Sports Info Solutions

Log In

Category: Baseball

  • Quantifying Aaron Nola’s Early Season Command

    Quantifying Aaron Nola’s Early Season Command

    By ANDREW KYNE

    Aaron Nola was one of baseball’s best pitchers in 2018, but currently has a 6.84 ERA and 6.05 FIP through five starts in 2019. His walk rate, which sat around 6 to 7 percent in his first few seasons, is currently north of 10 percent.

    After his April 15th start against the Mets, Nola’s manager Gabe Kapler said the following (via MLB.com):

    “I’m concerned about his command. His command is his calling card. He’s got movement, deception, life, and those things are still there. He just needs to put the ball where he wants to throw it.”

    As part of our pitch charting operation at Baseball Info Solutions, we chart not only the location of the pitch, but also the location of the catcher’s mitt. Comparing the distance between the pitch location and mitt location gives us, on average, a pretty good proxy for a pitcher’s command.

    For the entire league, we take all of those distances from the mitt and divide pitches into four buckets, based on percentiles:

    • Closest to Mitt (25th percentile and lower)
    • Close to Mitt (26th to 50th percentile)
    • Far from Mitt (51st to 75th percentile)
    • Farthest from Mitt (76th percentile and higher)

    Some of the leaders in 2018 in throwing close to the catcher’s target (combining the Closest to Mitt + Close to Mitt buckets) were names you would probably expect: Dallas Keuchel, Kyle Hendricks, Zack Greinke, etc.

    Nola was pretty good, too. Among the 100 pitchers who threw the most innings last year, Nola ranked 18th in “close percentage” (Closest to Mitt + Close to Mitt).

    This year, however? Among the 100 pitchers who have thrown the most innings so far, Nola ranks 75th in close percentage.

    Here’s a look at his distribution of pitches in terms of proximity to the catcher’s target, using those league-wide percentile buckets:

    SeasonClosestCloseFarFarthest
    201831%25%23%21%
    201924%23%31%22%

    Last year, 31 percent of Nola’s pitches were in the Closest to Mitt category, but that percentage is down to 24 percent so far this year. Additionally, he’s seen a big jump in the Far from Mitt pitches (those that fall in the 51st to 75th percentile across the league).

    That said, Nola’s start against Colorado on Saturday may have been a step in the right direction. He struck out nine and walked only one, and the command numbers were much better than his previous outings:

    DateClosestCloseFarFarthest
    3/28 (ATL)19%21%39%20%
    4/3 (WSH)29%18%25%29%
    4/9 (WSH)19%28%28%24%
    4/15 (NYM)19%11%40%30%
    4/20 (COL)35%33%20%12%

    Perhaps he’s shaking off some early season rust or still getting acclimated with his new catcher, but Nola’s command will be something to keep an eye on as the Phillies look to make a push for the NL East.

  • The Cost of Defensive Mistakes

    The Cost of Defensive Mistakes

    By ANDREW KYNE

    As mentioned on our podcast this week, one of my biggest takeaways from the first few weeks of the season was how the Mariners were winning in spite of their defensive performance.

    Seattle started 13-2 before a six-game losing streak this week. The team’s offense is cooling off after its hot start, but the defense has been poor the entire way.

    The Mariners currently have the worst Defensive Runs Saved in Major League Baseball (-27) and have committed the most Defensive Misplays (49).

    Our Defensive Misplays system at Baseball Info Solutions tracks all mistakes in the field, beyond your typical fielding or throwing errors. Defensive Misplays range from dropping flyballs to taking poor routes to missing the cutoff man. A Defensive Misplay must have a cost to the defense — either the loss of a base, an out, or the opportunity to make an out.

    Do Defensive Misplays correlate with winning? Similar to Scott Lindholm’s Baseball Mistake Index, let’s see if our defensive mistakes correlate with team success.

    The above plot shows a team’s Defensive Misplays correlated with its winning percentage for seasons between 2016 and 2018. Not a super strong correlation (R-squared = 0.28), but clearly a trend where, on average, teams with fewer misplays tend to win more games and teams with more misplays tend to win fewer games.

    What if we look at net Defensive Misplays? Let’s look at opponent misplays minus team misplays (so a higher number is favorable).

    The correlation is stronger here (R-squared = 0.40). Over the course of the season, it generally helps to have your opponents making more defensive mistakes than you do.

    But what about in individual games? Here’s how teams fared over the last three seasons when making different amounts of Defensive Misplays in a game:

    Of course there are many variables that decide a baseball game, but when teams didn’t make a Defensive Misplay, they won almost 63 percent of the time. And as teams made more and more misplays, they won less and less.

