📌 Join the BPCrew Chapter in your city and meet up with more Yankees fans! 👉 CLICK HERE

A smarter way to watch early season baseball

As a sabermetrics oriented columnist, it’s always tricky to write early season articles. With such a small sample size to draw from, it can often feel like that early season slump is the beginning of the end for an older player, or like a hot streak is a guarantee of an MVP caliber season. The always hot Yankees Twitterverse is emblematic of what most diehard fans are feeling right now, where every win or loss feels like the validation of our grandest dreams or darkest nightmares.

And while I can’t sit here and claim that I’m not slightly panicked after a series loss to the AAA Orioles to begin the year, I do believe there’s a smarter way to watch early season baseball. Rather than trying to draw sweeping conclusions from each at-bat or game, it’s important to dig into the underlying data and fundamental statistics to get a better picture of a team or player’s early performance.

As much as I’d love Tulo and LeMahieu to maintain an OPS of 1.301 and 1.611 respectively throughout the year, we all know it’s about as likely as Ottavino finishing with a 0.00 ERA or 16.20 K/9 rate. Throughout this first month, don’t be surprised if random fluctuations lead to some spectacular or atrocious stat lines for Yankee players. Rather than focusing on the stats you traditionally track to measure player performance, here’s a process you can use to better evaluate the Bombers early season performance.

All statistics referenced below are available on Fangraphs or BaseballSavant unless otherwise noted. I hope for even those of you who don’t love the new age statistics, this can provide a primer into advanced analytics and player evaluation.

Figure out your expectations

Whenever you’re trying to draw conclusions from inherently noisy and small data sets, it is crucial to first have a baseline in mind. Thankfully with MLB players, we have entire careers of data to draw from to set our expectations for each member of the team to determine if early results are a fluke, or indicative of some change in performance level.

In other words, what you should be looking for in each player varies based on their style and where they are in their career arc. For Brett Gardner, power numbers and exit velocity are probably not the most important statistics since his value is pretty clearly dictated by his ability to get on base and his defense. A big change in his O-Swing% (chase rate) or UZR/150 (ultimate zone rating) would be much more alarming than an expected decline in power.

By contrast, a low xwOBA (expected weighted on-base percentage) or declining Hard% (hard hit rate) would be major red flags for players like Aaron Judge and Giancarlo Stanton who provide their value primarily through their ability to slug the ball. The key insight is that some traditional statistics are not one-size-fits-all, and are especially bad at capturing true performance when they have few data points to work with. Determining your baseline expectations for each player and what constitutes success for them is step one in evaluating early season performance.

Go to the raw data

At its core, the sabermetrics revolution is about trying to find independent data points that are almost entirely in a player’s control. A pitcher’s ERA, the gold standard of traditional pitching stats, is now known to be imperfect because exogenous factors like the quality of the defense or the size of the ballpark are highly determinative. They don’t get to the true talent level or performance of a pitcher. Similarly for hitting, batting average and RBIs are heavily reliant on external factors like luck or the quality of a hitter’s teammates.

And while it’s true that ultimately, at the end of a season, these stats are not meaningless as measures of success, it is especially important to dive deeper into the data when these numbers are highly variable in the early stages of the year. So what are some good places to turn to in lieu of these traditional statistics?

O-Swing% and Z-Contact%: Metrics for how often a player chases a pitch out of the zone, and how often they make contact on swings at pitches in the zone. These are great baseline statistics as a replacement for BB% to measure plate discipline and how well a hitter is seeing the ball.

Hard% and xwOBA: Metrics for the quality of contact a player is making when they do put the ball in play. These are great statistics to substitute for the old “eye test”. If your eyes tell you a player has been tearing the cover off the ball but their OPS is only .600, this data can reassure you that they are indeed hitting the ball well and a positive regression is coming soon. Analogously, if a guy is batting .350 but it doesn’t feel like they’re smacking the ball, this data might convince you that it’s simply BABIP (batting average on balls in play) luck and perhaps a batting title isn’t on the way.

At its core, hitting is all about putting these two things together. If you are disciplined and rarely chase balls out of the zone, and likewise consistently punish the ball when you do in fact swing, chances are you’re a damn good hitter. I can guarantee that looking at these numbers is going to tell you a lot more about how a player is performing early in the season than their RBI total or OPS toward the middle of April.

O-Swing%, Z-Contact%, and Whiff%: These should look familiar this time around. As you can imagine, these same stats are very instructive for a pitcher as well as a hitter. A pitcher inducing a number of chases and who can limit contact on pitches in the zone is likely successfully deceiving hitters and keeping them guessing. The added twist is whiff rate (available through Brooks Baseball) which tracks the number swinging strikes over the total number of swings by a hitter. This is perhaps the single most instructive data point for a pitcher, since fooling hitters into swinging and limiting their contact is, ultimately, the goal of pitching.

Average EV, Hard%, GB%, and FB%: By this point, I hope that you can see where I’m taking this. If so, you’re now a seasoned sabermetrics veteran! Looking at pitcher contact is inherently more of an art than a science. You can be highly successful allowing hard contact if you always keep the ball on the ground. You can be a good fly ball pitcher if batters have very low exit velocities and are making soft contact. Ultimately though, some combination of low exit velocity, low Hard%, and more ground balls than fly balls is a winning formula. Again, cater your reading of the data to your expectations of a given player to try to glean insights into their performance.

Glance at the “catch-all” advanced statistics

For all that I’ve preached about raw data and statistics, the reality is that the joy of baseball comes down to production. But if you’re constantly staring at Judge’s home run total when he comes to the plate, you’ll drive yourself crazy and fail to see the true value he adds with his bat. Here are, in my opinion, the two best statistics to look at if you want to simply measure a player’s production at any given point in the season.

wRC+ : This is the holy grail of offensive statistics. It captures the value of a player on both a weighted and absolute basis, measuring each contribution based on the timing of the hit, the type of hit, and the number of runs created by that at bat. It then adjusts them all based on ballparks and league averages, to create a scale where 100 is league average, 150 and above is excellent, and 75 and below is poor. Again, for early season performance the underlying statistics are more important. Will die on this hill. But if you want a measure of a batter’s performance, this is the metric to use.

FIP- : FIP- is for pitchers what wRC+ is to hitters. This is a catch all statistic that captures a players performance in a more meaningful way than the traditional statistic (ERA), and then adjusts it to capture ballpark effects and league averages. Here, 100 is still league average, but lower is better. Thus 80 and below constitutes high quality performance, and 115 and above indicates bad performance. Don’t drive yourself crazy staring at bloated ERAs like JA Happ’s after today’s brutal start. Start by looking at the types of data listed above, but then come to FIP- to see how a pitcher’s performance stacks up against the league.


I have plenty of buddies who like to poke fun at my obsession with sabermetrics. And trust me, I know that citing a player’s wOBA sounds bizarre compared to an easy and sexy stat like dingers or RBIs. But the reality is that this is the way that front offices are evaluating players now, and for good reason. The data revolution in baseball has given analysts and casual fans like us equal access to high quality data about every MLB player.

If you want to save yourself a lot of headaches and give yourself a framework for analyzing player performance, the tools above apply throughout the baseball season. It is particularly helpful to use these stats now when the traditional statistics tell us next to nothing. But you can use these same data points in your next Twitter argument over whether Aaron Judge or Mike Trout should be the MVP, or whether Xander Bogaerts is worth that monster extension.

Questions? Email me at [email protected]