A lot of baseball statistics take a while to reach what is called a “stabilization point.” That means a stat needs to have enough of a sample size to be considered reliable. Baseball stats and metrics are subject to a lot of randomness and variance, but we can start drawing much stronger conclusions as the number of data points increases. Some stats, like K%, BB%, ground ball percentage (GB%), fly ball percentage (FB%) and exit velocity for hitters, do stabilize very quickly. All of those stats have reached a point of significance this season and can be viewed as reliable.
Other stats simply cannot reach a stabilization point based on the length of the season. Stats such as batting average on balls in play (BABIP) require about 2,000 batted ball events to reach a point of significance. Batting average requires 650 batted ball events. Cubs pitcher Kyle Hendricks had the most batted ball events last season with 597.
As a result, stats such as BA and BABIP are always open to interpretation and are going to fluctuate. This is especially true of stats in certain splits, like with men on base or runners in scoring position. Isolating outliers in all sorts of stats can help us find pitchers, hitters and teams in line for positive or negative regression. That is the purpose of “The Regression Report.”
For the second installment of this feature, here are the teams and players on my mind.
I won’t do this every week, but to illustrate how the concept works, let’s look at the Toronto Blue Jays offense that I wrote about last week. Toronto had a .188 BABIP and a 56 wRC+ with RISP through last Monday’s games. Now, Toronto has a .211 BABIP and a 65 wRC+ with RISP through this Monday’s games. I also wrote about how Toronto had the fifth-highest strikeout percentage with RISP, despite the 15th-highest K% overall. They now have the seventh highest.
While none of these stats has improved in a big way, you can see that they are gradually getting better, and that’s the hope when trying to isolate areas of positive regression.
Similarly, I looked at the Minnesota Twins pitching staff. Twins pitchers led the league with an 80.3% LOB% and the starters were even better at 81.3%. Minnesota’s starters have regressed to a 77.4% LOB% and the team LOB% has gone down slightly to 80%. So the expected regression hasn’t fully hit yet, but monitoring the progress of something that you’ve keyed in is essential to having success with this exercise.
Positive: Chicago White Sox offense
This season has not gone according to plan for the White Sox. They’ve dealt with a myriad of injuries and some questionable managerial decisions, but the numbers point to a team that should be having far more success. Through Monday, the White Sox had a .284 wOBA and a 90 wRC+ , which ranked 24th and 23rd, respectively, in the majors.
The White Sox were also third in Hard Hit% at 43.3% and sixth in Barrel% at 9.8%. The quality of their contact should be leading to far better offensive numbers than a .222/.281/.350 slash line. One of my early-season observations has been that the teams in the Midwest and on the East Coast have been hurt more by the humidor than others. Based on their contact quality, the White Sox have an xSLG (expected slugging percentage) of 1.122 on batted balls of 95+ mph. Their actual SLG on those batted balls is .762. The gap of 360 points is easily the biggest in the league.
The White Sox are about 45 points below the league average in batting average on high-velocity contact and also have one of the biggest discrepancies between wOBA and xwOBA. Interestingly, teams around them are the Orioles, Tigers, Red Sox and Royals — all teams in cold-weather cities. How quickly positive regression happens is a mystery, but Chicago’s results on high-quality batted balls are a big reason why this team is off to a slow start.
Negative: Los Angeles Angels pitching
The Angels have gotten off to an outstanding start with Mike Trout doing Troutian things, plus the big breakouts from Brandon Marsh and Taylor Ward. The offense does appear to have some staying power, but this pitching staff is one that I have major concerns about going forward.
With men in scoring position, the Angels pitching staff has allowed just a .176/.256/.298 slash line with a .247 wOBA. The BABIP against Angels pitchers in that split is .191, the best mark in the league. This is not a strikeout-heavy pitching staff, which means a lot of balls in play, so as these numbers go through a period of regression, this is going to be a team that gives up a lot more runs. The Angels are just 18th in K% with RISP.
The Angels bullpen worries me the most. Newcomers Ryan Tepera, Aaron Loup and Archie Bradley have largely been great, but this is a group with a 3.45 ERA and a 4.10 FIP, which represents one of the largest gaps between ERA and FIP. The relievers have a .234 BABIP against, which accounts for the biggest difference. The Angels are far better defensively than previous seasons, but I don’t think they’re this good and they’re going to have to rely more on that offense for run support sooner than later.
Positive: Logan Webb (San Francisco Giants)
A 3.82 ERA is nothing to scoff at, even in this depressed run environment. Webb has been solid enough over his 35 1/3 innings, but there is room for improvement, and the numbers show the likelihood of it coming soon. One obvious starting place is that Webb has an xERA of 3.24 and a FIP of 2.60. Anytime you see an xERA or a FIP lower than an ERA, that’s a guy that has the chance to improve.
Webb’s allowing a lot more balls in play this season, as his K% has dropped from 26.5% to 18%. I don’t see any reason for that to be the case. Hitters are making more contact with pitches outside the zone but less contact with pitches inside the zone. Better numbers with Z-Contact% is a good indicator of the quality of a pitcher’s stuff because he’s been able to get whiffs on pitches that would have been strikes anyway. Webb’s swinging strike rate (SwStr%) is down from 12.4% to 11.4%, but that shouldn’t lead to such a big drop in K%. The reason his K% is so important is that he has allowed a .351 BABIP. Even though his average exit velocity against is a career best and his Hard Hit% is down from last season, he’s still allowing a much higher rate of hits.
I would expect that to change soon and also for Webb to increase his strikeout rate in short order. He’s a great pitcher to buy stock in right now.
Negative: Chad Kuhl (Colorado Rockies)
There were a lot of candidates for this spot, but Kuhl stands out. The Rockies righty has a 1.82 ERA with a 2.89 xERA and a 3.38 FIP on the season. Those would still be great numbers if he regressed toward them, but the staying power of those numbers is a major question mark. There are 124 pitchers with at least 20 innings pitched this season. Kuhl ranks in the bottom 45 in K%, yet he ranks 22nd in LOB% at 85.6%.
It is much easier to strand runners as a high-strikeout pitcher. I typically look for faster, more significant regression in the LOB% department from guys who don’t miss bats. Kuhl fits that description. He also has a .179 BABIP against with a Hard Hit% of 42.5%, which is well above the league average. His average exit velocity against is 90.1 mph. On batted balls of 90 mph or more, the league is batting .403. Kuhl’s batting average against is .179 and his batting average against on batted balls of 90+ mph is just .260.
Running a low BABIP is extremely difficult at Coors Field because of the thin air and the spacious outfield. Early season stats can be a byproduct of luck and positive variance, but also the level of competition. Kuhl has made starts against the Rangers, Phillies, Tigers, Reds and Diamondbacks, so only one quality offense. He also owns a career HR/FB% of 12.8% and is at 6.7% for this season. I’ll be looking to bet against Kuhl in his next few outings as these numbers normalize.