Statcast defines different types of batted balls based on launch angle (LA) http://m.mlb.com/glossary/statcast/launch-angle. They call under 10 degrees a grounder, 10-25 a liner, 25 to 50 a fly ball and over 50 a pop-up. Those are not new terms, of course, those definitions have existed forever. Merriam Webster uses this definition for a line drive https://www.merriam-webster.com/dictionary/line%20drive.

To evaluate the boundaries I first looked at some characteristics of batted ball types using the boundaries that Statcast uses: https://imgur.com/a/wlWbsNE

The categories were BABIP, ISO and BA (used BA instead of BABIP to include homers) dependency on EV. You can see that “grounders” under 10 degrees have a BABIP of around .280, a very low ISO and a steady positive relationship of EV and BA. Liners have a very high BABIP, a high ISO of .435 and a relatively low impact of different EVs. On fly balls (25-50 degrees) you have a very low BABIP, a very high ISO and you have the “donut hole” where you have the bloopers on very low EVs, mostly outs at medium EVs (80-95) and then again (extra base)-hits at high EVs.

To test the existing boundaries I now did the same tests with other boundaries.
LA characteristics

Under zero degrees the BABIP is mostly under .200 except for very hard hit balls where it is around .300. From 0 to 5 and 5 to 10 that changes, at low EVs the BA on contact is low and at medium and hard contact it gets pretty high (around .500 and higher). That means those batted balls around 5 degrees behave like grounders at low EVs and like liners at medium and high EVs.

At 20-25 degrees the BA on contact is .682 on soft contact, just .251 on medium contact and around .700 on hard contact which is about the same as fly balls. That means balls hit at those angles behave like a liner on soft contact and like a fly ball on medium and hard contact.

I also compared the range of 5-20 with the 10-25 range and an alternative range of 5-20 https://imgur.com/a/qjaVrIP
What you can see is that the 5-20 range matches more closely with the “core line drive range” of 10-20 than the 10-25 range.

Overall it is not totally clear what is better. Neither range is perfect as both edge ranges (5-10 and 20-25) are more velocity dependent than the “core range”. The lower edge behaves like a grounder on soft contact and the upper range behaves more like a fly ball on medium and hard contact showing the famous donut hole.

IMO there are a lot of reasons to either narrow liners to the 10-20 range or alternatively use 5-20 if you want to keep the same angle range because soft contact is only 20% of all MLB contact. That means on 80% of all contact 5-20 behaves like a liner and on 20% like a grounder. The 20-25 range, however, behaves like a fly-ball 80% of the time and just 20% like a line drive.

So the changed ranges would be either:

<10 GB
10-20 LD
20-50 FB

or:

<5 GB
5-20 LD
20-50 FB

You could also introduce additional batted ball types to make it more precise as Andrew Perpetua did here but I think the easier solution would be to cut off the upper 5 degrees of the original range because that range behaves like a FB on most batted balls.

In Part 1, I discussed what batter xwOBA does, what data feeds into it (using Statcast’s quality of contact types, including barrels), and thus what some sources of noise or “batted ball luck” are contained within xwOBA despite what it strips out in terms of defense and park impacts.

Some takeaways to set up Part 2:

• Barrel% (AKA barrels per plate appearance) appears to be in a similar range of year-to-year reliability as K% and BB% for batters, while other Statcast quality of contact categories that produce positive results had far worse year-to-year consistency.
• When comparing roughly the first third of a season to the rest of a season, it appears that barrels remain one of the more reliable metrics we examined, while “flares and burners,” which are worth a bit less than half the wOBA of a barrel on average, are very unreliable despite making up 25% of all batted balls. Thus, variation in the number of flares and burners a batter hits was identified as a likely source of noise that would still exist in xwOBA. (The same, to a lesser extent, could be said of “solid contact.”)

Now that we understand what feeds into xwOBA better, I want to look at the descriptive and predictive capabilities of xwOBA. To be clear, while xwOBA regresses results on batted balls based on exit velocity and launch angle, it is not a projection of future results / a predictive metric.

The “expected” element refers to what you would have expected a player’s past results to be. In this manner, it is similar to FIP, though FIP is much simpler. And, just like with FIP, you cannot simply look at a player’s FIP and then anticipate them replicating that going forward. However, as FIP is to ERA, theoretically xwOBA could be to wOBA – it considers elements potentially more indicative of skill while cutting out some noise, and thus could predict wOBA going forward better than actual past wOBA.

We should have something else to compare it to though, mostly for fun. We could compare it to past projections here at FanGraphs, but that sounds like a great deal of additional work for me that I would not know where to start with, so let’s conduct an experiment by producing a simple model that describes, but does not attempt to project, batter wOBA.

For Comparison to xwOBA: Our (Very Simple) Model

I constructed a quick linear model using all batter data from 2015-2017.

I saw, from Part 1, that three of the most reliable batter statistics we looked at were K%, BB%, and Barrel%. Therefore, I used batter K%, BB%, and Barrel% to describe wOBA in a linear model weighted by the number of plate appearances each batter had. (When we later test the predictive capabilities of Our Model, we will benefit from Statcast here by being able to look at barrels instead of home runs, as barrels appear to be more trustworthy than actual home runs.)

After rounding the coefficients, that gave me the following equation:

wOBA = 0.309 – 0.36*K% + 0.45*BB% + 1.24*Barrel%

or

wOBA = 0.309 + (-0.36*K + 0.45*BB + 1.24*Barrels)/PA

Unlike xwOBA, Our Model ignores the well over 90% of batted balls that are not barrels. Our Model also ignores the specifics of how each barrel is hit, unlike xwOBA. Not all barrels are created equal. For example, for barreled balls hit at a launch angle of 28 degrees, a 102 mph exit velocity has produced home runs on about 3 out of every 4 batted balls, while at 112 mph there have been 100% home runs.