    And here are the net Defensive Misplays, again presented as opponent misplays minus team misplays. These are the winning percentages on the favorable side of it — when your opponent makes more mistakes than you do:

    When two teams had the same amount of Defensive Misplays in a game, they went exactly .500. But as the balance of misplays tipped in one team’s direction, the team’s winning percentage went up and up.

    Needless to say, it pays to limit your mistakes — and helps to have your opponent making mistakes, too.

  • Breaking Down Plate Discipline by Pitch Type

    Breaking Down Plate Discipline by Pitch Type

    By ANDREW KYNE

    Last week, I presented a way to visualize plate discipline, referencing some of the numbers on FanGraphs like O-Swing% and Z-Swing% that make use of Baseball Info Solutions’ pitch charting data.

    For each player, those plate discipline metrics obviously come at an aggregated level across pitch types. When we see that Joey Votto swung at 16 percent of pitches thrown outside the strike zone last year, that’s his total performance — against fastballs, curveballs, sliders, changeups, etc.

    But what if we split those out? What kind of discipline does Votto — and the rest of the league — have against different types of pitches?

    Let’s consider the chase rates — the percentage of swings taken on pitches outside the zone (O-Swing%) — for the 179 batters who saw at least 1,000 total pitches out of the zone in 2018. For now, let’s just break it down by O-Swing% on fastballs and O-Swing% on breaking balls (curves and sliders).

    Here’s a look at the leaderboard for each:

    And here’s the correlation between those two metrics for these players:

    And some takeaways:

    — There’s a positive relationship, which is expected — batters who chase (or don’t chase) fastballs probably also chase (or don’t chase) a lot of breaking balls.

    — However, the rate of chasing breaking balls is, understandably, higher. Using the trend line on the graph, note that a 20 percent fastball O-Swing% corresponds with something more like a 26 percent breaking ball O-Swing%, on average.

    — Overall, the batters in this sample swung at out-of-zone fastballs 27 percent of the time, compared to 33 percent for out-of-zone breaking balls.

    — Of the 179 batters, 29 had a lower O-Swing% against breaking balls than against fastballs. Three notables are those labeled on the graph: Votto, Juan Soto, and Jose Ramirez. All three showed excellent discipline against all pitch types, but were particularly impressive against breaking balls. And all three are great offensively.

    — On the other side of that is Yasmani Grandal. Overall, he had a very good O-Swing% at 23 percent. But he was outstanding against fastballs (14 percent), and only about average against breaking balls (33 percent). Last season, he hit .279 with a 1.046 OPS in at-bats ending with a fastball, but only .210 with a .555 OPS in at-bats ending with a slider or curveball.

  • The Orioles are fully embracing shifting

    BY MARK SIMON

    While you were distracted by Chris Davis’ futility pursuit, you might not have noticed something interesting.

    The Mike Elias/Sig Mejdal/Brandon Hyde-run Orioles have gone full-scale Houston Astros when it comes to defensive shifting.

    The Orioles have used a defensive shift on just under 60 percent of the balls in play against them this season. That’s the third-highest rate in the majors behind the Marlins (73 percent), whose outlier shifting performance we wrote about, and the Brewers (62 percent), who have been willing proponents of that defensive approach.

    Here are the top five teams in shift usage this season.

    Team20192018
    Marlins73%33%
    Brewers62%42%
    Orioles60%30%
    Twins58%48%
    Pirates57%39%

    It should be pointed out that the previous regime, led by GM Dan Duquette and and manager Buck Showalter, wasn’t necessarily anti-shift. The Orioles were middle of the pack in shift usage last season at about 30 percent. It’s just that this year they’ve taken it to an extreme in the first couple of weeks.

    What’s distinct about the Orioles’ approach is what they’re doing against right-handed hitters. They’re shifting against them on 55 percent of the balls in play. That’s more than double the MLB average of 25 percent and nearly double the Astros’ 28 percent rate. We mention the Astros, not just because they’re Elias’ and Megdal’s former employer, but because they’ve ranked top five in shift usage against right-handed batters in each season from 2013 to 2018.

    Who are they shifting?

    Where have the Orioles changed? Here are a few examples:

    Yankees catcher Gary Sanchez was shifted 12 times on the 14 balls he put into play. Last season, the Orioles shifted him in only three of 15 such instances.

