## Using Statcast Data to Predict Future Results

Introduction

Using Statcast data, we are able to quantify and analyze baseball in ways that were recently immeasurable and uncertain. In particular, with data points such as Exit Velocity (EV) and Launch Angle (LA) we can determine an offensive player’s true level of production and use this information to predict future performance. By “true level of production,” I am referring to understanding the outcomes a batter should have experienced, based on how he hit the ball throughout the season, rather than the actual outcomes he experienced. As we are now better equipped to understand the roles EV and LA play in the outcome of batted balls, we can use tools like Statcast to better comprehend performance and now have the ability to better predict future results.

Batted Ball Outcomes

Having read several related posts and projection models, particularly Andrew Perpetua’s xStats and Baseball Info Solutions Defense-Independent Batting Statistic (DIBS), I sought to visualize the effect that EV and LA had on batted balls. For those unfamiliar with the Statcast measurements, EV is represented in MPH off the bat, while LA represents the trajectory of the batted ball in Vertical Degrees (°) with 0° being parallel to the ground.

The following graph visualizes how EV and LA together can visually explain batted ball outcomes and allows us to identify pockets and trends among different ball in play (BIP) types.

The following two density graphs were created to show the density of batted ball outcomes by EV and LA, without the influence of one another.

As expected, our peaks in density are located where we notice pockets in Graph 1. Whereas home runs tend to peak at 105 MPH and roughly 25°, we see that outs and singles are more evenly distributed throughout and doubles and triples fall somewhere in between, with peaks around 100 MPH and 19°. These graphs served as a substantiation to the understanding that hitting the ball hard and in the air correlates to a higher likelihood of extra-base hits. I found it particularly interesting to see triples resembled doubles more than any other batted-ball outcome in regards to EV and LA densities. Triples are often the byproduct of a variable such as larger outfields, defensive misplays, and batter sprint speed, which are three factors not taken into account during this project.

Expected Results

My original objective in this project was to create a table of expected production for the 2017 season using data from 2017 BIP. Through trial and error, I shifted my focus towards the idea that I could use this methodology to better understand the influence expected stats using EV/LA can have in predicting future results. With the implementation of Statcast in all 30 Major League ballparks beginning in 2015, I gathered data on all BIP from 2015 and 2016 from Baseball Savant’s Statcast search database. In addition, I created customized batting tables on FanGraphs for individual seasons in 2015, 2016, and 2017 for all players with a plate appearance (PA).

After cleaning the abundance of Statcast data that I had downloaded, I assigned values of 0 and 1 to all BIP, representing No Hit or Hit respectively, and values of 1, 2, 3, and 4 for Single/ Double/Triple/Home Run respectively. Comparing hits and total bases to their FanGraphs statistics for all individuals, I made sure all BIP were accounted for and their real-life counting statistics matched. Following this, I created a table of EV and LA buckets of 3 MPH and 3°, along with bat side (L/R), and landing location of the batted ball (Pull, Middle, Opposite), using Bill Petti’s horizontal spray angle equation. While projection tools often take into account age, park factors, and other variables, my intention was to find the impact of my four data points and to tell how much information this newly quantifiable batted-ball data can give us.

By calculating Batting Average (BA) and Slugging Percentage (SLG) for every bucket, we can more accurately represent a player’s true production by substituting in these averages for the actual outcomes of similar batted balls. For instance, a ball hit the opposite way by a RHB in 2015 and 2016 between 102 and 105 MPH and 21° and 24° was worth .878 BA and a 2.624 SLG, representing the values I will substitute for any batted ball hit in this bucket.

While a player’s skills may be unchanged, opportunity in one season can be tremendously different from the following, affecting individual counting statistics. With a wide range of factors that can lead to changes in playing time, from injuries to trades to position battles, rate statistics are steadier when looking at year-to-year correlation than counting statistics. Typically rate statistics, such as BA and SLG, will correlate better because they remove themselves from the variability and uncertainty of playing time, which counting statistics are predicated heavily on. Totaling the BA and SLG for each individual batter’s BIP from the 2015 and 2016 season, I was able to then divide by their respective at-bats for that year to determine their expected BA (xBA) and SLG (xSLG).