These details undoubtedly matter for estimating past results – Our Model should be easily worse than xwOBA in that respect. But how will this impact predictive capabilities? Will Our Model’s lack of knowledge of what happens in about two-thirds of all batter plate appearances significantly worsen predictive qualities, or will it cut out the noise to the point that predictive qualities improve?

wOBA vs. xwOBA vs. Our Model

Naturally, to find out, let’s go to some tables. First, how do the models describe wOBA in a full season and at the 2-month level? (i.e. What is the R² between 2015 wOBA and 2015 xwOBA? Or between Apr-May 2015 wOBA and Apr-May 2015 for Our Model? And so on…)

R² of Models Describing wOBA – Full Year

R² of Models Describing wOBA – First Two Months of Season Only

^using batters with min. 300 batted balls for full years and 100 batted balls for two month periods.

Of course wOBA perfectly describes itself. No other model can beat that! As was assumed, xwOBA is clearly a tier above Our Model in terms of descriptive capabilities.

xwOBA loses to wOBA because, for example, xwOBA doesn’t know when the defense made or did not make a play; when a ball that might have cleared the fence on an average day was actually blown in by the wind and caught; or whether a lumbering lefty pulled yet another hard-hit grounder straight into the shift.

Our Model, in turn, loses to xwOBA, because it leaves out the same things as xwOBA plus it knows nothing about whether a liner, a pop up, or whatever else was hit on the vast majority of batted balls. Still, Our Model is not way less successful.

Finally, on to the most interesting part: predictive capabilities.

People have been comparing batter xwOBA to wOBA when discussing breakout or slumping hitters and whether or not they may continue to succeed or fail. To test the appropriateness of this, let’s see how well our three batting value models (wOBA, xwOBA, and Our Model) predict future batting value (future wOBA) on a “year-to-year” and a “pre-June 1st to June 1st onward” basis.

R² Between One Year of Models and the Following Year’s wOBA

^Same sample as in Part 1: Batters with min. 300 batted balls in both years being compared.

At the year-to-year level, none of these metrics are magic at predicting future wOBA. It is not clear from this fairly simplistic analysis whether one year’s wOBA or xwOBA will tell you more about the next year’s wOBA. Our Model may be the worst (well, it at least did a poor job 2016-2017).

R² Between Models Pre-June 1st and wOBA June 1st Onward

^Same sample as in Part 1: Batters with min. 100 batted balls before June 1st and 200 batted balls from June 1st onward.

In our smaller in-season sample, there is a difference. It appears using wOBA from the first two months of a season to predict rest of season wOBA is the worst idea out of the three.

It also appears that using xwOBA or Our Model from the first two months of a season to predict rest of season wOBA isn’t really any different, despite Our Model ignoring so much information! (I’m not going to say Our Model is better, because this is fairly imprecise analysis and the R² values are very similar.)

Conclusions

Similar to the lessons of FIP for pitchers, we can see how leaving out large amounts of data can be appropriate when you have not figured out how to use it effectively yet. Even though wOBA itself clearly benefits from feeling the impact of certain reliable things that are ignored by the other models we examined, such as a batter outperforming their quality of contact due to playing in a hitter’s park or being fast, xwOBA and Our Model cut out other elements that muddy the data in small samples to make up for missing that info.

However, neither xwOBA nor Our Model is built to be projections of future performance. I already linked to this Tom Tango tweet in Part 1, which says that the minimum condition to make a statistic predictive is to weight it by the number of trials, which for batters here we could use plate appearances. In a simple form, this would consist of a model that incrementally adjusts the expectations for a batter to be based more on their tracked performance and less on the league average rate as more data (i.e. plate appearances) for that batter become available.

One can see how you could go about using Statcast data to build a projection system for wOBA on batted balls. For example, one could project the rate of barrels hit based more on a batter’s past barrel rate than the league average rate even in a relatively small number of PA, while one would have to heavily regress the projected rate of flares and burners a batter would hit toward the league average rate.

We have a number of projection systems available at FanGraphs that are great and constantly updated. Using Statcast data is attractive, but it is all very new, so we need to wait a bit longer before we see a similar Statcast-based projection system. Also, we probably simply need more years of Statcast data before we can be too confident in any such projection system regardless.

If you want your batter analysis to benefit from Statcast data in the meantime, maybe check out how a batter’s barrels per plate appearance have changed. Have they gone from about average to well-above average? Their ability to hit for power may have legitimately changed. (Speaking of which, this Mookie Betts power surge is crazy. 2015 to 2017 Barrel% = 4.2%. This year through July 7th: 11.9%!!!)

Enjoy xwOBA and what it does, but be careful using it to adjust your future expectations for players without diving deeper or relying on the powerful information we already have.

Nolan Arenado has been one of the elite hitters in avoiding the ground in his career with a GB rate of just 36% which is well below the league average of around 44% during that time frame. Especially impressive is his pull LA on low pitches of 9.3 degrees vs the league of 3.6 degrees. Those are the pitches the league rolls over when it tries to pull it and he drives just straight through them and pulls them in the air without hooking or rolling over.

The question is how does he do that. First, he does have a slight uppercut through the zone like most good hitters but it is not an extreme upswing.

Overall his swing is pretty flat, maybe a 10-degree positive attack angle or so, there are definitely swings with more uppercut out there. Also, his posture is rather vertical in the front to back direction. He does tilt his upper body over the plate but he doesn’t lean back toward the catcher.