    Sanchez’s teammate Luke Voit was shifted once on 10 balls in play against the Orioles last season. He was full shifted on 13 of the 15 balls in play in the first two series’ against the Yankees this season.

    Blue Jays outfielder Randal Grichuk drew a shift alignment from the Orioles 13 times on 47 balls in play last season. In the first series against the Orioles, Grichuk was shifted on all five balls he put in play in which we could determine if a shift took place.

    All of these players pulled at least 83 percent of their last 120 grounders and short liners, thus making them legitimately shift-worthy. The Orioles also regularly shifted right-handed hitting Brandon Drury (79 percent) who falls just below the shift-recommendation threshold of 80 percent.

    They did draw the line at spray hitter DJ LeMahieu, though he did see a partial shift (two fielders significantly deviating from normal spots) once.

    Sound strategy

    Is it working?

    The Orioles have converted 74 percent of ground balls hit against them when shifting into outs, comparable to last year’s 75 percent, which happened to be the MLB average. But shifts have been a much better option than non-shifts. The Orioles’ ground ball out rate in non-shifts is 63 percent (at least one out on 41 of 65 ground balls). That ranks last in MLB.

    Let’s also note that the Orioles are playing with a highly-inexperienced shortstop, Rule V pick Richie Martin. He’s at -4 DRS in non-shifts so it seems that the Orioles are doing what they can to help him succeed.

    Like with most things, the jury is still out on whether this will be a long-term commitment by the Orioles. If nothing else, it does send a message that Baltimore is willing to think differently and aggressively when it comes to trying to compete in the AL East.

  • Visualizing Plate Discipline

    Visualizing Plate Discipline

    By ANDREW KYNE

    Baseball Info Solutions’ pitch charting data allows for many interesting applications — one being as a way to evaluate a player’s plate discipline.

    By checking out the plate discipline leaderboards at FanGraphs, we can do things like confirm that Joey Votto rarely swings at pitches out of the strike zone or learn that Freddie Freeman swings at a ton of pitches in the strike zone. (O-Swing% shows how frequently hitters chased pitches out of the zone; Z-Swing% shows how frequently hitters swing at pitches in the zone.)

    We can put a number on it. But what does it actually look like? Where are hitters taking their swings, and how far do they extend their zones?

    Using methods similar to what Jim Albert has demonstrated on the Exploring Baseball Data with R blog with generalized additive models and what FanGraphs has on its site, let’s visualize the swing tendencies of baseball’s most and least disciplined hitters.

    The plots below show how likely a batter is to swing if a pitch is thrown in a certain location, using data from the 2018 season. As the legends show, anything above a 25% expected swing rate is colored in red, with darker red indicating a higher swing rate. All plots are from the pitcher’s perspective.

    Out-of-Zone Differences

    First, let’s compare some of the extreme hitters who either swung a little or a lot at pitches out of the zone last year.

    Low O-Swing%: Joey Votto

    The stat: Swung at 16% of pitches outside the zone in 2018 (lowest among qualified batters)

    Votto’s plate discipline has long been elite, and his expected swing rate locations almost perfectly fit the strike zone borders. While he covered all in-zone pitches to some degree in 2018, it appears that there’s a bit higher expected swing rate on down-and-in pitches (like this pitch he got from Edgar Santana).

    Low O-Swing%: Andrew McCutchen

    The stat: Swung at 19% of pitches outside the zone in 2018 (second-lowest among qualified batters)

    McCutchen followed Votto among the O-Swing% leaders last year. His 19.4% O-Swing was a career-best. There’s a small patch of dark red there in the heart of the zone, about belt-high, that appears to have been his most likely swing (like this one against Andrew Chafin).

    Now, a couple players who do extend beyond the zone…

    High O-Swing%: Salvador Perez

    The stat: Swung at 48% of pitches outside the zone in 2018 (highest among qualified batters)

    He’ll miss the 2019 season after undergoing Tommy John surgery, but Perez’s tendency is too extreme not to include here. The plot shows how willing he was to swing at anything within the vicinity of the zone.

    High O-Swing%: Javier Baez

    The stat: Swung at 46% of pitches outside the zone in 2018 (second-highest among qualified batters)

    Baez extends the zone in similar fashion, though maybe not as much on far outside pitches as Perez. Interestingly, the up-and-in pitch looks like one that he offers at a lot (or at least did in 2018). Here are some examples of that: a swinging strike against Brandon Woodruff; a popup against Luke Weaver; a home run against Gerson Bautista.