Year-to-Year Correlation Rates For BA/SLG/xBA/xSLG to Next Season BA/SLG, 2015 to 2016 / 2016 to 2017

 Season (Min. 200 AB Per Season) Statistic 2015 to 2016 2016 to 2017 BA 0.140 0.173 xBA 0.163 0.179 SLG 0.244 0.167 xSLG 0.301 0.204

While our correlation rates for xBA and xSLG are not terribly strong from season to season over their BA and SLG counterparts, we are seeing some positive steps towards predicting future performance. The thing that stands out here is the decline in SLG and xSLG from 2015/2016 to 2016/2017 and my suspicions are that batters are beginning to use Statcast data. It is widely known that a “fly-ball revolution” has been taking place and many players are embracing this by changing their swings and trying to elevate and drive the ball more than ever. With a new record in MLB home runs in 2017, I would not be surprised to see our correlation rates jump back up next season as the trend has now been identified and our batted-ball data should reflect that.

By turning singles, doubles, triples, and home runs into rate statistics per BIP, we are able to put aside the playing time variables and apply these rates to actual opportunities. Similar to calculating xBA and xSLG, I created a matrix of expected BIP rates (xBIP%) for each possible BIP outcome (x1B%, x2B%, x3B%, xHR%, xOut%). In other words, for each bucket of EV/LA/Stand/Location, I calculated the percentage of all batted-ball outcomes that occurred in that bucket (i.e. 99-102 MPH/18-21°/RHB/Middle: x1B% = 0.012, x2B% = 0.373, x3B% = 0.069, xHR% = .007, xOut% = .536), and summed the outcomes for each batter, giving their expected batting line for that season.

Using this information, I wanted to find the actual and expected rates per BIP for each possible outcome (actual = 1B/BIP, expected = x1B/BIP, etc.) and apply these to the next seasons BIP totals. For example, by taking the 2B/BIP and x2B/BIP for 2015 and multiplying by 2016BIP, I can find the correlation rates for actual and expected results, with disregard to opportunity and playing time in either season. Below are the correlations from 2015 to 2016 and 2016 to 2017, with both their actual and expected rates applied to the BIP from the following season.

Correlation Rates For Actual and Expected Batted Ball Outcomes, 2015 to 2016 /

2016 to 2017

 Season (200 BIP Per Season) Statistic 2015 to 2016 2016 to 2017 1B 0.851 0.843 x1B 0.871 0.865 2B 0.559 0.594 x2B 0.624 0.644 3B 0.173 0.262 x3B 0.107 0.098 HR 0.628 0.608 xHR 0.662 0.617

Looking at the above table, the expected statistics have a higher correlation to the following seasons production than a player’s actual stats. The lone area where actual stats prevail in our year-to-year correlations is projecting triples, which should come as no surprise. Two noticeable areas that this study neglects to take into account are park factors and batter sprint speed. Triples, more than any other batted-ball outcome, rely on these two factors, as expansive power alleys and elite speed can influence doubles becoming triples very easily.

One interesting area where this projection tool flourishes is x2B/BIP to home runs in the following season. By taking the x2B/BIP and multiplying by the following seasons’ BIP and then running a correlation to the home runs in that second season, we see a tremendous jump from the actual rate in season one to the expected rate in season one.

Correlation Rates of 2B/x2B To HR In Following Season, 2015 to 2016 / 2016 to 2017

 Season (200 BIP Per Season) Statistic 2015 to 2016 2016 to 2017 2B -> HR 0.381 0.322 x2B -> HR 0.535 0.420

Conclusion

With this information, we can continue to understand the underlying skills and more accurately determine expected future offensive production. By continuing to add variables to tools like this, including age, speed, park factors, as many projection models have done, we can incrementally gain a better understanding to the question at hand. This research attempted to show the effect EV/LA/Stand/Location have on batted balls and how that data can help us find tendencies, underlying skills, and namely, competitive advantages.

Having strong correlation rates on xBIP% to the next season’s actual results, it is exciting to find another area of baseball that gives the information and ability to better understand players and their abilities. With the use of Statcast, we are looking to create a better comprehension of what has happened and how can we use that to know what will happen, and it appears that we have.

## The 2016 Strike Zone and the Umpires Who Control It

Introduction

One of the most-discussed issues in Major League Baseball is the consistency of the strike zone. The rule-book strike zone states “The STRIKE ZONE is that area over home plate the upper limit of which is a horizontal line at the midpoint between the top of the shoulders and the top of the uniform pants, and the lower level is a line at the hollow beneath the kneecap. The Strike Zone shall be determined from the batter’s stance as the batter is prepared to swing at a pitched ball.” After watching games throughout the regular season and playoffs, it is easy to realize this is not the strike zone that is called. Each umpire has tendencies and dictates his own strike zone and how he will call a game. With the rise of PITCHf/x and Trackman in the last few years, umpires have been increasingly monitored and judged for their accuracy and impartiality. For this reason, umpires are criticized for incorrect calls more than ever before and I believe are now trending towards enforcing the rule-book strike zone more than in years past.