This is different from many big uppercut hitters. The swing is generally pretty perpendicular to the spine thus the tilt over the plate changes the bat angle and the lean toward the catcher creates more uppercut in the plane as the natural direction faces up while a guy using just lean over the plate will have the bat going flatter and then up in the end out front compared to flatter barrel guys who will have the swing often getting flatter in the end when they roll over. In this picture, you see him vs Bellinger. Bellinger leans back much more and thus has a natural built-in lift. However, you can also see that Cody’s bat angle is flatter and his bat is already starting to roll over here. This might be why Cody- while an elite launch angle guy has a low pitch pull LA of 6.6 vs 9.3 for Nolan, who doesn’t have as much uppercut but is better in avoiding the rollover on low pitches even if he is fooled and out front.

View post on imgur.com

So how does he still create elite lift rates? One thing he does is having a very steep almost Ferris Wheel like bat angle. Even on high pitches his bat is pointing down and he swings more under the shoulders rather than around them.

Most other hitters will flatten the bat out more on high pitches like Pujols on a similarly high pitch

Here is another comparison Arenado vs Beltre on a pitch away and slightly above the belt

View post on imgur.com

Beltre also has some shoulder tilt but the bat and shoulders rotate on a much more level plane while Nolan has a lot of side bend in the spine and has the hands extremely high with the barrel pointing down (you can’t even see his face) while Adrian has the hands about lower chest high and the barrel just under the hands.

Because of this Arenado has a very straight direction through the ball and almost never rolls over. At the end of the swing the bat of every hitter will roll over to the other shoulder and if you hit balls out front there is a chance that you catch the ball during that rollover. That is the reason why pulled balls are hit on the ground more often https://www.fangraphs.com/community/the-effect-of-batted-ball-direction-on-launch-angle/ the rolling over creates a top spin.

Arenado due to his steep bat angle, however, delays that roll over extremely long. In this picture you can see that he almost is at full extension and the barrel is still below his hands and from the front, you can also see it still slightly points toward the other batters box, so it hasn’t started to roll over yet. So Arenado can be very out front and still not roll over, even in some swings where he loses his posture and lunges.

I wrote in this article how this is an important skill that holds some hitters back on low pitches
https://www.fangraphs.com/community/finding-keys-to-elevate-the-ball-more/

So Arenado does have a slight uppercut but the thing that makes him elite is that he rarely rolls over as his bat comes straight through the zone from below and not across the ball. He really gets the most out of his attack angle by rarely rolling over and across the ball but driving through it and either hit it straight or backspin instead of topspin.

There is a slight cost of this of course, on very high pitches this Ferris Wheel bat path is hard to do. Arenado does have a slight weakness very up in the zone https://www.fangraphs.com/zonegrid.aspx?playerid=9777&position=3B&ss=2018-01-01&se=2018-12-01&type=5&hand=all&count=all&blur=1&grid=10&view=bat&pitch=&season=2018&data=pi
However the first video of the article shows that he is able to pull this off until about belt high, so there is not much room for the pitcher up.

Overall this is an interesting and slightly unusual swing with some great strengths and weaknesses mitigated by great flexibility (especially in the spine) and the ability to contort himself to still get to the high pitch with a steep bat angle. This swing allows him to lift low pitches and make contact way out front without rolling over what most can’t do. There is a small space to attack him up in the zone but the margin for error is not high.

Overall, of course, we know that Arenado is an elite hitter. While I would not recommend his style for pitches belt high up it is definitely interesting how he refuses to roll over baseballs and drives them in the air consistently especially against lower pitches.

I was born, raised and am currently residing in South Korea, where Chan Ho Park, Shin-Soo Choo, and some Korean-born Major League players spent their school days before moving to the US with a big dream. Many Korean MLB fans recognize that their free agent contracts with the Texas Rangers have been something that very far from what we call a success.  Choo is currently having a great year and has been rewarded with his first trip to the All-Star Game.

Spending more than a decade of time to observe the ongoing conversations about those free-agent contracts, I have been wondering if it is possible to view this matter rather mathematically and/or statistically. And I think I have found a way for it, if it is not an absolute solution. I have named it proW/sal, which stands for proportional fWAR divided by proportional salary.

The equation is rather simple: proW/sal = (fWAR of the selected player in a particular year/fWAR of the player whose fWAR was the highest in that year)/(salary of the selected player in a particular year/salary of the most expensive player of that year).

The following are the exemplary applications:

1. This can be used in comparing individual players’ performances in one year in their FA contracts. Alex Rodriguez won the AL MVP award in 2007 after having a monster season with 54 homers and 156 RBIs, batting stellar .314, .422, .645. This enabled him to sign another 10-year FA contract with the Yankees. Let’s compare his year to that of Albert Pujols, who was in the middle of the extension contract with the St. Louis Cardinals.

A-Rod recorded the highest fWAR of the year: 9.6. The most expensive player of the year was Jason Giambi, whose salary of the year was $23,428,571. Since A-Rod’s salary for that year was $22,708,525, his proW/sal would be (A-Rod’s fWAR in 2007/A-Rod’s fWAR in 2007)/(A-Rod’s salary in 2007/Giambi’s salary in 2007) = (9.6/9.6)/($22,708,525/$23,428,571), which would be rounded to about 1.07. This is a great number because this means A-Rod’s fWAR and salary made an equilibrium with the highest fWAR and the highest salary of the year.

Let’s look at Pujols in 2007. His fWAR was 7.7, which is amazing, but not quite like that of A-Rod. His salary, however, was only $12,937,813. So his proW/sal of the year would be (Pujols’ fWAR in 2007/A-Rod’s fWAR in 2007)/(Pujols’ salary in 2007/Giambi’s salary in 2007) = (7.7/9.6)/($12,937,813/$23,428,571), which is about 1.45.

