Willi Castro creates bases at a league-average clip — better than 46% of qualified hitters. The strikeouts pile up, though — only 30% of qualified hitters whiff more often.
Each spoke is a skill estimate adjusted for sample size — not raw season stats — so farther out = better.
Percentiles vs. 463 qualified hitters.
Data through 2026-07-15.
Plays like…
Same archetype, nearest by rate — the hitters whose profile looks most like this one.
Every plate appearance ends one of three ways: a strikeout (0 bases), a walk (1 base), or a ball in play worth its estimated bases.
The bar splits his value into those pieces, versus the league.
More bases is better. In the table, 100th percentile = best in MLB.
Ball-in-play basesWalk bases
Bases per plate appearance
.336
.084
Willi Castro
.351
.084
League average
What a replacement bat would have done
over Willi Castro's 320 plate appearances
134
Willi Castro
115
Replacement level
▲ +19 bases above replacement (created 134 vs 115)
Willi Castro
League
Percentile
Bases per ball in play(higher = better)
.540
.510
46th
Walk rate(higher = better)
8%
8%
46th
Strikeout rate(fewer = better)
28%
23%
30th
Platoon splits
Platoon splits (EB/PA)
Dot = shrunk true-talent estimate; bar = 89% credible interval.
Faint diamond = his raw, unshrunk rate against that hand — small samples make it noisy, which is why the model pulls the dot toward a more trustworthy estimate.
Shrunk splits beat raw splits decisively at low PA counts and converge with them as playing time builds — individual platoon gaps need roughly 1,000+ PA vs a hand before the raw number alone can be trusted.
Willi Castro switch-hits — he can choose the platoon-advantaged side almost every plate appearance, so a small gap here is expected. It is not missing data or a modeling error.
Headline EB/PA (0.437)
vs LHP107 PA0.420
vs LHP: shrunk estimate 0.420 EB/PA, 89% credible interval 0.367 to 0.475, based on 107 plate appearances. Raw (unshrunk) rate: 0.301 EB/PA.
Raw vs LHP rate (0.301) is off this scale — small samples like this are exactly why the shrunk estimate (the dot) is the trustworthy number, not the raw one.
vs RHP202 PA0.445
vs RHP: shrunk estimate 0.445 EB/PA, 89% credible interval 0.397 to 0.491, based on 202 plate appearances. Raw (unshrunk) rate: 0.497 EB/PA.
0.3520.3910.4290.4670.505
0.025 EB/PA better vs RHP than vs LHP (shrunk estimate).
89% credible interval: -0.021 to +0.067
That interval crosses zero, so the model can’t confidently say which side he’s actually better against.
How he hits
Hits it to Pull most often, and does the most damage there too.
Best batted balls
Where every ball Willi Castro put in play landed this season.
Dots are colored by estimated bases (EB) — pale slate (easy out) to dark teal (home-run territory).
Filled dots are hits; hollow rings are outs — a dark ring in the outfield is a crushed ball that got caught.
Hover a dot or a row to link the two; click to watch on Baseball Savant.