The purpose of this research will be to do two things. First, I will focus on identifying overarching themes where I look at finding how umpires are adjusting to modern technology but also how the rule-book strike zone is not the strike zone we know. After this, I will dive into a few umpire-specific tendencies. The latter would be helpful to teams in preparing their advance reports by knowing how certain umpires call “their” strike zone dictated by situations in a game.

Analysis

Using PITCHf/x downloaded through Baseball Savant, I have looked at major-league umpires since 2012 in regards to their accuracy in correctly labeling pitches, primarily strikes, and their tendencies dictated by specific situations. While the height of the strike zone is often influenced by the height of the batter, there are other factors to take into account such as the how the batter readies himself to swing at a pitch. Unfortunately, the information publicly available to conduct this research does not include the batter handedness, pitcher name, or measurements of individual strike-zone limits. For this reason, a stagnant strike zone serves our needs best. The height of the strike zone shall be known as 1.5 feet from the ground to 3.6 feet from the ground. This is the given strike zone of a batter while using the pitchRx package through RStudio when individual batter height is not included.

All PITCHf/x data is from the Catcher/Umpire perspective, having negative horizontal location to the left and positive to the right. The width of home plate is 17 inches, 8.5 inches to both sides where the middle of the plate represents 0 inches. After calculating the average diameter of a baseball at 2.91 inches, we add this to the width of the plate. Therefore our strike-zone width will be 17 + 5.82, or 22.82 inches. The limits we will then set are going to be -.951 to .951 feet (or 11.41/12 inches). Throughout the paper I will be referring to pitches that fall within the boundaries of our zone as “Actual Strikes” and pitches correctly identified as strikes within this zone as “Correctly Called Strikes.”

Called Strike Accuracy By Year

As Table 1 shows, correctly identifying strikes that fall in the parameters of the rule-book strike zone has risen substantially. While 2015 has a higher percentage of correctly called strikes, 2016 PITCHf/x data from Baseball Savant was incomplete, with 28 days’ worth of games unavailable at the time of this research. A rise of 5.90 percent correctly called strikes from 2012 to 2015 shows the rule-book strike zone is being more strictly enforced.

While this provides some information, we can also look into where strikes are correctly being called using binned zones. Understanding that the evolution of umpires over the last five years is taking place and trending toward correctly identifying strikes more today than in years past, we can analyze where, in the strike zone, strikes have been correctly labeled.

Called Strike Accuracy by Pitch Location

In Table 2, we can see a tendency among umpires. Strikes are called strikes more routinely over the middle of the plate and to the left (from umpire perspective). As I have mentioned before, the publicly available PITCHf/x data I used did not include batter handedness and I am unable to determine who is receiving the benefit or disadvantage of these calls. Presumably from previous research on the subject, lefties are having the away strike called more than their right-handed counterparts, explaining the separation between correctly identifying strikes in zones 11 and 13 versus 12 and 14.

Binned Strike Zone

While one may argue that there should not be strikes in these bordering zones, we consider any pitch that crosses any portion of the plate a strike. Due to our zone including the diameter of the baseball on both sides of the plate, the outer portion of the plate includes pitches where the majority of the ball is located in one of these zones.

Called Strike Accuracy by Individual Umpire

When gauging an umpire’s ability to correctly identify a rule-book strike, an 85.67% success rate sets the mark with Bill Miller, while Tim Tschida ranks at the bottom of this list, only calling 71.57% correctly. We can infer from Tables Three and Four along with Table One, that while umpires are calling strikes within the strike zone more often, they are still missing over 17% of these pitches. It is important to note that this information does not take into account incorrectly identifying pitches outside the rule-book strike zone as strikes, which when considering an umpire’s overall accuracy, should absolutely be taken into account.

