How Do We Know If a Player's Hot Start Will Continue?
Some numbers can help us spot a player who may be fooling us by a "great" start to the season.
At what point in the baseball season does it turn from “too early to tell” to “legit”? Is it May 1st? First day of Summer? All-Star Break? September? Everyone has their own opinion, but what is true is that small snippets of data can often lead us astray. This is especially true in baseball, where comments like “he’s 0 for his last 12” or “he’s got an ERA under 1.00 in his past 3 starts”, could suggest a player who's simply hit a streak of good or bad luck is far from his true ability.
One of the many fantastic things about baseball is the sheer volume of data and the scrutinizing research that goes into every number. Now I’m not talking about numbers like “Juan Soto bats .400 on Sundays in months beginning in a vowel when the temperature is between 78 and 84.1 I’m talking about stats like expected weighted on-base percentage (xwOBA), batting average on balls in play (BABIP), and fielding independent pitching (FIP). These actually offer insight into the true performance of a pitcher by neutralizing batted ball variance and accounting for quality of batted ball contact. For more in-depth descriptions of many of these advanced stats, check out the FanGraphs glossary. For the purposes of this article, I will focus on wOBA, xwOBA, BABIP, and FIP.
Weighted On-Base Average (wOBA), created by Tom Tango, is a statistic which attempts to give a better credit for the value of each outcome of hitting (single, home run, walk, etc.) instead of treating all hits or times getting on base equally. It essentially combines all the aspects of hitting into one metric and weights each in proportion to their actual respective run value.2 Ultimately, it measures and summarizes offensive value more accurately and comprehensively than other stats such as batting average, on-base percentage, and OPS. Expected Weighted On-Base Average (xwOBA) is similar to wOBA but is formulated using exit velocity, launch angle and, in certain scenarios of batted balls, sprint speed. MLB.com states “xwOBA is more indicative of a player's skill than regular wOBA, as xwOBA removes defense from the equation. Hitters, and likewise pitchers, are able to influence exit velocity and launch angle but have no control over what happens to a batted ball once it is put into play.”3 Batting Average on Balls in Play (BABIP) is fairly self-explanatory. It is a batter’s batting average on all balls hit in play, meaning it excludes events like strikeouts, home runs, walks, and hit-by-pitches, in which the ball did not land in the field of play. Fielding Independent Pitching (FIP) is somewhat of the reverse in that it assesses a pitcher’s run prevention independent of the defense behind him. FIP only accounts for strikeouts, walks, hit batsmen, and home runs. These are metrics that are nearly entirely within the realm of the pitcher’s control. FIP has been found to be more accurate assessment of a pitcher’s performance than ERA.4 These statistics can all be useful in uncovering what players who seem to be performing exceptionally are really frauds.
How to Spot a Fraud
Research has shown that pitchers whose BABIP allowed deviated from the league average tended to regress towards the league average as time progressed.5 Therefore, a pitcher whose BABIP is particularly low will more than likely see that number climb back up in the direction of the league average in the future. This is due to the phenomenon of batted ball variance. Once that ball is hit in play, the end result is largely out of the control of the pitcher. Sure, he can craft his pitches to try and force more weak ground balls or routine pop-ups, but he has no control over the ball that squeaks through the hole or a lackluster second baseman. A pitcher may simply have a streak of bad luck, but that will eventually be cancelled out by the law of averages. So, a pitcher with an exceptionally low BABIP versus the league average and verses his career average is primed for regression.
Spotting frauds is a little easier when looking at expected statistics. A player whose expected number is far worse than his actual number is primed for regression. So, a batter with an xwOBA much less than his wOBA may appear to be performing better than he actually is. Similarly, a pitcher with an xFIP higher than his FIP is likely to see results decline.
In summary, to spot this year’s early potential “frauds”, I looked at…
Batter xwOBA vs. wOBA
Batter BABIP vs. Career Average BABIP6
Pitcher xFIP vs. FIP
Pitcher xERA vs. ERA
Pitcher BABIP vs. League Average BABIP
The Batters
Three players who fell highly on both charts are Orlando Arcia of the Braves, Ryan McMahon of the Rockies, and Jose Altuve of the Astros. All are in the upper portion of the league in both stats, but their deviance from expected stats or career averages signal they may not keep up these hot starts.
The Pitchers
Jose Berrios, Dane Dunning, Lance Lynn, and Paul Blackburn have all been off to very strong starts to the season, but these pitchers all land in the top 6 when it comes to deviance from expected stats and league average BABIP. I excluded Jose Quintana from the aforementioned list because frankly he has been fairly pedestrian to start the season (3.05 ERA and 3.63 FIP). Our goal in this is to identify players who have gotten off to a hot start but will likely regress. Quintana’s start hasn’t been particularly “hot”.
Conclusion
Well first of all, if you have any of the players I discussed on your fantasy team, you may want to trade them now before their value declines (or others read this article). Ultimately, I hope you now have an idea of how to spot a player who may not be quite as good as his numbers suggest. It will be interesting to look back on this later in the season and see how these players have ended up doing.
Totally made-up statistic
Slowinski, P. (2010, February 15). wOBA. FanGraphs. Retrieved April 21, 2024, from https://library.fangraphs.com/offense/woba/.
MLB Advanced Media, LP. (n.d.). Expected Weighted On-base Average (XWOBA). MLB.com. Retrieved April 21, 2024, from https://www.mlb.com/glossary/statcast/expected-woba
Slowinski, P. (2010, February 15). XFIP. FanGraphs. Retrieved April 21, 2024, from https://library.fangraphs.com/pitching/xfip/
Law, K. (2017). Smart baseball: The Story Behind the Old Stats That Are Ruining the Game, the New Ones That Are Running It, and the Right Way to Think About Baseball. William Morrow.
Note - the BABIP “regression to the league average” rule does not apply to hitters the same way it does for pitchers. In his book Smart Baseball Keith Law suggests “There are absolutely hitters who can make better or worse quality contact on a consistent basis and thus regularly post BABIPs that vary from the league average”. Therefore, I compared hitters’ current BABIP to career average BABIP, as a hitter’s career average is more indicative of his true performance than a small sample of the early 2024 season.