Let’s Talk About Pitcher Luck
Luck invariably plays a role in all baseball players’ individual performances. Try as they might, baseball fans, players, and analysts will always compete with those fickle forces, which obscure the pure performance of any given player. That said, there have been significant strides made to isolate players’ true skills. Descriptive metrics like FIP and xwOBA capture the performance of players after stripping out the value of the defense on the field and describe player skill based on those factors they specifically can control.
Luck, however you might conceive of it, generally does not last forever, though. Regression goes hand in hand with luck; should a player benefit from a stretch of good luck, logic (and statistics) would have it that, over time, that player’s performance will regress toward his actual skill level. The same principles apply to those players whose luck has been of the negative variety.
To that end, this post aims to highlight a few basic, well known, metrics for spotting particularly fortunate – or unfortunate – performances among starting pitchers. I highlight those players who have benefitted from, or been hindered by, forces largely outside of their control. First, I look at those most and least fortunate seasons since 2002 (when data for all three metrics was tracked on FanGraphs). Then, data are examined exclusively from 2021 in order to highlight those players potentially primed for performance swings in 2022.
Those metrics of interest are the following:
None of these metrics are wholly and completely outside of a pitcher’s control. There is a slight association between higher BABIP and ground-ball pitchers, for instance; HR/FB is a more important factor for fly ball pitchers, and less problematic for those pitchers with a knack for limiting hard contact. Still, there is clear evidence that metrics such as these are by and large outside a pitchers control. Taking a look at the convergence of these metrics offers an interesting glimpse into those seasons wherein luck played a relatively large role (in either a positive or negative sense) in eventual pitcher value.
A couple notes on the charts that will follow. First, I have included the values for each of the three metrics above for each pitcher-season. Next, I have taken the percentile value of those three metrics across all qualifying starters since 2002, of which there are 1,540. Finally, the “Average Luck” metric, which these leaderboards are sorted by, is the simple average value of those three percentile rankings.
Below is a chart of those 15 player-seasons that were considered to be the most “lucky”, as measured by these three metrics, since 2002. Put another way, these pitchers all benefited from some combination of low BABIP figures, a high LOB%, and low HR/FB.
It should be noted, first and foremost, that some incredible seasons are included in the chart above – sorting by those luck-based metrics clearly creates some sampling bias. Still, even exceptionally talented athletes can use some good fortune to go with their considerable skills. Roger Clemens‘ 2005 season, Zack Greinke‘s 2015, and Dallas Keuchel‘s 2020 are tied at the top. Not coincidentally, those seasons from Clemens and Greinke resulted in ERA titles. In fact, 5 of 15 player-seasons listed above (Carpenter, Schmidt, Kershaw (’13)) resulted in ERA titles.
Below is a chart of those players on the other end of the spectrum, which is to say those whose seasonal luck was quite poor relative to other recent player-seasons.
Not a single player-season featured above has a BABIP allowed below .300, a HR/FB below 12.5%, or a LOB% above 70.6%. In other words, balls put in play were more often falling for hits, balls hit in the air were more often going over fences, and players on base were more often scoring, relative to league average. Anecdotally, I might have expected more seasons by Colorado pitchers aside from Jon Gray‘s 2018.
Of all those 1,540 qualifying seasons, no 2021 player-season made the top or bottom 15 in “average luck,” which isn’t particularly surprising based on sheer numbers alone. Now though, those charts above are replicated while only including 2021 seasons. It should be noted that there were just 38 qualifying pitchers in 2021, so this top and bottom 15 charts in fact capture the considerable majority of qualifying pitchers.
Walker Buehler, Max Scherzer, and Julio Urías, three players that made up the Dodgers playoff rotation, led baseball in “Average Luck” in 2021. However, they did so in different ways as Buehler and Scherzer relied on very low BABIPs and LOB% relative to their HR/FB figures. Urías, meanwhile, had above average, but not exceptional, luck in all three categories.
Finally, those unlucky pitchers in 2021.
Dallas Keuchel’s transition from 2020 to 2021 is a clear example of the fickle nature of these three metrics: in 2020, Keuchel tied the highest luck score since 2002 while in 2021, by this simple average measure, he was the least lucky qualifying pitcher. With a smaller sample, these lists include pitchers whose luck balanced inn some ways: Sean Manaea, for instance, was stuck with poor BABIP and HR/FB figures, but actually benefitted from a strong strand rate.
In summation, this was a really simple exercise that cast a lot of nuance to the wayside. Luck, in its many forms, is not created equal, and taking a simple average of these three metrics is not particularly rigorous. Nor are those three metrics included close to the only ways to capture luck. Regardless, seeing them together can be informative and create context for a pitcher’s performance. Whether those 2021 pitchers see the regression that so often trails outlier performances remains to be seen, but there is a not insignificant chance that it does.
Recent Comments