did I mention that he’ll be 27 this year and already has a season where he hit 27 home runs? I didn’t? Oh, well he does and he’s being drafted as the 20th ranked shortstop. GET WITH THIS ALL DAY!

Player E
Being draft six picks behind my boy is Player E. The numbers here aren’t too bad as projected by ZIPS. He’s one HR shy of a 10-20 season with a near .260 BA. This is an NL SS with elite talent defensively and well above average speed. He basically going to be given the job because of his defense but needs some work offensively and there are veterans there to take his spot if he falters. Player E is Amed Rosario, previously the Mets’ number one prospect in 2017. Rosario had an awful debut hitting .248 with an atrocious .271 OBP and K-BB rate of 27%! Yes, you read that right. After last season, I did not like Rosario coming into 2018; I worry about the plate discipline and I’m not sure the power shows up this year. He also needs to get on base to steal and he will be batting at the bottom of the lineup, so he won’t have a ton of opportunities. I do think he will be a solid hitter in a few years something like .275-15-25, but not now. DON’T GET WITH THIS!

Don’t Sleep on Marcus Semien

Tags: Marcus SemienJorge Mateo, Sleeper, 2018 Projections

Marcus Semien is often overlooked in fantasy baseball.  With the young talented crop of shortstops, it’s easy to see why.  It might seem like Semien has been around a while and that’s true.  He got his first taste of the bigs with the White Sox in 2013 but didn’t become a full time player until he went to Oakland in 2015.  Since then he’s played in 399 games where he has hit 52 HRs and stolen 33 bases.  I’ll do the math for you. That averages out to 21 HRs and 13 steals per 162.  That doesn’t sound all that bad for a shortstop going around pick 225.  I actually think he’ll surpass those numbers in 2018 and here’s why.

First off, remember how I said it seemed like he’s been around for a while?  He’s ONLY going to be 27 in 2018.  So he’s right in the middle of his prime.  He did only hit 10 HRs in 85 games last year but missed a lot of time due to a wrist injury.  That wrist injury sapped a lot of his power when he came back off the DL hitting only 1 HR in August but bounced back with 5 in September.  That was backed up by his increased fly ball% to 43% in the 2nd half and his hard hit% back up over 30%.  The wrist injury is no longer a concern going into 2018.  Oh and did I mention he hit 27 dingers in 2016?  So there’s clearly power in his bat.

That brings us to his approach.  His K rate hovers around 22% but in 2017 his walk rate jumped to a career high 9.8%!  That’s good, he’s being more selective not chasing pitches out of the zone as much and as a result making better contact.  The increase in walks seems like it will stick, he had shown patience in the White Sox minor league system so his OBP should be on the rise. This is necessary because he doesn’t hit for a high average partially due to the elevated FB%.  I do think he got a little unlucky with his BABIP which was .300 in 2017.  He’s got good speed and hits line drives at an over 20% clip.  I mention the speed and while his 12 steals were a career high in 2017, it was only in 85 games and he’s efficient AF on the bases stealing 22 bases in his last 25 attemptsThat’s like 88%!  I’m thinking more like 15-18 in 2018.

I know Jorge Mateo is waiting to take over and he’s one of the more exciting players in the minors right now.  He came over from the Yankees in the Sonny Gray deal.  He’s got Billy Hamilton speed with developing power!  Scary.  Semien though, is under control through 2020 with the A’s and I don’t see Mateo up in 2018 except maybe in September for a cup of coffee as he’s only played 60 games above high A ball.  So playing time shouldn’t be an issue for Semien who batted almost exclusively in the #1 or 2 spot in the lineup last year.  He’ll have to keep up the walk rate to secure one of those spots for all of 2018.  The good news is, the A’s don’t really have anyone else to hit atop the order.  So, here’s what we are looking at for 2018:

.260/.337 22 HRs, 15 steals, 90+ runs and 68 RBI.  He’s going around pick #225 in Early Mocks.