    Votto and Baez each had a 131 wRC+ in 2018, but took different approaches to get there. Here’s a GIF to show their swing rate differences back-to-back:

    In-Zone Differences

    Next, let’s look at two batters who were at the extremes of swinging at pitches in the zone.

    High Z-Swing%: Freddie Freeman

    The stat: Swung at 85% of pitches inside the zone in 2018 (highest among qualified batters)

    If a pitch is anywhere in the strike zone, Freeman is likely to swing. Prior to his 85% rate last year, he swung at 84% of pitches in the zone in 2017 and 81% in 2016.

    Low Z-Swing%: Brett Gardner

    The stat: Swung at 53% of pitches inside the zone in 2018 (second-lowest among qualified batters, behind the now-retired Joe Mauer)

    Gardner’s plot features a much lighter shade of red all around, indicating how much less likely he is to swing at pitches in the zone than Freeman. His Z-Swing% has consistently hovered in the 50-55% range for his career. There appears to be a slightly darker band of red across the middle of the zone, but Gardner is definitely willing to let pitches pass through everywhere.

    And here’s a GIF to compare Freeman and Gardner:

  • What is Strike Zone Runs Saved?

    You might notice a few extra columns of stats now available on Fangraphs.com stat pages. They’ll have glossary explanations up in the near future, but for now, here is the definition for Strike Zone Runs Saved.

    Strike Zone Runs Saved (rSZ) is a stat created by Baseball Info Solutions (BIS) that ascertains the contribution each player is making in getting more or fewer called strikes than average, converted to a run value. It can be found on the Fangraphs stat pages. The catcher, pitcher, batter and umpire all have an independent influence on whether a pitch is called a strike or a ball. For the purposes of the Fangraphs stats pages, our concern is with the performance of the catcher and whether he is performing above or below average in this regard.

    The idea of framing pitches is that the catcher is trying to ensure that taken pitches in the strike zone are called strikes and attempting to get called strikes on pitches thrown out of the strike zone through a variety of means (pre-pitch body positioning, subtle body movements upon catching the pitch, or keeping the body and glove still when catching the pitch).

    Catchers who are good at this can have a significant positive impact on the performance of their pitching staff. They have good rSZ numbers. Conversely, catchers who are not good in this area can negatively impact their pitchers’ performance and have a poor rSZ value.

    Calculation

    Before computing rSZ, the calculation starts with Strike Zone Plus/Minus.

    To compute Strike Zone Plus/Minus, the pitch is categorized by its location, the count the pitch was thrown in, the proximity of the pitch to the catcher’s target, and the batter’s handedness. That allows the determination of the percent likelihood that each pitch is to be called a strike. The full array of these strike percentages represents our Strike Zone Plus/Minus Basis, i.e., the basis by which credit is assigned if the pitch is called a strike or debit if the pitch is called a ball. This uses a rolling basis of four years of data to determine how to bucket our pitches.

    If a pitch is called a strike, there is positive credit to be awarded (plus) to the players involved. If a pitch is called a ball, there is negative credit to be assigned (minus). The amount of positive or negative credit given depends on how likely that pitch was to be called a strike in the first place, which is known from the Basis that was previously calculated.

    For example, if a pitch is thrown that is one inch off the outside edge of the plate and 20 inches off the ground, to a left-handed batter, in a 3-2 count, and misses the catcher’s horizontal target  by 3.5 inches, there is an estimated 46 percent chance that the pitch will be called a strike. Therefore, if the umpire calls the pitch a strike, then there are plus-0.54 Strike Zone Plus/Minus points (or “extra strikes”) to be allotted to the participants on the pitch. However, if the umpire calls the pitch a ball, then there are minus-0.46 points to be divvied up among the four parties.

    An iterative approach is used to divide up the credit or debit for each pitch. You can learn more about that here.

    Each catcher’s seasonal Strike Zone Plus/Minus is the sum of his Plus/Minus for every pitch he caught that season. From there, the Plus/Minus can be converted to a run value. The run value is calculated by taking our pitch results from the four-year period, calculating the run expectancy associated with each ball/strike count, and then finding the difference in the change in run expectancy between the next pitch being called a ball and the next pitch being called a strike. Those changes in run expectancy are averaged into a single run value.

    Why Strike Zone Runs Saved?

    As is noted in the definition for Defensive Runs Saved, the value of this statistic is in its allowing you to compare all players using the same set of parameters. A run value stat allows you to assess a player’s ability relative to how much it helped his team win.

    rSZ also is notable for its ability to divvy up the value between all involved parties, rather than give all the credit to the pitcher, batter, catcher or umpire. As such, rSZ tends to value catchers less extremely than other publicly available pitch-framing metrics.