Yes. Pujols’ fWAR was lower than that of A-Rod. But due to his low salary, Pujols can be considered to have had a way more number-efficient year than that of A-Rod.

2. This also can be used to determine the efficiency of an individual FA contract in a year or as a whole. Let’s look at Randy Johnson during his first six years as a Diamondback.

Johnson’s FA contract with the Arizona Diamondbacks in December 1998 is considered one of the most successive FA contracts ever. But how good was it? Let’s determine it by using proW/sal.

In 1999, Johnson’s first year as a Diamondback, the player who recorded the highest fWAR was Pedro Martinez(11.6). Randy was at 9.5. And the most expensive player of that year was Albert Belle, whose salary was $12,868,670, while that of Randy was only $9.7M. So his proW/sal in 1999 was (9.5/11.6)/($9.7M/$12,868,670), which is about 1.09.

In the same manner, Johnson’s proW/sal in 2000, 2001, 2002, 2003, and 2004 can be calculated. According to my calculation, they were 1.18, 1.65, 1.44, 0.43, and 1.41, respectively. The big jump was due to A-Rod’s record-breaking FA contract with the Rangers, as well as Johnson’s great performances.

So the average of those six years’ proW/sal would demonstrate the efficiency of Johnson’s first six years in Phoenix. It’s at 1.19–a huge number because this means he was either the best in fWAR in those years, while his salary was quite below that of the most expensive player of each year.

For comparison, the average proW/sal of Carlos Lee’s FA contract with the Houston Astros was 0.29, and that of Choo’s FA contract with the Rangers was 0.26 until last year. I would not even bother with Park’s contract with the Rangers.

This equation, however, is far from perfection. First, I have not considered about distinction between fWARs of pitchers and batters, as well as salaries of pitchers and batters. I wonder if it would be necessary to differentiate them so that the number becomes fairer.

In addition, this equation is yet to function properly in regards to the fWARs in negative numbers. Using this equation may improperly evaluate a player with a negative fWAR with a cheap salary. For example, let’s suppose a player, A, whose fWAR was -1.0, and whose salary was $1M, another player, B, with the same fWAR but $10M of salary. And let’s assume that the highest fWAR that year was 1.0, and the most expensive salary was $10M. Despite the same fWAR and the lower salary, A’s proW/sal would be -10, while that of B is only -1. I have not found a solution to this yet due to my lack of proficiency in mathematics and statistics.

Moreover, this equation only functions well within the comparison between/among FA contracts. If this gets to rookie players, the value skyrockets  to an absurd degree. For example, the proW/sal of Mike Trout in his Rookie-of-the-Year year of 2012 is about 569.63. Every FA contract will be viewed as a bust if it is compared to that.

So here it is. I hope people with much proficiency take a look at this and improve it.

Before the start of the season, back in late March, I wrote an article on three Padres relievers who I thought had a chance to form a triumvirate worthy of comparison to that of the Yankees’ 2016 trio of Chapman, Miller, and Betances. While I was wrong about Kazuhisa Makita, two of those three pitchers have indeed turned in impressive campaigns, and their fellow teammates in the ‘pen for San Diego must be commended for being quite good as well. The Padres bullpen has made what is otherwise a pretty lackluster team, a lot more interesting.

With a variety of different looks and approaches to getting hitters out in the late innings, this bullpen has performed really well this season. Their FIP, or fielding independent pitching, ranks fifth in the league at 3.47. This is a metric that is on the same scale as ERA, and can be interpreted in the same way. What makes it more useful than ERA is the fact that it tells you what a pitcher’s ERA would be if he had average luck on batted balls. The idea is that some pitchers have a higher or lower ERA because of their good or bad defense, or luck, etc.

Going back to the Padres — The team is rebuilding, and one would think GM A.J. Preller would be open to discussing a trade for any of his relievers. Though there are a number of questions he’s going to have to answer before making any moves, because the Padres are in an interesting situation at the moment. Is the team close enough to contention, to justify trading valuable relievers under team control for multiple years beyond this season? Does it make sense to trade for a prospect who wouldn’t necessarily be ready to contribute to the major league team in 2019? These questions will clearly have to be addressed, yet the most intriguing part about the whole situation is envisioning the return San Diego could get in return for some of their prized bullpen arms.

The price of relievers on the free-agent market has continued to rise in the last few years, so a team could see trading for relievers under team control for multiple years — as a cheaper alternative to shelling out money for older relievers in free agency. If a team had the chance to give up a big prospect but knew they were getting, say, three quality relievers in their prime from the Padres for the next three years, wouldn’t that be better than breaking the bank to sign a veteran this offseason? It would at least be a strong alternative solution to the problem of having a mediocre bullpen.

Furthermore, the Padres could actually trade multiple hurlers in their bullpen to try and get a significant haul back for them in a trade. Of course any trade of this nature, potentially involving one of the “super-teams” in the league — Could end up having significant ramifications throughout the final stretch of the season and in the playoffs. These are the likely trade candidates for the Padres:

Their resumes above basically speak for themselves.

It seems like the only player the Padres would be less interested in trading would be Brad Hand, considering that he’s signed to a team-friendly long-term deal and has become the anchor of their bullpen. The question then becomes, what teams should be looking for relief help? The teams that come to mind are the Nationals, Phillies, Cardinals, and Indians. The depth of quality pitchers in the San Diego ‘pen should allow the team to possibly make a trade for a true impact player if they include multiple hurlers in a deal.