Called Strike and Ball Accuracy by Count

One of the most influential factors in whether a taken pitch is called a strike or a ball is the count of the at-bat. We have all seen pitches in a 3-0 count substantially off of the plate called a strike, just as we have seen 0-2 pitches over the plate ruled balls. Table Five shows the correct percentage of strikes and balls by pitch count. While this shows that umpires are overwhelmingly more accurate at identifying strikes as strikes in a 3-0 count (91.06%) as compared to an 0-2 count (56.66%), we must acknowledge this only paints part of the picture. Umpires are conversely most likely to correctly labels balls in 0-2 (98.73%) counts and misidentify balls in 3-0 (90.32%) counts. I included their accuracy of correctly identifying both strikes and balls here as opposed to throughout the entire paper because we can clearly tell through this information that umpires are giving hitters the benefit of the doubt over pitchers. Umpires are far more likely overall to correctly identify a ball than a strike, as evidenced by the fact that there are no counts during which umpires correctly call less than 90% of balls.

The data in Table Five is corroborated by the visualizations in Figure One and Figure Two. These visualizations of the strike zone include pitches off of the plate and we can see that in a 3-0 count, a more substantial portion of the rule-book strike zone is called strikes while also incorrectly identifying balls as strikes. While in a 0-2 count, a smaller shaded area of the rule-book strike zone works with our findings that less strikes are identified correctly but more balls are correctly called.

Called Strike Accuracy by Pitch Type

The next area I looked at was whether pitch type significantly altered the accuracy of umpires. In order to do this, I grouped all variations of fastballs into “Fastball” and all other pitches into “Offspeed”, while omitting pitch outs and intentional balls. I was able to see how umpires fared in correctly identifying strikes by pitch type in Table Six.

Not surprisingly, we see Bill Miller near the top of the list with both Offspeed and Fastball accuracy. For umpires as a whole, the difference in accuracy between the two is not large (79.05% Offspeed accuracy vs. 78.91% Fastball strike accuracy). On the other hand, what may come as a surprise is the fact that eight of the top ten highest accuracies were for Offspeed pitches.

Called Strike Accuracy for Home and Away

One of the most-mentioned tendencies of referees or umpires in any sport is home-team favoritism. Whether a foul or no-foul call in basketball, in or out-of-bounds call in football, or a strike or ball ruling in baseball, many think that the home team receives more of an advantage than their visiting counterparts. Looking at top and bottom half of innings, away and home team respectively, we can identify trends and favoritism in major-league umpire strike zones.

While a difference of .62% accuracy may seem like a lot, especially in a sample size of over 650,000 total pitches, we can look at this on a game-by-game level to see the actual discrepancies. For simplicity’s sake, we can assume 162 games a season, making for roughly 11780 games played in our data set (this subtracts all games from the unavailable 2016 data). This leaves us with 23.03 Correctly Called Strikes out of 29.05 Actual Strikes for away teams per game, meaning that 6.02 strikes were not called. As for home teams, we have 22.04 Correctly Called Strikes a game with 28.02 as the Actual Strikes, averaging 5.98 missed strikes a game. By this measurement we can see that more hitter leniency was given to the away team than the home team.

During this time frame, while a higher percentage of strikes were judged correctly, hitters were given more leniency as the away team than the home team on a game-by-game basis.

Called Strike Likeliness in Specific Game Situation

Included in Table Eight are the three most and least likely umpires to call any non-fastball a strike below the vertical midpoint of our zone. I split the strike zone at 2.55 vertical feet and looked at any pitch (not necessarily within the zone) below that height. Here, we are not judging an umpire’s accuracy of correctly identifying pitches, but rather looking at where a certain umpire may call specific pitches. We can see that Doug Eddings is 5.34% more likely to call a strike on a non-fastball as compared to Carlos Torres.

While this does not paint the entire picture, we are able to see how their tendencies can play an important role in the game. Information like this may be valuable to a team in deciding how to pitch a specific batter, which reliever to bring into a game, or factor into being more patient or aggressive while at the plate.

Conclusion

External pressures and increased standards are undoubtable effects on umpire strike zones. As evidenced throughout this paper, strike zones are called smaller than the rule-book strike zone specifies. And while umpires are trending toward correctly identifying strikes, situations such as count and pitch type can affect their judgment.

While the system in place is not 100%, we must understand that these umpires are judging the fastest and most visually-deceptive pitches in the world and are the best at what they do. Major League Baseball must use modern technology to their advantage and provide the best training for umpires to achieve the goal of calling the rule-book strike zone. Another option, while more drastic and difficult to implement, may include adapting the definition of the rule-book strike zone, something that has not been changed since 1996.