    How To Use Strike Zone Runs Saved

    The best way to think of rSZ is relative to the average catcher. A player with 10 rsZ is 10 runs better than the average catcher.

    rSZ works best as a player accumulates more sample size for the season. A catcher could have had a couple of good days or a couple of bad days within a small sample that would distort his value.

    One thing worth remembering is that if you have two catchers with the same rSZ value, but one has caught 100 games and the other has caught 80, the two may be tied in rsZ, but the catcher with 80 games caught is faring better on a per-pitch basis.

    Context

    In 2014, Hank Conger and Mike Zunino shared the major league lead by saving 16 runs with their pitch framing. Dioner Navarro ranked last. He cost his team 17 runs with his framing. The gap between best and worst was 33 runs. The differential dropped to 25 runs in 2015 and 24 in 2016 before climbing back to 31 runs in 2017.

    In 2018, Max Stassi, Yasmani Grandal and Tyler Flowers led the majors with 10 Strike Zone Runs Saved. Tucker Barnhart and Nick Hundley ranked last with -9. The gap between best and worst shrunk to 19 runs.

    With that in mind, rSZ can be broken into the following tiers:

    Strike Zone Runs Saved Rules of Thumb

    Defensive Ability rSZ
    Gold Glove Caliber +10
    Above Average +5
    Average 0
    Below Average -5
    Poor -10

    Things to Remember

    – Pitch locations for each pitch are hand-charted by Baseball Info Solutions video scouts. BIS does its best to ensure the accuracy of this information and that it is the best information that is publicly available.

    – rSZ is one component of catcher defense that goes into Defensive Runs Saved. BIS also charts how catchers fare at thwarting stolen bases (rSB), blocking pitches (which makes up most of the value of rGFP), and handling a pitching staff (rCERA).

     Links for further reading

    Who is responsible for a called strike? Sports Info Solutions

    Which catchers are best at stealing strikes? Sports Info Solutions

    Is pitch-framing cheating? Fangraphs

  • March’s Most Interesting Team Was… the Marlins?

    March’s Most Interesting Team Was… the Marlins?

    BY ALEX VIGDERMAN

    The NL East might have been the buzziest division in baseball over the offseason, with every team making big trades and/or signings. You know all that because you’ve been paying attention. You’re reading this blog after all.

    The Marlins’ primary buzz surrounded their pursuit of offloading J.T. Realmuto, so once that was done all the NL East attention went back to the four teams jockeying for position at the top.

    What if I told you that the Marlins are on pace to use infield shifts on balls in play more than any team in history? How about over 1,000 more times than the next-closest team?

    Well, first you’d say stop playing the on-pace game. But next you’d probably say, “Really? The Marlins?”

    (As should come with any April baseball article, all small sample caveats apply to the following paragraphs.)

    The Marlins had 78 shifts on balls in play in their series against the Rockies. That’s a pace of 3,159 for the season.

    Remember that the prorated numbers are particularly silly this early. But to use that many shifts is a pretty crazy start to the season. The shiftiest teams on record recorded totals in the 1,800’s.

    In fact, this year’s Marlins team shifted on 73 of its first hundred balls in play, which outpaces any team on record (the 2017 Angels had 65).

    Perhaps just as interesting, Miami has really committed to full shifts (the traditional three-to-one-side version). Since the start of 2018, the league as a whole has favored full shifts to partial shifts (two on each side but with a noticeable shift in positioning), but only slightly.

    This runs counter to what BIS has recommended many times in the past, which is that partial shifts are not nearly as effective. Over the last four-plus seasons (quite a large sample), groundballs and short line drives hit into full shifts have a .222 batting average, compared to .274 with partial shifts.

    Bucking the league trend, 61 of the Marlins’ 73 shifts mentioned above were full shifts. That’s 12 more than any team has ever used in their first hundred balls in play.

    So what gives? It’s not like the Marlins’ opponent this past weekend, the Rockies, is chock full of shift candidates. Current Rockies hitters combined to be shifted on just 23 percent of their balls in play last season, eighth-fewest in MLB.

    It turns out this has been a bit of an M.O. for the Marlins the last few seasons. In each of 2017, 2018, and (as of now) 2019, they have increased their shift rate substantially to start the season. Compared to their shift rates in both the previous season as a whole and just the last month of the previous season, they’ve added 10 percentage points or more each April the last three years. They’re the only team to do that.