More likely to make a trade with the Padres than the aforementioned teams, are the Nationals. Seeking to maximize the last guaranteed season with Bryce Harper around, the franchise probably has very legitimate motivation to go all-in on trying to win a championship. Max Scherzer and Stephen Strasburg won’t be in their respective primes forever, which is another reason for the team to really go for it now! The Nationals seem like a team that would make sense to trade with.

The Nationals don’t have a very deep system, and their top prospects would likely be untouchable. That doesn’t mean they couldn’t give the Padres a number of decent prospects, though. A package centering around Carter Kieboom, and perhaps Seth Romero or Yasel Antuna could be had by San Diego. A deal with DC would likely be more about quantity over quality in prospects received by the Padres. Other teams, however, could send fewer, yet more individually talented prospects.

If the Cardinals feel like they have a chance at the Wild Card, they could realistically trade for some relief help to use in the playoffs. Their system isn’t very impressive, but Adolis Garcia and Conner Greene could net them one of the Padres pitchers.

The Phillies’ bullpen has been fine so far, though they just demoted their closer Hector Neris to Triple-A. It would seem like they’d want to feel more secure about their bullpen, so maybe the Padres could trade a few of their relievers for blue-chip prospect Sixto Sanchez. More likely, they could take Adonis Medina, Franklyn Kilome, and buy low on hometown outfielder Mickey Moniak. Those three latter prospects would be a nice return even if Sanchez is unavailable.

The Indians have a horrendous bullpen situation, so they would be the most obvious candidate as a team interested in making a deal with A.J. Preller. Getting multiple talented relievers in one trade likely seems very attractive to the Indians at this point in time. While Francisco Mejia is likely off limits in trades, it would be reasonable for the Padres to get some combination of Shane Bieber, Nolan Jones, and Will Benson in return for some of their relief aces.

All of the pitchers being dealt are under team control beyond this season, too:

The years of control on all the pitchers are what should excite contending teams —  As much as their talents likely already do. The Padres are still rebuilding, and they have plenty of arms in the system to fill out this bullpen in the years to come.

Not all of the starting pitchers in the system will pan out completely, so expecting some of them to be in the bullpen is realistic. While it is tempting to say that the team should hold onto their star relievers, the team simply isn’t close enough to contention for that to be a viable excuse to keep them around. Add on the fact that reliever performance is volatile, and you can see that trading some of these guys now would make sense while their value is so high.

Considering the number of quality relievers the Padres possess, it would be surprising if none of them were traded at the deadline in the coming weeks. Hopefully, for the Padres, they’ll be able to continue to build for the future in exchange for some of the quality bullpen arms they have.

This post was originally posted on https://fathomablefriarfactsresearch.wordpress.com — Check out the site for more Padres related posts.

Short answer: No.

Slightly longer answer: No idiot – they won the World Series.

Much longer answer: No. But he may have cost them a game. And with one more win the team would have tied the franchise record for wins. Which, in light of winning the World Series, probably doesn’t matter at all. But, as an exploration, let’s pretend this game really, really mattered.

It was just more than one year ago, on a night just like tonight. July 17 in Houston, Texas. Astros hosting the Mariners. Bottom of the ninth. Two outs. Score tied. Bases loaded. Alex Bregman at the plate. He smacks a hard-hit ground ball to the left side that gets past a diving Kyle Seager at third, but shortstop Jean Segura snags the ball in the hole and his strong throw just beats a sliding Brian McCann for the force out at second. (Seager would lead off the 10th with a long ball, and the Mariners would win in 10.)

Slick-fielding play by Segura, bang bang result. So why put the blame on McCann? Let’s go to the tape.

The throw beats McCann in by a split second. But, revolutionary suggestion alert, what if instead of slowing down to slide, McCann had simply sprinted right through second base? Without slowing down to slide, he would have beaten the ball to the bag, avoiding the force out. Sure, he’d get tagged out soon after, but the game-winning run already would have scored from third. GAME OVER – ASTROS WIN! Running through second base sounds counterintuitive to baseball logic, but, had McCann beaten the force out, Yuli Gurriel would already have scored the game winning run and McCann getting tagged out would be irrelevant. Instead of a -.16 WPA play, the outcome would be a .34 WPA play that ended with a W.

Which begs the larger question – why don’t we ever see runners run through second base? Turns out, it’s because this is a very situational opportunity. For running through second to make sense, a team would need a runner on third, a runner on first, (an optional runner on second), two outs, a ground ball, and a close force out at second base. In the 2017 regular season, a Baseball Savant search reveals that there were only 252 ground balls in that situation that ended with an out at second or third.

So, in theory, 252 potential chances for a runner to overrun the bag at second (or third), to get an extra run out of the inning. Granted, there is an increased risk of collision with a middle infielder covering the bag, or even injury from the unfamiliar act of straight-line sprinting past second before hitting the brakes. So maybe this base running maneuver is only worth it to bring home the tying or go-ahead run. Which brings us to roughly 54 chances in the season. Of those, Baseball Savant has video for 20. And of those 20, it turns out that on only three of those plays (all in July) were close enough that the runner would have potentially benefited from running through second base to allow the tying or go-ahead run to score.

Once again, let’s head to the tape:

Lucas Duda:

Austin Barnes:

These examples have two things in common: Both are early in the game (third and first inning, respectively) and both would involve the baserunner colliding with the second baseman at full speed. Neither seems worth the chance of injury to score an early run. And it’s unclear on how an umpire would call that type of collision.

That leaves us with one example from the 2017 season, where running through the bag would have won the game while reasonably avoided a collision. Let’s watch it again.

