Players whose results may lag their skills, ranked by the wins they add to the Pittsburgh the Winter Plan builds. Every candidate shows why he’s flagged, what
the data says about it (where he chases, whether a split holds up once regressed, what a fix is worth), whether his club would trade him, and the price in two
currencies never added together: the prospects Pittsburgh would send (the least value that meets his club’s ask, or up to 10% over it when a package fits both clubs’ depth clearly better, from the farm the plan leaves) and his salary,
with what it costs under each CBA lens. Pittsburgh’s own free agents (Kirby Yates, Gregory Soto, Brandon Lowe) aren’t listed: re-signing them is in the Roster Builder, whose “Lowe accepts the QO” preset is the Winter Plan’s branch for Lowe.
For general manager, pro scouting Model card2026 season · contracts as of Oct 21, 2026
Measured against the Winter Plan: 81.7 wins, 44% playoff odds, 39 on the 40-man, 2027 payroll $138.6M (owners’ basis $137.2M: $34.0M under the owners’ proposed floor, on their terms). Prospect pool after the plan: 48 prospects, $278.9M; Jhostynxon Garcia and Jacob Gonzalez already go to Isaac Paredes.
30 of 159 candidates · ranked by wins added to the Winter Plan
Wins added on the plan’s roster (its second and third basemen held; the package’s 40-man players come off first). Players their clubs are unlikely to trade, and trades no package of three prospects can pay for, come after every other row. The page carries the 80 best who could be had and the 20 best of the rest; filters search those. Click a row for the checks.
#PlayerStuff · best pitchAvailability · plan needPriceWins added
Wins added: the plan’s depth-chart engine, one season of team playing time, the same measurement as the Free-Agent Board. Trade price: his club’s ask is his surplus value
over his remaining control ($9.0M per WAR in 2027, 4% growth, 8% discount; a prospect at the Trade Builder’s value), never less than
half his 2027 value less any overpay, plus 10%; Pittsburgh sends the least prospect value that meets it, up to three prospects, top
3 protected (2026 draftees count, valued at their pick on the draft curve scaled to FanGraphs’ table until FanGraphs grades them); packages are prospects only, and major leaguers can be added in the Trade Builder. Free-agent price: the Free-Agent Board’s contract model, plus the draft pick a qualifying-offer signing would cost Pittsburgh (the whole pick when an offer is likely, the pick at its odds when it’s a toss-up; shown, not added to the salary). His “in 2027” salary is his own; when his package sends 40-man players, their salaries leaving is shown beside it and counted in the payroll after him.
A player missing from the contract data gets the least control his MLB debut allows, marked “(est.)”, with arbitration seasons priced by Contract Studio’s arbitration model. Salary under the
owners’ proposal (a proposal, not a rule): dollars that keep a below-floor payroll below the floor cost nothing if the club has no other use for them, and face value if another use would take
them, such as absorbing a contract for talent or an extension; both ends are shown. Plan needs are read off the plan’s roster. Pittsburgh's needs from 2028 on: each position the depth model's Winter Plan grid (depth.json) has thin (best option under half the regular bar) or open (nobody controlled above replacement). A candidate fills one in a season when his expected WAR there, valued as the depth model values a major leaguer over his contract or control, reaches half the bar and beats what the spot holds. Tags and a filter only: the rank is wins added in 2027. With one partner club, the package also weighs the trade fit: one that suits his club’s needs and Pittsburgh’s depth clearly better may cost up to 10% over the ask, and the row says so, with the difference from the rule’s default package (more or, as that is a tie-break pick among near-equal totals, sometimes less). The signals explain why a player is listed; the checks test them;
neither sets the rank. “Stuff ahead of results (no forecast gain)” means the gap is there but the results check finds less than a run of upside in it, so it never lists a pitcher on its own. Model card →