    Marlins Shift Percentage on Balls In Play
    (March/April vs. Previous Seasons)

    Per Sports Info Solutions

    As you might notice from this table, though, even with substantial upticks in shifting to start each season, they’ve ended up with much lower rates by the end of the season. In 2018, for example, they shifted on 33 percent of their balls in play through April, but ended up at 29 percent (meaning they were below that from May on).

    Should we expect anything like this going forward? Of course not. Given that annual trend, it’s probably fair to assume that the Marlins aren’t going to blow away shift records and single-handedly bring the league to ban the shift or anything like that.

    That’s especially true if their performance with shifts doesn’t improve. Thus far they have zero Shift Runs Saved, which measures shift effectiveness compared to an average defense. Considering the volume with which they’ve shifted, that’s not great.

    The big trouble with their performance in the shift is that the players are botching plays where the shift might otherwise have been successful. The Marlins have committed nine Defensive Misplays or Errors with the shift on, six more than any other team (acknowledging that they also have more shifts than anyone).

    The Phillies had a similar situation last year when their shifts were being criticized (Andrew Kyne wrote about it last spring). The shift can only help so much when the fielders don’t complete the plays presented to them. Philadelphia saw substantial improvement in their performance over the second half of 2018, so it’s fair to assume that the Marlins will get more out of their shifts going forward as well.

    Obviously this is a wait-and-see situation. But given how outlandishly aggressive Miami has been in shifting so far, it will be very interesting to see how they follow up against the Mets this week.

  • What is Team Shift Runs Saved?

    You might have noticed a few new columns in the stat pages on Fangraphs.com. Fangraphs will have glossary definitions posted in the near future, but for now, you can learn more about these stats here. Here’s our explanation on Team Shift Runs Saved.

    In 2010, Baseball Info Solutions (BIS) began tracking defensive shifts for the first time. That season, Video Scouts observed 2,463 shifts on balls in play from watching video. That number decreased slightly to 2,350 in 2011, but after that shift usage exploded. In 2012, the number of shifts nearly doubled from the previous season. By 2014, that number was up to more than 13,000. In 2016, it jumped to more than 28,000.  In the 2018 season, there were nearly 35,000 observed shifts on balls in play.

    With it widely known that shift usage has skyrocketed, the obvious question that arises is: Which teams are most effective in using it? For that, there is the stat devised by BIS, Shift Runs Saved. It is listed on the Fangraphs team fielding pages as rTS.

    Calculation

    rTS measures whether a shift is doing what it is supposed to do – get outs on ground balls and short line drives that wouldn’t have been achieved with the traditional infield alignment. The calculation is done at the team level rather than the individual level.

    This stat takes how hard the ball was hit and what direction the ball was hit, and then it compares how often a team was able to make an out on that type of ball in play while using a defensive shift to how often the league was able to make an out on that type of ball in play on the whole. Positive credit is assigned for successful plays made and negative credit is given for plays not made, which is eventually converted to a Runs Saved value. The values are summed for all grounders and short liners against defensive shifts to get a seasonal value.

    For example, let’s say a batter hits a groundball 70 miles per hour through the infield at an angle 17 degrees off the first base line. This play may be made 25 percent of the time in an unshifted defense within a two-year period specific to that date. If the play is made, the defensive team gets a collective credit of 0.75 (1 minus 0.25). If the play is not made, the defensive team receives a 0.25 debit. The credit or debit is then converted into a run value, and all of those are summed into a seasonal total.

    Note that velocity is measured as distance divided by time, rather than speed off the bat

    Why Shift Runs Saved?

    Shift Runs Saved became necessary because of The Lawrie Defense. In 2012, the Blue Jays began playing third baseman Brett Lawrie in shallow right field during their defensive shifts. If Lawrie made a play, he was receiving an abnormal amount of credit in the BIS Range and Positioning system because third basemen don’t typically make plays on balls hit to short right field. Lawrie racked up value that other third basemen could not because their teams were not playing them in that area.

    Given that it is the team’s decision to use shifts and to position fielders, it was best to assign the value to the team as a collective unit.

    Shift Runs Saved became the solution to that issue and is the best way to measure performance specific to when a team is utilizing a defensive shift.

    How to Use Shift Runs Saved

    rTS shows how many runs a team saved or cost themselves on groundballs and short liners against a shift in a given season.

    rTS is more of a cumulative metric than an efficiency metric. Often teams that shift the most will have the most rTS, but that doesn’t mean they have the highest frequency of getting outs on a per-shift basis.