Brian McCann:

The ball beats McCann to the bag by a fraction of a second. It’s a fraction of a second he could have avoided had he just kept running.

In conclusion: at times, running through second base instead of slowing down to slide would be a smart move. These times just appear to be very rare. Out of 188,074 plate appearances in the 2017 regular season, only 1,979 (or just over 1%) had the necessary requirements for a base runner on first to be thinking of running through second base on a ground ball. Only 252 (or .13%) resulted in a ground ball force out to second or third. And only 1 (.0005%) would have materially changed the game.

So, for everyone else in the big leagues, you probably don’t have to think about running through second. But for Brian McCann, by not being alert to the situation, you cost your team from tying the franchise record in wins. (Since he did catch every inning in October while helping bring home the Astros first World Series title, I guess we can give McCann a pass on this one).

If you’re following the Atlanta Braves this season — and it would be hard not to, as they’ve lead the NL East for a large portion of the first hal — there’s a lot that may draw your eye. Ozzie Albies, Ronald Acuña have provided anticipation. Nick Markakis has surprised. Freddie Freeman has been himself. A host of pitchers, like Mike Foltynewicz, Sean Newcomb, Mike Soroka, Shane Carle, AJ Minter, and Dan Winkler, have all emerged as more than expected in some respect. But another name should also grab your attention: 24-year-old, switch-hitting Johan Camargo.

The Atlanta system has been among the best in baseball the last couple years, boasting both depth and top end talent. The litany of players above largely verifies that. Two years ago, the last time Camargo was eligible to be on a prospect list, he was effectively ranked as the 52nd-best prospect in the team’s system by FanGraphs. He was said to be “a plus defender at third” but also that “his feel for hitting and lack of balance at the plate are both non-starters.” He was ultimately compared to Abraham Nunez.

While Nunez enjoyed a long professional career, he also retired being worse than a replacement level player. His career fWAR was -1.4. Upon arriving in the Majors last year, Camargo seemed to immediately dispel any such comparison. His defense between shortstop and third base was passable, but his bat was more than anyone ever seemed to imagine. He mustered a 102 wRC+ in 82 games, which was 14% better than Nunez ever achieved.

Camargo performed that way largely on the tails of a .368 average on balls in play. Sustainable? Probably not, but it was something, and way more than what was ever expected of him. That’s already a win for a team’s 52nd-best prospect. But this year he’s gone from something to something to write home about.

All of his numbers so far jump off the chart. Last season he whiffed five times more than he walked. This year he’s walking an additional 8% and striking out less. He’s driving the ball at a clip that’s 33% higher than last year. He’s been 15% better than the average Major League hitter, and that’s with his average on balls in play dropping more than 80 points! That’s fantastic! So for the second time in as many years, Johan Camargo is forcing us to beg the question: is he for real?

Well, dang. His walk rate skyrocketing seems legitimate with how much less he’s swinging at balls out of the zone. Spitting on offerings that are inherently less hittable will influence the rest of his batted ball profile, too. He’s traded in weaker contact for harder contact. Hard contact throughout the league is up by nearly 4% from last year. That’s substantial because the amount of balls in play is in the thousands — think of it like getting a 4% raise in a single year, compared to, say, 1.5% for cost of living. Alex Chamberlain recently examined how it’s meant less overall, but this much is clear: Camargo is still knocking the crap out of the ball.

This authority has lead to improved exit velocity on line drives and fly balls. Last year, on average, Camargo hit balls in the air at 91.4 mph. This year he’s doing it at 93.8 mph. The tick and a half might not seem like much but it moves him from the 23rd percentile in all the Majors to the 76th. And considering his average launch angle on those balls in play — 25.4 degrees — it’s significant. Rob Arthur has found that “the very best hitters in MLB tend to smack lots of balls with launch angles around 25 degrees and exit velocities above 90 miles per hour,” and so far Camargo is only trending upward.

We might be able to contribute this next gear at the plate from Camargo to a more exaggerated leg kick. See below for yourself.

On the left is Camargo in 2017 when he first showed us he might be more than we thought. On the right is him in 2018, as he insists that he is. Leg kicks like this are timing mechanisms players use to establish rhythm at the dish. His teammate Ozzie Albies, who is also a switch hitter blasting by his projections, employs a similarly pronounced leg kick. Camargo seems to have found one that does the job for him, providing him the balance and feel at the plate he lacked as a minor leaguer.

Maybe we’d have heard more about Camargo by now if he was on a different team, or if Atlanta hadn’t surged to contention so quickly. Maybe it’s tougher to see how far he’s come given that he started so far off everyone’s radar, or that he’s supposed to be a utility man and placeholder for prospect Austin Riley. But Johan Camargo is more than any of that, and he’s showing us how.

—

Exit velocity, launch angles, and stills from Baseball Savant. All other data from FanGraphs.

Zack Wheeler has been waiting years for an article to be written with this headline. Despite Tommy John Surgery’s increasing success rate, Wheeler is evidence of the still lingering uncertainty revolving around the procedure. Once one of the bright up-and-coming flamethrowers in the game, Wheeler was demoralized by the devastating diagnosis that he had a torn ulnar collateral ligament in March 2015, mandating repair by surgery. The typical recovery timeline issued to pitchers who receive Tommy John is a 12-16 month period, all encompassing the grueling process of post-surgery rest, physical therapy strengthening, and an interval throwing program that spans many months, finally followed by a return-to-mound program.  Wheeler attempted to make his comeback to professional baseball in August 2016, slightly more than 16-months post-surgery, but he was shut down after just 17 pitches in his first rehab start. It wasn’t until April 7, 2017, more than two years after his surgery, that Wheeler returned to an MLB mound.

