Average with runners in scoring position is noise
No number gets pulled out of a broadcast graphic with more moral weight than batting average with runners in scoring position. A hitter is up with a man on second, and the announcer reads it like a character reference: he’s hitting .320 with runners in scoring position, this is his moment — or, the mirror image, he’s down at .210 in these spots, watch him press. The stat is treated as a window into temperament, a measure of who rises and who shrinks. But RISP average is one of the least stable numbers in a hitter’s line. It is a small slice of an already small season, it barely repeats from one year to the next, and it drifts back toward the hitter’s ordinary batting average with almost mechanical reliability. Most of what it “reveals” is the ordinary scatter of a few dozen at-bats.
The sample is tiny by construction
A full-time hitter gets somewhere in the neighborhood of six hundred plate appearances a season. Only a fraction of those arrive with a runner on second or third, and once you strip out walks and other non-batted outcomes, the RISP batting average is often built on well under a hundred and fifty at-bats. Batting average is already a jumpy statistic over a full season — the whole reason it refuses to die as a headline number is that people mistake its volatility for meaning. Cut it down to a third of the sample and you have amplified every wobble. The difference between a .320 and a .240 RISP line can be six or seven hits across an entire summer: one bloop a month, a couple of hard outs that found gloves. That is not a personality. That is variance doing what variance does in small samples.
It doesn’t repeat
The cleanest test of whether a stat measures a real, durable skill is whether it predicts itself — whether this year’s number tells you anything about next year’s. RISP average fails that test badly. Hitters who post gaudy scoring-position averages one season are, as a group, unremarkable in those spots the next, and the hitters who “couldn’t come through” bounce right back. The lack of year-to-year correlation is the signature of noise. If clutch hitting in the RISP sense were a stable trait, the same names would sit atop the leaderboard season after season, the way they do for exit velocity or strikeout rate. They don’t. The leaderboard reshuffles every year, which is exactly what you would expect from a coin that lands heads a few extra times and then doesn’t.
Regression is not a moral failing
The reason a hot RISP average cools is not that a hitter loses his nerve. It is regression to the mean, the same force that governs BABIP and every other average built on a limited number of balls in play. A hitter’s true talent in scoring-position spots is, for almost everyone, very close to his true talent everywhere else — the pitcher is still throwing, the bat is still the bat. When the small-sample number sits far above or below that true level, the safest bet is that it will drift back toward the hitter’s overall average as the at-bats accumulate. Announcers narrate that drift as a slump or a clutch awakening, assigning agency to what is really just arithmetic catching up. The hitter did not find or lose a skill. The sample got bigger.
The context isn’t neutral either
Even the raw comparison is contaminated, because at-bats with runners in scoring position are not a random draw from a hitter’s season. Defenses may hold runners and change positioning; managers deploy their best available relievers in exactly these spots; first base is sometimes open, inviting a pitch-around or an outright intentional walk that removes the best hitters’ most dangerous chances from the sample entirely. The RISP at-bats a star sees are disproportionately against a team’s toughest arm in a leverage moment, while a weaker hitter down the order may face a tiring starter with the bases loaded. Comparing raw RISP averages across hitters compares different pitching, different game states, and different willingness to let them swing at all.
Clutch, if it exists, is tiny
None of this proves that no hitter is ever genuinely better or worse with the game on the line. Careful work on clutch performance has looked hard for a persistent skill and generally found that if it exists, its effect is small — far smaller than the season-to-season swings in RISP average that fans treat as proof of it. This is the same lesson the smallest samples in sports keep teaching: the more dramatic the situation, the fewer the repetitions, and the louder the noise relative to any real signal. A few dozen high-leverage at-bats simply cannot separate a true clutch talent from a lucky month, and the stat sheet does not come with error bars to remind you of that.
What actually predicts runs
If the goal is to know who will drive in runs going forward, the honest answer is boring: the hitter’s overall quality. A batter’s full-season line — his on-base skills, his power, his strikeout and contact rates — predicts his future scoring-position production better than his past scoring-position production does. The best hitter in ordinary spots is, over any meaningful stretch, the best hitter in big spots too, because the underlying skill is the same and the true-talent gap between “normal” and “clutch” is nearly zero for almost everyone. Chase a high RISP average and you are chasing the part of the line most likely to vanish. Trust the whole line and you are trusting the part that persists.
Why the myth is so sticky
If the number is this unreliable, why does it survive every season? Partly because it is available — it sits right there on the broadcast rail, ready to be read into any at-bat — and partly because human memory is built to find clutch narratives. We remember the hit that drove in the winning run and forget the nine flyouts that preceded it, and a hot RISP line gives that memory a number to point at. The stat also arrives pre-loaded with a story: effort, nerve, wanting it more, all the qualities fans want to believe separate players. Handing that story a decimal makes it feel measured rather than imagined. The result is a metric that persists not because it predicts anything but because it confirms what we already wanted to say about a hitter, which is the most durable kind of bad statistic there is.
The honest read
Batting average with runners in scoring position is a real record of what happened, and over a career the sample eventually grows large enough to carry a faint signal. But in any single season it is a thin, jumpy slice that does not repeat, regresses hard toward the hitter’s ordinary average, and is skewed by the pitching and game states those at-bats happen to contain. Read as a clutch rating, it invents personalities out of variance and rewards or punishes hitters for a handful of bounces. Read it for what it is — a small, noisy split — and lean on the full-season line when you want to know who will actually come through. The hitter you can trust with a runner on second is, almost always, just the hitter you can trust.