    Keep in mind that a team’s reasons for being good or bad in rTS could be due to other factors, such as how it positions its fielders within shifts and the quality of the players on the field at that time.  

    Context

    With the number of shifts continually increasing, the totals at the top of the rTS leaderboard are similarly increasing. Context is best determined by rank within the league within that given season. Almost every team has a positive rTS total because shifts do what they are supposed to do — they allow a team to defend more effectively against those hitters who tend to pull a lot of their grounders and short liners.

    The 2017 season was the first one in which a team recorded at least 30 rTS. In 2018, five teams had at least 30 rTS. The Diamondbacks led the majors with 39, followed by the Athletics (36), Rays (31), Twins (31), and Tigers (30). Four teams had negative rTS, meaning that using the shift cost them runs — the Phillies (-10), Pirates (-6), Nationals (-4), and Mariners (-4).

    Things to Remember

    – BIS categorizes shifts as follows:

    * If 3 infielders are on one side of the infield, it is considered a Full Shift. If at least 2 infielders have deviated significantly from their usual positioning, or if one infielder is playing deep into the outfield (Usually the second baseman playing shallow right field), that is considered a partial shift These are combined into one category on the Splits Pages, “Traditional Shifts.”

    * “Non-Traditional Shifts” are Situational shifts not covered under the definition of traditional shifts, such as playing the infield in.

    – Shifts are determined from video review by trained video scouts based on observations of game broadcasts.

    – Shifts are not recorded for balls that are not put into play (strikeouts, walks, home runs) and measuring shift effectiveness does not take anything other than grounders and short line drives into account.

    Links For Further Reading

    Shift Data! – Fangraphs

    2019 Shift Update – Bill James Baseball Handbook

    Why Baseball Revived a 60-Year-Old Strategy Designed to Stop Ted Williams – FiveThirtyEight

  • Michael Conforto the Met For Whom Beating the Shift Matters Most

    Michael Conforto the Met For Whom Beating the Shift Matters Most

    By BRIAN DEVINE

    The best way for defenses to contain Met outfielder Michael Conforto is to put a full Ted Williams shift on against him. When three infielders were aligned to the right of second base, Conforto’s BABIP on short line drives and groundballs dropped to .189 last season.

    Like many left-handed power hitters, Conforto is susceptible to the shift because he hits the ball to the right side of the infield at a high rate. The 26-year-old outfielder pulled 78 percent of his short liners and ground balls last season, so that explains why the full shift worked so successfully against him. Below is his ground ball/line drive distribution.

    Partial shifts, however, have proven to be ineffective against Conforto. When two infielders were aligned significantly out of position, Conforto had a .375 BABIP last season (12-for-40). While a BABIP this high against partial shifts certainly isn’t sustainable, it demonstrates that the partial shifts don’t work against him, which has been the case his entire career.

    Conforto entered 2019 with a career .301 BABIP on short line drives and groundballs against partial shifts, but he owns a .198 BABIP against full shifts. These numbers indicate that teams should abandon the partial shift against Conforto, and that they should use the full shift against him more frequently.

    A promising sign for Conforto is that he is now committed to adjusting his approach. Working this spring with the Mets’ new hitting coach Chili Davis, Conforto is trying to hit to the opposite field against the shift.

    Davis wants Conforto, and the rest of the team, to take a more situational and contact oriented approach. Conforto might be the Met for whom this is most important.

    Davis’ thinking represents a change in philosophy from previous Mets’ hitting coaches, like Kevin Long, who emphasized the home run. This new strategy could be exactly what Conforto needs to beat the shift, and help him reach his full potential.

    While Conforto posted solid overall numbers in 2018, the full shift stifled his production to a degree. According to SIS’s data, the full shift robbed Conforto of six hits (though he did gain two from partials).

    While that six may not seem significant on the surface, it drags his average down to .243 instead of .254. Small differences like this can make an impact over a course of a 162-game season where games are often won and lost by small margins.

    And thanks to Conforto’s power and excellent eye at the plate, he still was very productive despite his low average. Conforto posted a 120 wRC+ with 28 homeruns in 638 plate appearances last season, and he also managed to get on base at a solid .350 clip because of his high walk percentage. Conforto took ball four in 13 percent of his plate appearances, which ranked 21st in the majors among qualified hitters.

    The former first round pick has the power and patience to rank among the game’s elite. And now that he is fully recovered from the shoulder injury that cut his 2017 campaign short, the full shift is one of the few things that’s holding him back.