Wheeler’s 2017 season could be considered a success from a health perspective, as he was able to hurl 86.1 IP in his 17 starts, but it was clear the right-hander was still figuring things out from a performance perspective. The final line on the former top prospect’s season was an ugly 5.13 ERA/5.03 FIP, amounting to a dreadful 19.5 % HR/FB rate, while his 4.17 BB/9 made it evident that his control was lacking. Wheeler’s average velocity had plummeted from its pre-surgery readings, from the 96.2 MPH he had averaged with his heater before surgery, to 95.4 MPH last season. It may seem like a 0.8 MPH drop-off should lead to a negligible difference in on-field performance, but the missing extra life on all of Wheeler’s pitches resulted in a season to forget. Wheeler’s peripherals were in agreement with the high ERA he boasted, so this was surely discouraging to everyone in the Mets organization.

Wheeler’s 2018 has been a completely different narrative. While he seemed to be getting acclimated in the first few months of the season, Wheeler has turned it up to full throttle since late May, as his average fastball velocity has accelerated to 96.3 MPH in his 9 GS in that span, from the 94.5 MPH it averaged in his first 7 GS. In Wheeler’s first 7 starts this season, his fastball’s average velocity was capped out at 95.2 MPH. Wheeler’s revival as one of the hardest throwing starting pitchers in the game began when he took the bump against Miami on May 22. Turning his fastball up to average 96.1 MPH, Wheeler punched out 9 Marlins. Five days later against Milwaukee, he sustained the extra juice on his fastball, averaging 95.9 MPH. However, the season and future still looked menacing for Zack Wheeler. His 5.40 ERA insinuated he was doomed for another dreadful season. Largely the result of an unsettling .331 BABIP induced and a diminutive 66% strand rate, Wheeler carried a far more promising 4.03 FIP/3.88 xFIP. Regardless, Zack Wheeler has been a different pitcher since the calendar flipped over to June.

On June 1, Wheeler’s fastball was zipping out of his hand even faster, clocking in at an average of 96.7 MPH, quicker than any outing since his big-league debut in 2013. The heat has been status quo for the lanky righty ever since; all signs point to Wheeler’s return to his pre-injury form that once inspired a momentous Mets-fanbase-sized wave of hype.

The plot directly below exhibits the velocity of Zack Wheeler’s fastball by month this season, and confirms that as Wheeler received his sudden boost in velocity, his ERA and FIP have both plummeted. While Wheeler’s 2018 season started on a shaky note, he has righted the ship in recent starts, to the point that he should now be an attractive asset to teams at the trade deadline.

While Wheeler has seen the largest velocity gains on his fastball since the season began, his split-finger, slider, and curveball have all marginally gained steam too. While latter changes are less noticeable to the naked eye, due to the physical nature of these offerings being slower, they are still meaningful. However, just saying that a pitcher gained velocity in isolation is meaningless. MLB hitters would prefer a pitcher throw harder if it signifies they are sacrificing some control, movement, or both.

In addition to being the beneficiary of a sudden velocity boost, Wheeler appears to have successfully reformed his approach to attacking hitters. The transition began in May, when Wheeler revived his previously abandoned sinker and started to decrease his fastball usage drastically. In April, Wheeler threw 203 fastballs, and hitters clocked this offering for a .439 xwOBA, with just a 20.6 % whiff rate (Per Baseball Savant). Wheeler’s fastball was the most prevalent pitch in his arsenal, at 56.5 %, which was a welcomed sight to batters. It was more of the same for Wheeler, as he threw 57.1 % fastballs in May. Throwing harder than ever before, Wheeler has dropped off his fastball usage in subsequent months, at a 41.3 % clip in June, and an even less frequent 37.3% rate through 2 July starts. Wheeler’s 3.44 ERA since June 1 shouldn’t blow anyone away, but it’s a vast improvement from his performance in the first 9 starts he made. His 3.39 FIP in this span indicates that the performance is legitimate.

As far as pitching goes, one of the skills that exhibits the greatest correlation from year-to-year is the ability to generate swings and misses. Wheeler has induced a Swinging Strike rate 0f 11.8% over 8 GS since June 1, higher than the 10.3% he averaged over the first 9 stars. As Al Melchior deliberated in his recent piece titled “Can We Count on Strikeouts From Mike Foltynewicz?,” there is a strong correlation between swinging strike rates and strikeout rates, with the former accounting for 69 % of the variance of the latter.   Despite this heavily-correlated relationship, Wheeler’s punch out numbers are yet to follow, as evidenced by his decline from the 23.6 % rate in the first 9 starts to 22.7 % since June 1. The correlation equation from the graph below suggests that Wheeler’s swinging strike rates since June 1 should entail a strikeout rate about 1.78 percentage points higher. With the increase in velocity accompanied, it shouldn’t be a shock that the fastball has been one of the main root causes of this increase in swinging strike rate. While averaging just 24.1 % between April and May, Wheeler’s fastball whiff rate has skyrocketed to 26.6 % in June, then again to 28.4 % in July.

As we near crunch time in the trade acquisition season, Zack Wheeler is an underrated asset who is hitting his stride at the perfect time. The month of June produced a mere .296 xwOBA for Wheeler, the lowest it’s been in any month since his return. With the disastrous 2017 season and a Tommy John surgery in his past, Wheeler is a cheaper alternative to other likely deadline trade candidates, such as Michael Fulmer, J.A. Happ, and Cole Hamels, and comes with 1 1/2 years of control. With the sudden return of his heat, Zack Wheeler is pitching as well as ever, and should provide great value to any team that acquires his services at the trade deadline.

