There was much to admire when José Abreu signed with the Astros earlier this season. He fits the general team structure, he’s a great hitter and the contract is looking more and more reasonable in the context of the rest of the free agency market. However, there was an alarming note in several corners of the baseball internet: Abreu’s performance against fastballs, especially high-speed balls, dropped significantly in 2022.
I do not attribute this observation to one particular person, for I have seen it in many different places. This is indisputably true. According to Baseball Savant, Abreu’s numbers against both four marines and all fastballs thrown at 95 mph or faster are:
José Abreu vs. Fastballs
|Year||4-stitch RV||4-Stitch RV/100||High Speed RV||High Speed RV/100|
Oh no! The trends seem pretty clear; Abreu didn’t hit the fastballs very well in 2022 and had been somewhat of a drag against them the previous season. Is it just cooked? Is this fastball performance the famous canary in the coal mine that warns us that bad days are coming?
I really had no idea how to think about it, so I decided to dig deeper into the data. My first question was simple: What does first-year performance against fastballs mean as measured by Statcast’s run values for second-year performance against fastballs?
I did a very simple test to investigate this. Two years in a row, I called every player who saw 100 fastballs of the required type: either quads or fastballs that were launched at 95 miles per hour or higher. I did this for four-year couples: 2018-19, 2019-20, 2020-21 and 2021-22. The first question I asked was simple: What is the relationship between first year work value and second year work value? In other words, if you produce tons of runs against the fastballs in a year, as Statcast measures, should we expect that success to continue?
If you stop and think about it, it looks like a dunk. The question is not whether one is comparatively better against fast balls or secondary courts; how much they’re worth against the fastballs each year. To give you an example, Aaron Judge was an excellent hitter against four Marines this year, 3.2 laps above average in 100 shots. Last year, he was an excellent hitter against fast balls—two runs above average on 100 courses. You guessed it in 2020 – an above-average run of 6.1 per 100 steps. In 2019 – you get the idea.
The bad news though: Even with obvious gains like the Judge, there isn’t much correlation between first-year production and second-year production on four sewing machines. More specifically, it has a correlation coefficient of 0.17 and thus an r-square of 0.03. In layman terms, you can explain 3% of the change in next year’s production for four-stitch machines by looking at this year’s production of four-stitch machines. This is not too much! To eliminate potential noise, r-square only increases to 0.067, even if you limit it to hitters that see at least 500 four-stitch machines every two years. In other words, if you want to describe how hitters will perform against four-stitch fastballs in 2023, how they will perform against them in 2022 is not an adequate predictor.
As you can imagine, the data is no better for fastballs throwing 95 miles per hour or faster. While Abreu is down in both categories in 2022, that alone isn’t enough about what we should expect from him next year. Performance over the last three years is a much better indicator, but Abreu is better off there. Maybe he’s refusing, but if he’s refusing, we have to find another way to show it.
I decided to look at the odor ratio to examine the speed ball performance. The main problem with run value is that it is noisy, being results-oriented rather than process-oriented. The BABIP is rightly considered an unreliable and largely regressing indicator, and running a high BABIP when you engage a certain pitch will increase your run value and vice versa. If you’re focusing on the process rather than the results, the most obvious place to look is the olfactory ratio. How often a striker makes contact while swinging does not depend on a round ball hitting a round ball or where the defense is standing or which way the wind is blowing. It’s easier than that: hit or don’t hit.
Repeating the same exercise from above, I looked for swing hitters in 100 four stitchers in a row. The results are much more in line with the spirit of what we are looking for. The correlation between the olfactory rate of one year and the rate of the next year in four-stitch speedballs is quite high; r-squared checks out at a solid 0.52. If you want to explain the change in this year’s rate, looking at last year’s rate will get you more than half the way. This is an excellent mark compared to the almost pure noise of its operating value. As you can imagine, the data looks similar for high-speed balls.
Another useful metric is how hard the batter hits when he puts fastballs in play. In fact, it’s even more correlated from year to year than fastball olfaction. If you’re looking for stats to tell you how good a hitter is against fastballs, you should look at sniff rate or hard hit rate, not actual production.
Abreu did not get worse when it came to contacting the four sailors; He came up empty in 24.7% of his hits against them in 2022. This is slightly above the average for the last five years (23.4%) as well as the league average (21.8%). That’s pretty much his game for the last five years: he misses a little more often than average, but makes up for it by hitting the ball hard when he makes contact.
There is little evidence that the second part of the equation is changing. In line with his typical excellence, he was in the 86th percentile for hard hitting rate in four sailors this year. The same is true for high-speed balls; He dodges and misses slightly above the league average, but makes up for it by hitting them hard.
Will Abreu regress next year? I’ll give him a solid “maybe”. To be honest, I have no idea. A baseball player in his 30s, which means he’s always at risk of getting worse. Time wounds all heels and all. But if you’re looking for a way to tell this story, don’t use its fastball data to do so. Despite the drop in his results in 2022, I am as always impressed with Abreu’s batting ability.