// DRAFT FILTER STUDY — 2026 NCAA CLASS
A self-directed research project testing statistical filters against real draft outcomes, then applying the surviving filters to the current 2026 season to build a forward-looking watchlist.
Four spreadsheets, two questions. First: of the traits I could screen for, which combinations actually separated drafted players from everyone else? Second: who does that same screen flag right now?
Filter Research (FR) runs four candidate filters against every freshman-year (2024) hitter season belonging to a player who ended up in the 2026 draft class, and checks what share of each filtered group actually got drafted.
Filter Research (SO) does the same thing using those same players' sophomore (2025) seasons.
2026 Season FR Watchlist applies the freshman filters to current true freshmen (2026 season stats) — hitters who won't be draft-eligible for years, but who already show the traits that correlated with getting picked.
2026 Season SO Watchlist applies the sophomore filters to current sophomores — the group closest to actual draft eligibility.
Every filter sheet is its own screen (e.g. K% ≤ 20%, ISO ≥ .075, BB% ≥ 7%...). A player can clear more than one screen — the 3/4 filters badge on a row shows how many of the four he satisfies within that dataset. On the historical sheets, an amber left-edge + DRAFTED tag mark players who were actually picked.
Three patterns showed up consistently across both the freshman and sophomore back-tests.
Share of each filtered group that was actually drafted in 2026, ranked loosest → tightest.
Draft rate for players grouped by how many of the four filters (within the same season) they satisfied at once.
Combined FR + SO filtered pool. Up-the-middle and 3B outperform the corner-outfield / "IF" catch-all profiles.
| Name | Pos | School | FR filters met | SO filters met | 2026 Draft result |
|---|
Every filter below was run against the freshman-year numbers of players who ended up in the 2026 draft. Rows with an amber edge were drafted; the tag next to each drafted name is (round, overall pick) from the 2026 MLB Draft.
Same exercise, one year later — sophomore-year numbers for the same 2026 draft class. Tags show (round, overall pick).
Outcome metric: Day 1 rate — the share of each filter's pool taken in Rounds 1–3 (picks 1–103), the portion of the draft actually televised and treated as "Day 1" by MLB. Raw draft rate (drafted anywhere in 20 rounds) mixes in mostly organizational-depth late picks and overstates how well a filter finds impact talent, so Day 1 rate is the number used to rank and judge the filters below. Median pick (among drafted players) is shown for extra context.
Of the 15 college hitters taken in picks 1–37, 10 appear in a back-test pool. Vahn Lackey (#3), Derek Curiel (#5), Zion Rose (#6), Tyler Bell (#10) and Jake Schaffner (#20) appear in neither pool. These are high-contact / high-average profiles; the power-and-walk filters undervalue them.
Current true freshmen whose 2026 season already matches the filters that historically correlated with getting drafted. These players are years from eligibility — this is an early flag, not a near-term projection.
Current sophomores clearing the sophomore-year filters — the group nearest actual draft eligibility (2027–28 range).