(All Data from FanGraphs and BaseballSavant)

We’re two weeks out from the trade deadline. It may be quiet for most of baseball, given the state of the Haves and Have-nots shaping a less traditional mid-season urgency than in the past. Most of the AL playoff picture appears to be nearly set, at least to many observers. Meanwhile, the NL is up for grabs. As of July 14, the Phillies hold the biggest divisional lead at just 1.5 games over Atlanta, while the Dodgers have only a half game lead over the DBacks and the Cubs are in a dead heat with the Brewers. Manny Machado is the trade deadline’s biggest fish and he’s been connected to nearly all of those teams.

Given the state of competition in the NL, Machado could dramatically impact the league’s playoff race. He’s projected to be worth at least two more wins. That’s a bigger gap than any current divisional lead. It could be easy to argue that he’s a critical addition for any club, but it may not be more important for anyone than the Phillies.

Of course, there’s the short term considerations for the Phillies to acquire Machado. The team is competing earlier than anticipated. Their top tier farm system could handle the cost of acquiring a star on an expiring contract and still be excellent. It doesn’t hurt when the star in this case has intense connections to the current Phillies front office, from its director of scouting to its general manager to its president. But then there’s this:

That’s every first and second place team in the NL right now. The Phillies have had some terrible shortstop production in 2018. That could be because their expected starter, JP Crawford, has only managed to appear in 34 games this year, of which only 25 have come at short. The team’s primary replacement has been Scott Kingery, who’s appeared at short in 68 games. He was bally-hooed in Spring Training as he pushed for a roster spot and was signed to a long-term extension to accommodate him making the team, but he’s been miserable in his Major League debut. He’s mustered a 66 wRC+. In other words, he’s been 34% worse than average.

Beyond just being an upgrade at shortstop, Machado could help the Phillies become a more efficient offense overall. To date, they’ve left 654 runners on base, which is 11th-worst in the Majors. But they’ve also share the league’s 10th-highest OBP at .320. So they’re one of the best teams at getting guys on base, and one of the worst at driving them in. Machado has a wRC+ of 131 with men on base, and that may be a bit muted because Baltimore has been so bad. He’s garnered 11 intentional walks in those situations this year, which is already two more than he’s ever had in a full season.

Trading for Machado does more than just improve the Phillies and their chances this year, too. It keeps him away from every other team that would stand to get better by acquiring him. Maybe you read that and thought, “duh.” But if you notice in the chart above, the Brewers may especially feel the urgency to make a big move. They’re the only contender which has been worse at shortstop than the Phillies. They’re also trying to stave off the Cubs, who everyone seems to be waiting to click again and run off with the division, just like last year.

Long-term, Machado serves additional purpose for Philadelphia if they can sign him to an extension, which they may stand a good chance to do. Atlanta’s top tier farm system has put them in position to churn out role players and superstars with staying power. Even if the Nationals lose Bryce Harper this winter, they still have Juan Soto and the rest of the cast that’s good enough to compete. The Phillies system has produced talented Major League pieces the last couple years and is still ranked highly, but it lacks players who are projected to be stars on the level of the other teams in the NL East.

Acquiring Machado now is a move the Phillies can make with confidence because of how it impacts the present and scales for later. The iron is hot. They should strike.

—

LOB data from Baseball Reference. All other data from FanGraphs.

A few years ago, I created a distance model to evaluate unexpected distance. The purpose was primarily to evaluate spin on hit balls but there seems to be a lot more interest in juiced balls and home runs so here we are again.

When I recently updated everything, the results were shocking. If the ball was “juiced” in the second half of 2015 and remained so until last season, then in 2018, it has been extra de-juiced. (Actually, the recent study concluded it wasn’t “juiced” but rather, the added distance was the result of an unexplained reduction in drag).

I present the distance model and method at the end of this post since I believe the recent data and results are far more interesting.  Given the reduction in unexplained distance at all but one stadium, the magnitude of the change is mind boggling. For the past few years, results for the full year have been within a foot of the model. So far this year, all but one stadium is showing a negative unexpected distance and all but one are also showing a negative change over the same period last year. Even more, the change of -5.1 ft. for 2018 over 2017 is greater than the unexplained distance gain of 3.1 ft. from 2015 to 2017. All numbers are based on YTD June 20^th for comparative purposes. The full year averages are also shown below which indicate an expected weather related distance pickup in the second half.

First, let’s take a look at average exit velocity (EV) , launch angle (LA), and distance based on a June comparison of “well-hit” balls (defined as EV>=90 and LA>=15 <=45).

Given that the comparison is based on YTD June data for both periods, it is highly unlikely that weather is the major cause.

Unexplained Distance vs. Model (in ft.). All Years are June 20 YTD.

Note: Coors Field Excluded

As you will see in the model at the end, distance is significantly impacted by the horizontal hit direction. Here is a summary of unexplained distance for the same June analysis periods:

This is really quite remarkable. Based on the breadth of what is happening, it seems the most logical conclusion is that something is up – again! with the ball.

Model Construction

Average distances for well-hit fly balls (≥90 MPH, LA≥15°<45) at each exit velocity and launch angle combination between 90-115 MPH and 15°-45°, respectively were used to create a model of expected distance. The model was then expanded by including EV and LA combinations in tenths.The dataset was 2015-2016 (excluding all balls hit at Coors Field). The distance difference for each hit was then examined based on the horizontal angle of the hit. The pattern of the distance differences indicates there is a significant directional bias likely caused by spin as illustrated below. For the total unexplained distances referenced in the above chart, all hits were adjusted for directional impact. The illustration below is based on right-handed hitters only (A left handed model is used to evaluate left standing hits).

Source of all data is Baseball Savant