    Conforto’s upside will be limited if he can’t adjust to the shift. And once teams realize that partial shifts aren’t effective against him, He will face more of a challenge as he will see more full shifts (he only saw full shift the opening weekend). Therefore, it is important that Conforto heeds Davis’ advice and starts driving the ball to the opposite field more often.

  • What we can learn from Rougned Odor’s bunts against the shift

    What we can learn from Rougned Odor’s bunts against the shift

    By ANDREW KYNE

    With a new season upon us, teams are hoping that 2019 is the year they can beat the infield shift. Consider this out of Mets camp last week, via MLB.com’s Anthony DiComo:

    From the Dominican Summer League on up to the Majors, the Mets’ new front office is placing increased emphasis on bunting and situational hitting. When teams employ defensive overshifts on their Minor Leaguers, the Mets — for the first time — are encouraging their players to bunt to beat them.

    Bunting against the shift isn’t a new idea, but hitters haven’t really embraced it as a strategy.

    One batter, however, did it far more than most in 2018: Rougned Odor of the Texas Rangers. Odor’s 20 bunts against infield shifts were by far the most in MLB:

    So, I decided to watch those 20 bunts and figure out what I could learn. Does it work? How do teams react? Let’s find out.

    It’s not foolproof

    First, how often was Odor actually successful?

    It turns out that he only went 7-for-18 on these bunts. (Two of the outs had men on base and went down as sacrifices.)

    Of course, .389 isn’t a bad batting average to own, but you have to consider these are only going for singles, and therefore providing nothing in the slugging department.

    The league as a whole hit .584 (122-for-209) on bunts against shifts last season. Not bad, but it’s not perfect and relates to a point Matt Carpenter made to ESPN.com last summer. Though he’s specifically talking about hitting grounders to the left side, it also applies to dropping down a bunt:

    “Let’s just say I sell out tonight, and I try it four times. The likelihood of me hitting four straight ground balls to short and ending up 4-for-4 are very slim. If I succeed once or maybe twice, at best I’m going to go 2-for-4 with two singles, where if I just play the game, I might go 2-for-4 with a homer and a double.”

    The accuracy needs to be pinpoint

    So how do you improve on Odor’s .389 average or even the league’s .584? The execution needs to be strong.

    Consider a play like this one against the White Sox. Odor bunts it to the third base side, yet Reynaldo Lopez gets off the mound and throws him out with relative ease.

    Ideally, the bunt will be further away from the pitcher than that one, and preferably as close to the third base line as possible. That obviously requires great accuracy.

    Here’s a look at a spray chart of all the bunts against shifts from left-handed hitters in 2018. The red points represent hits, and the blue points represent outs. There are a lot of hits down the third base line, especially if it can get past the pitcher.

    But even then, you can find pitchers who are able to get off the mound and make plays. Check out this play by Jose Berrios, who would have thrown out Odor here if Joe Mauer held onto the ball at first base.

    The defense will be ready

    This was my biggest takeaway from watching these plays. I think a lot of times when people talk about beating the shift and they suggest laying a bunt down, they’re imagining a defensive alignment like this:

    Either the shortstop or third baseman moves over to the right side, and the one who stays on the left side plays where the shortstop would traditionally be.

    Yet out of these 20 Rougned Odor bunts, this one against Baltimore was the only one I saw that didn’t have a fielder on the left side of the infield playing up near the edge of the infield dirt, protecting against the bunt. (And he still didn’t get a hit on it!)

    Kyle Seager mentioned this in the same ESPN.com article that’s referenced above: “I’ve tried to bunt a few times, and I’ve had a few successes. But the third baseman is usually still in there for the first two strikes, so the bunt is not as big a factor as it could be.“

    This appears to be true for Odor as well. Our company charts the starting positions of infielders on grounders and short liners.

    Here’s a look at where third basemen, when in a shifted alignment, played against Odor in 2018 when he put grounder/liner into play with fewer than two strikes:

    3B Positioning vs. Odor: < 2 Strikes

    And here’s a look at where third basemen, when in a shifted alignment, played against Odor when he put a ball in play with two strikes:

    3B Positioning vs. Odor: 2 Strikes

    (The straight lines represent the base lines, rather than the true edges of the infield grass.)

    You can see that with fewer than two strikes, the 3B typically stayed at home to protect against the bunt. But with two strikes — and the threat of a bunt all but eliminated — the 3B would more often move back.

    The case of Rougned Odor shows that there’s difficulty in bunting against the shift. Perhaps the Mets will gain an edge by emphasizing it at the minor league level and developing accurate bunters, but opposing defenses will continue protecting against it to some degree.