Another Way to Look at the Twins’ Lifeless Bats

“We’re not going to win unless we make adjustments at the plate and, ultimately, barrel a lot of balls and hit.”

– Rocco Baldelli, yesterday

The Twins were swept in Atlanta, outscored 13-3 in three games. Minnesota stranded 20 runners on base, hit 0-for-23 with runners in scoring position, and struck out 31 times.

After the game, Rocco Baldelli closed the clubhouse for a players’ only meeting and put his offense on blast in the media, publicly lamenting the team’s record-setting rate of strikeouts and inability to make adjustments to the pitching they are facing for the third time in less than a month.

Baldelli’s frustration finally bubbled over after the Twins struck out 8 times in less than five innings (14 total in the game) against Atlanta left-hander Kolby Allard, a soft-tossing spot starter who has never averaged even a strikeout per inning in his career.

The Twins management’s patient approach and belief that a good process will portend better future results haven’t come to fruition, and Baldelli indicated he is finally tired of waiting and sticking to the same, one-size-fits-all plans.

The Exceptions that Prove the Rule

Dan Hayes noted on Twitter that the Twins’ offense has scored 118 of their 343 total runs (34.4%) in just 12 games (14.6%) of their 82 played.

The large outburst of runs every 10 days or so, and the Statcast metrics that show the Twins more favorably, like hard-hit rate (40.9%, 14th), barrel rate (9.9%, 3rd), and expected production on contact (xwOBAcon, .399, 4th) haven’t proven to be indicators of better things to come.

They’ve shown to be the exceptions that prove the rule about their three true outcomes-oriented booms or busts approach. The results consistently have not been there.

While it generally pays to be process-oriented instead of results-oriented, it’s also important to understand the limitations of the measurements being used. Stats like the ones I listed just above from Statcast use a denominator of batted ball events, not all plate appearances.

That makes them prone to being misleading. Sure, the Twins have tended to hit the ball hard when they hit it, but they also don’t hit it very much, thanks to the league’s worst swing-and-miss rate (29.3%). If we adjust their barrel rate to be on a per-PA basis, it falls to 6th. Where they rank 8th in wOBAcon, they are just 21st in wOBA. You get the idea.

Measuring the Ideal Approach with IPA%

Most of Baldelli’s complaints have been about the Twins’ approach at the plate. What constitutes a good approach? When I wrote about the Twins’ three-true-outcomes orientation earlier this month, we had a good discussion in the comments about how approaches should be adjustable based on situations and how we don’t really have any metrics that give us insight into this.

We discussed that amateur baseball coaches have long made popular the idea of the “Quality At Bat” or QAB. The exact criteria vary from coach to coach, but typically include things like walks, hit-by-pitches, reaching on errors, hard-hit balls, advancing runners into scoring position, making the pitcher throw a lot of pitches (6 or more), driving in a run, and sacrifice flies and bunts. I kept charts of such things when I played in high school and college.

A couple of writers at Pitcher List, Johnathan Metzelaar and Christian Mack, explored the idea of “Ideal Plate Appearances” at the Major League Level over the past couple of years. Metzelaar based his version on Statcast’s most productive contact quality bins — barrels, solid, and flares/burners.

MLB Glossary

Add up those events and divide by plate appearances to get IPA%, a metric that is hosted on Pitcher List’s site at the player level.

Mack built on that to include walks in the numerator, a sort of obvious improvement given that so much of the Moneyball revolution was driven by recognizing the value of a walk and getting on base.

Earlier this week, I expanded on their earlier work and proposed the addition of sac flies, hit-by-pitches, and sac bunts to the numerator, and explored whether hard-hit batted balls (95+ mph exit velocity) were a useful alternative to the more complicated barrels, solid contact, and flares and burners.

In the end, I found that the following formula has a strong positive relationship with offensive production (as measured by wOBA) at a player level, based on qualified player data in the Statcast era (since 2015).

• IPA% = (Barrels + Solid Contact + Flares & Burners + BB + HBP + SF + SH) divided by PA

The correlation of the above metric with wOBA across 968 player seasons in my data set was r = 0.736. That gives an r2 of 0.542, which is a stronger relationship than exists using the ideal contact quality bins (r2 = .2005) or hard-hit rate (r2 = .2765).

Replacing the ideal contact bins with hard-hit batted balls in the formula (so, IPA% = (Hard Hit Balls + BB + HBP + SF + SH) / Plate Appearances) reduced the strength of the relationship to r2 = 0.427, but also came with the advantage that it becomes reliable (via Cronbach’s alpha) in just 190 PAs, which is faster than the earlier formula (440 PAs), hard-hit rate (210 PAs), and expected weighted on-base average (xwOBA, 340 PAs).

This metric is a more complete representation of the good things a batter can do in a plate appearance than what we can often see via individual statistics like walk rate, strikeout rate, or hard-hit batted balls.

The 2023 Twins and IPA%

Here’s how the 2023 Twins stack up by this metric:

Here’s how the Twins’ individual players show up with this metric:

This isn’t surprising, given what we know about this offense. It confirms all the things we’ve observed while watching them and that Baldelli is complaining about.

The approach at the plate just isn’t very good.

It’s not poor luck or something that will turn around on its own. This offense needs a significant adjustment or it will sink this season just like the lack of pitching depth sank last season.

---

John is a writer for Twinkie Town and Pitcher List with an emphasis on analysis. He is a lifelong Twins fan and former college pitcher. You can follow him on Twitter @JohnFoley_21.