Welcome to “The Regression Report,” a new feature I’ll be writing throughout the MLB season. Finding indicators of positive and negative regression is a big part of my handicapping on a team and player level. The goal is not only to isolate areas in which teams and players will improve or decline in hopes of making winning wagers but to also uncover these things and get ahead of the lines in the betting market.
As I write about players and teams, you’ll be able to identify these signs and apply them in your own handicapping using FanGraphs, Baseball-Reference and similar sites as your statistical guides. The goal is to show readers how to use the stats and data and then analyze why they are areas of regression.
Sample sizes remain fairly small, so a lot of these indicators could swing quickly one way or the other, but there are some obvious ones that are good starting points for the initial column.
Positive: Toronto Blue Jays offense
The Blue Jays are off to an excellent start this season at 15-9, due in large part to a 9-3 record in one-run games. The pitching staff has carried a team that was expected to have one of the best offenses in baseball. There are a lot of offensive stats pointing toward substantial improvement soon.
First, the Blue Jays are second in MLB in Hard Hit% at 45.6%. To be classified as a “hard-hit ball,” it must be hit at least 95 mph in exit velocity. In the Statcast era, which dates to 2015, this type of batted ball has had a batting average of at least .500 every year until this season, when it has dropped to .465. Either way, that’s still much higher than the calculation of batting average minus strikeouts, sac bunts and sac flies of .318. Hard contact is a positive for an offense, and the Blue Jays have made a lot of it but rank just 17th in batting average on batted balls of 95+ mph.
Second, and this is the biggest one, the Blue Jays have been terrible — but also terribly unlucky — with men in scoring position. Their batting average on balls in play is the league’s lowest at .188. Their Weighted Runs Created Plus (wRC+ ) is 56, which means they are 44% below league average offensively with RISP, ranking last in that category.
The hitters are pressing in those high-leverage situations. Toronto ranks 15th in overall strikeout percentage but has the fifth-highest K% with RISP. When this team starts cashing in on its premium run-scoring chances and the offense takes off, it will be a sight to see.
Negative: Minnesota Twins pitching
I will preface this by saying the Twins are a very smart front office that makes a lot of savvy moves and has done a good job of identifying talent from other teams. That seems to be the case with the additions of guys like Dylan Bundy, Chris Paddack, Joe Ryan and Sonny Gray. Gray is on the injured list, but Bundy, Paddack and Ryan have all contributed quality innings to the cause.
However, I’m not sold on this pitching staff. The Twins lead the league in left on base percentage at 80.3%. The highest team last year in LOB% was the Dodgers at 76.5% with an elite pitching staff and a great defense. Natural regression will pull the Twins down, which will lead to more runs allowed. The starters stand out even more in this department with an 81.3% LOB%, which is among the league leaders.
Offense is down around the league, but the Twins rotation has a .243 BABIP against. The Dodgers were the only team under .272 last season. Even with a record-low batting average overall, the Twins won’t continue to be that fortunate on batted balls.
Perhaps Bundy has figured it out and found a home. He also owns a 4.67 ERA and a 4.64 FIP over 792 career innings and has posted an ERA under 4.00 only one time, which came in the COVID-19-shortened 2020 season. Maybe Paddack, a limited two-pitch pitcher, has also found a new home. He has allowed 52 home runs in 64 career games but hasn’t allowed one yet over 14 2/3 innings with Minnesota.
The Twins might be a good team, but their pitching rotation will not sustain this pace, and the bullpen already has a 3.65 ERA while having the fifth-best LOB% at 79.3%.
Positive: Eduardo Rodriguez, Detroit Tigers
Rodriguez’s Tigers tenure is off to a rough start. He has a 5.33 ERA in his 25 1/3 innings of work. However, he has a 3.76 FIP and a 3.88 xFIP. One really big indicator of positive or negative regression is the difference between ERA and FIP. A high ERA with a much lower FIP indicates positive regression, which is the case for E-Rod.
His K/BB rate is solid and there is nothing to be upset about with his home run rate. Even his BABIP is fine. Rodriguez’s two major problems have been level of competition and sequencing. His LOB% is the lowest in the league at 38.2%. League average tends to be around 72%, and the lowest LOB% among pitchers with at least 120 innings last season was 60.4% (Matt Harvey). Not only is Rodriguez a lot better than Harvey, he’s been unsustainably unlucky thus far.
Also, Rodriguez has faced the White Sox, Red Sox, Yankees, Twins and Dodgers. Most of those teams have really good right-handed hitters. When he faces teams like the Royals and Guardians and some of the league’s other bottom feeders, his numbers should naturally improve. He’s had no luck to this point.
Negative: Madison Bumgarner, Arizona Diamondbacks
With the depressed offensive environment, walks are really bad, a point I made last week when discussing bullpens. Bumgarner’s 11.7% walk percentage is the sixth-highest among pitchers with at least 20 innings pitched. To make matters worse, his K% is the 11th-lowest out of 85 pitchers.
Using the ERA-FIP shortcut, Bumgarner has a 1.17 ERA with a 4.35 FIP. That is a clear-cut sign of negative regression. Add the awful K/BB numbers and his unsustainably low .197 BABIP, coupled with an unsustainably high 86.2% LOB%, and you have all the makings of a pitcher ready to go on an ugly run.
Positive regression takes time. Negative regression can happen over a period of time or a pitcher can give up eight runs in two innings and have a lot of the numbers normalize, especially this early into the season. We may not have a lot of swings at the Bumgarner piñata, but we should get some candy with each hack. Even if you don’t want to lay the price against Arizona, these types of angles can be helpful with team total bets or pitcher props.