Older but not Wiser+
Analyzing how Hitting+ components change with age

This is the second part of the Hitting+ Series: Part one can be read here.
In my Spoiled Milk in Your Fine Wine piece last offseason I described the discipline growth young hitters experience as they get older. The quick hypothesis was that a hitter’s understanding of the strike zone - and they’re ability to determine if an incoming pitch is a ball or strike - increases with age, which means they should chase at balls less and swing at strikes more as they get older.
Here is the model version of what I described above; If you line up every hitter by age and take their average Decision+ (read part one to see how it was defined) at each age, plate discipline rises from 93.9 at age-22 to 105.1 at age-37. Read that curve at face value, and you would conclude that hitters spend fifteen years learning the strike zone, and that the old guys have finally figured it out. This was my understanding of a hitter’s journey and probably 99% of baseball fan’s understanding of the hitter’s journey.
The problem is the 37 year olds. A 37 year old who is still getting 800 pitches thrown at him is not a random 37 year old. He is the one who survived, and the reason he survived is that his eye was good enough to keep him employed while the rest of his cohort washed out at 31. The curve is not measuring hitters getting better. It is measuring bad hitters disappearing, savvy readers already know this phenomenon has a name and it’s called survivorship bias.

We analyzed the fighter planes and batters that safely returned to home base
I promised at the end of the last piece that I would get into this, because it is the whole reason the aging question is interesting. Since Decision+ can be computed all the way back to 2017, the question for today is simple to say, how does a hitter’s skill age based on these components?
Curve Your Enthusiasm
The way to fix survivorship bias is by following the same guy year-after-year.
Instead of comparing all the 30 year olds to all the 25 year olds right now, you take each hitter's season and pair it with his own next season, and you measure how much that specific hitter changed. Ryan McMahon at 30 against Ryan McMahon at 31. Do that for every consecutive pair in the data and average the changes at each age. A hitter who collapses and drops out still counts for the one transition he made before he left, instead of vanishing from the sample and taking his decline with him. This is the delta method, and it is not mine, writers like Tom Tango have been beating this drum for twenty years for measuring age variation instead of simply averaging by age.
Running the delta method on Decision+ across ten seasons and 1,704 pairs, and the rising curve flattens out and then tips slightly down. From age 22 to 36 the same hitters lose a cumulative 3.1 points of Decision+. On a metric with a standard deviation of about 15 points, that is a fifth of one standard deviation across an entire career. Plate discipline, followed honestly, is close to a flat line.

The grey line is the conventional wisdom I described at the beginning of the article, but the blue line is the adjusted data that shows the actual trend, which is somewhere between flat to slightly declining as a hitter gets older.
The Bat and the Eye
So plate discipline is flat, but what about the bat.
Bat speed is a physical tool. Bodies get older, fast twitch muscles wear down, and there is no version of the aging story where a hitter swings harder at 34 than he did at 26 (at least not without some medical funny business). If the framework is any good, the bat components should decline where the eye does not, and the split should be clean. Here is what the delta method says, in points per year of age, run across the full age range so the young gainers and the old decliners both count.

Going down the points per year column, Power+ falls at about a third of a point a year, because the bat speed underneath it is eroding, and it is the most certain single component on the page. Hitting+, the composite, slides three quarters of a point a year because the bat is doing the pulling. And Decision+ sits there at a slope you cannot distinguish from zero.
The takeaway from the table is that the bat may go but the eye stays. A hitter loses the tool he was issued and keeps the trait he brought. Which, if you will indulge me for one sentence, is exactly why you grade inputs separately in the first place. A single number that blends the bat and the eye tells you a 33 year old is declining. The components tell you which half of him is declining, and therefore what he is going to look like next year.
The clearest way to see it is not the slope but the two ends of the curve. From 22 to 24, the average hitter is still gaining bat speed, about three points of Power+ a year as he fills out. By 34 and up he is giving back more than a point a year. The eye does not do this. It has no 22 to 24 ramp and no cliff at 34. It just sits.

