In The Book: Playing the Percentages in Baseball, the authors investigate batter-pitcher matchups using a number of different statistics to see if choosing matchups using those statistics give either side an advantage. The statistics used are handedness, GB/FB rate, BB/K rate, contact rate, and player quality. This post re-examines those statistics using data from 2017-2019 to see if the trends observed hold true in 2019. Data Background In the original analysis, data was used only from batters and pitchers with 800 PAs or more between 1999 and 2002. That allowed for 340 batters and 314 pitchers and their matchups to be analyzed. To try and replicate the analysis as best as possible, I used all batters with 800 PAs or more from 2017 and 2019 and all pitchers with 200 IP from 2017 and 2019. That allowed my analysis to have 295 batters and 194 pitchers. All statistics for classifying players were collected from fangraphs.com and play-by-play data was collected from baseballsavant.com. Only matchups between qualified players were used in the analysis. Re-examining Statistics used in The Book Handedness Observations from 1999-2002 data: • 39 point swing in wOBA • Left-handed batters benefit more than right-handed batters by 11 points • Pitchers benefit equally from having the advantage (20 points) • Teams are using the information and get a left-left matchup in 28% of at-bats Observations from 2017-2019 data: • 47 point swing in wOBA • Left-handed batters benefit more than right-handed batters by 9 points • Left-handed pitchers benefit more than right-handed pitchers by 21 points • Teams get a left-left matchup in 25% of at-bats and have increased left-handed pitcher use overall The main difference in the data is the extra advantage left-handed pitchers are getting. There are more left-handed pitchers now (31%) than from 1999-2002 (24%), but the percentage of left-handed batters that left-handed pitchers face has decreased from 28% to 25%. This indicates that there are more lefties pitching but teams are using them less for spot relief against a left-handed batter. This has led to more at-bats against right-handed batters inflating the advantage lefties get from facing a left-handed batter. As baseball works to decrease the number of pitching changes in a game this decrease in left-left matchups will likely continue but the value of a lefty who can get outs against everyone will too. Key takeaways: • The platoon advantage from handedness still exists and has increased since 2002 • Left-handed pitchers are being used more overall but less to get specific left-left matchups Note: It is not clear how the authors divided players for classification in the original work but for my analysis I used 25th and 75th percentiles as cutoffs. For example, a hitter with a FB/GB rate above 1.46 (75th percentile) was considered a flyball hitter and a hitter with a FB/GB rate below 0.94 (25th percentile) was considered a groundball hitter. All other hitters were labeled 'neutral'. The same process was used for pitchers and on all other statistics used from here on out. Due to this unknown in the original data, any specific comparison cannot be made between the datasets. I will only note if the trends are the same or different and focus my analysis on data from 2017-2019. Groundball/Flyball Tendency Observations from 1999-2002 data: • 32 point advantage in wOBA • Flyball hitters performed better than groundball hitters regardless of the matchup • Pitchers gained an advantage from facing hitters with the same GB/FB tendency • Teams were not using this information although most batters and pitchers are classified as neutral so choosing a matchup is more difficult than with handedness Observations from 2017-2019 data: • 48 point advantage in wOBA • Flyball hitters hit best against groundball pitchers (32 point advantage) while groundball hitters hit best against flyball pitchers (16 point advantage) • Teams don’t appear to be using this information to get a higher percentage of advantageous matchups The main change observed is that flyball hitters are no longer the best hitter regardless of the matchup. This could be because players are now more focused on a swing or pitch type leading to more polarity among matchups. For example, flyball hitters might be what we now consider launch angle hitters. They are highly skilled at getting the ball into the air and overcoming a groundball pitcher's sinking pitches but are also more likely to pop-up or swing and miss against a flyball pitcher's elevated fastball. This also explains why flyball hitters gain a larger advantage than groundball hitters. A mistake against a flyball hitter is more likely to be a home run or extra-base hit, elevating wOBA and making that matchup the most important for choosing a pitcher. However, teams do not appear to be using this information to make decisions. Against a flyball hitter 22% of at-bats were against flyball pitchers and 23% against groundball pitchers. Key takeaways: • The platoon advantage from GB/FB tendency still exists • Flyball hitters create the largest swing in advantage but don’t outperform groundball hitters in all matchups BB/K Rates (Controlling the Strike Zone) Observations from 1999-2002 data: • No platoon effect • Better control of the strike zone is better for both pitchers and hitters regardless of matchup Observations from 2017-2019 data: • No platoon effect Key takeaways: • Hitters with high BB/K and pitchers with low BB/K rates have better control of the strike zone and will perform better regardless of the matchup. Contact Rate Observations from 1999-2002 data: • No platoon effect • Lower contact players tend to perform better – hitters because of walks and pitchers because of strikeouts Observations from 2017-2019 data: • Low contact pitchers perform better than high contact pitchers - 40 points vs high contact batters and 69 points vs low contact batters • Batters perform generally the same regardless of matchup except for one – low contact batters vs low contact pitchers where low contact batters are 24 points worse than a high contact batter and 20 points worse than a neutral contact batter There isn’t really a platoon effect here, but there is change in overall trend compared to the 1999-2002 data. Low contact pitchers are still preferred in all matchups. They are typically the best pitchers and generate a lot of strikeouts. In the 2017-2019 data this group included Jacob deGrom, Max Scherzer, and Gerrit Cole. Low contact hitters are still elite – Aaron Judge, Josh Donaldson, and Bryce Harper fall into this category – but they are no longer the best performers in all situations and some of the game’s most elite hitters like Mike Trout and Alex Bregman are high contact hitters. Each group of hitters performs relatively equal (9 points or less) with the exception of the low contact hitter vs a low contact pitcher where there is a 25 point drop in wOBA. This is valuable insight given the emphasis on strikeouts and home runs across the league. A batter with low contact rates may struggle against elite pitching more than other players, even if they are a top tier batter themselves. Key takeaways: • Elite hitters can emerge in all contact rate categories • Low contact pitchers are the best performers and they especially take advantage of low contact batters Player Quality Observations from 1999-2002 data:
• No platoon effect • “Good pitching beats good hitting as much as good hitting beats good pitching.” Observations from 2017-2019 data: • No platoon effect Key takeaways: • Good pitchers decrease all batter performance equally just as bad pitchers increase all batter performance equally Conclusions The trends observed and analyzed in The Book allow teams to create matchups that give the hitter or pitcher a distinct advantage. However, the game has changed since 2002 and some of the trends discussed have changed when using data from 2017-2019. Handedness is still the main factor used by teams and provides a clear advantage, but groundball/flyball tendency could also provide teams with valuable matchup advantages and is still not being used on a wide scale. Contact rate shows that low contact pitchers are still the best in baseball, but a shift in batting has led to the elite hitters mixed between high, neutral, and low contact tendencies. There is much more data available to the public than there was in 2002 and more statistics should be explored to look for matchup advantages teams can easily use. What statistics do you think would be worth exploring?
0 Comments
|