It’s a Lifelong Decision+
Back to the batter’s eye, because the flat line has a consequence that is more fun than it sounds.
If plate discipline does not improve with age, then a young hitter with a great eye did not earn it in the minors and he is not going to suddenly acquire one at 27. He showed up with it. And the data says this is exactly what happens. Follow hitters who started below a 95 Decision+ and they drift up about 5 points over their careers, which sounds like learning until you remember that a guy who posts a 92 was partly unlucky and would drift up on regression alone. Follow the ones who started above 105 and they drift down about 4. Both groups sag toward the middle by the amount you would predict from regression to the mean and no more. A hitter's first season Decision+ correlates 0.64 with his last. They mostly end where they began.
So when you see a rookie with an advanced eye, believe it early, because it is a trait and not simply small sample noise. Jackson Holliday posted a 111 Decision+ at 22. Nolan Schanuel put up a 107 at 22 and did it again the next year. Jacob Wilson, 106 at 23. These are not guys who are going to learn the zone, they already know it.
This is the fantasy inversion, and it cuts the way most aging content does not. Discipline is a buy you can make before the track record exists, and bat speed is a sell you should make while the track record still looks great. The young eye is real now. The old bat is already leaving even when the results have not noticed yet.
Miguel Vargas is what that trade looks like when it pays. Rewind to 2024 and his line was a poor .265 xwOBA and a 100 Hitting+, the kind of season that gets a former top prospect quietly labeled a bust, but the eye was already there, a 106 Decision+ at 24, sitting on top of a bat that had not shown up yet at a 98 Power+. Then the power arrived. His Power+ climbed to 98 in 2025 and 115 this year, and because it landed on a discipline foundation that was already poured, it did not just make him playable, it made him a 146 Hitting+ and a .387 xwOBA, one of the best hitters in baseball in 2026. The breakout everyone is watching now was legible in 2024 to anyone grading the eye instead of the power. You were not waiting to find out whether Vargas could hit. You were waiting for the bat to catch up to the eye, and this year it did.


The Anticipation That Wasn't
Timing+ looked, for a little while, like it might improve with age, and there is a lovely story available if it does. The veteran cannot catch up to the fastball anymore, so he learns to anticipate, and his contact point drifts back to where it should be through guile instead of quickness, but the story does not survive the data. The age trend in Timing+ is a rounding error. The two transition years point in opposite directions. Timing+ is stable, which the first piece showed, but it does not systematically get better as a hitter ages. Anticipation is not a skill the old bat grows into to replace the speed it lost.
The Graveyard Not in the Data
One more honest limit, and it is the one that matters most, because it runs the same direction for every number above.
The delta method fixes the survivor problem in the cross section, but it has a survivor problem of its own, pointing the other way. A hitter who truly falls off a cliff gets released and never produces a second season, so his decline is not in the pairs at all. Of the qualified seasons that could have had a follow up, 73.7 percent do. The ones who did not come back were worse across the board before they went, by about eight and a half points of Hitting+, and the no-show rate climbs steadily with age.
Which means every curve in this piece is too kind. The real aging decline is steeper than what is drawn here, because the hitters who aged worst are the ones who are missing. You can bound the problem and you can tell which direction it bends, but you cannot resurrect a season that was never played. So read the slopes as a floor. Bat speed falls at least this fast. Probably faster.
And Steven Kwan still says hello. The Kwan problem from part one, where Hitting+ underrates a no-power contact hitter, does not go away when you age him. If the metric already dislikes low-power contact guys, it will make some of them look like they aged badly when they simply were never the metric's kind of hitter. Keep that asterisk on any "aged poorly" leaderboard you build from this.
The Short Version
The rising plate discipline curve you have seen is survivor bias. Follow the same hitters and discipline is nearly flat across a career
Bat speed and Power+ decline with age at about a third of a point a year, the most certain thing in the data. The eye holds, the bat goes
Young hitters do not learn the zone. They show up with it. First season Decision+ correlates 0.64 with last. Believe a rookie's eye early
Timing+ does not improve with age. The story that veterans learn to anticipate does not survive testing
Every curve here is too optimistic, because the hitters who aged worst got released and are missing from the sample. Read the declines as a floor
Back soon with the next question these components can answer. The daily grades, aging views included, live at hitting-plus.vercel.app.