$3MCheap-talent surplus — $/yr of value from cost-controlled (pre-arb/arb) players above their pay; the selection engine, and the stickiest GM skill we track156th/195 execs
$-4MDrafted surplus — cheap-talent surplus from his own draftees (the develop-from-the-draft engine)192nd/195 execs
$-1MTraded-for surplus — cheap-talent surplus from cost-controlled players he acquired via trade170th/195 execs
$7MInt'l surplus — cheap-talent surplus from his own international signees7th/195 execs
-16.1WAB — realized wins (TAB + the non-repeatable luck of converting talent to wins)138th/195 execs
0.07Acquisition Leverage — playoff pts per acquired WAR, graded once in the year each deal's wins arrive (the repeatable decision skill)179th/195 execs
Roster value distribution — career average — depth vs. stars
Every roster slot ranked best→worst by WAR, averaged across Roy Smith's seasons. Bars = his rosters; the black line = the league-average player at each slot. Staying above the line deep into the roster is the depth edge.
Players acquired — 18 players, by realized fWAR for the franchise (scroll for more)
WAB = wins above what the team's payroll predicted (resource-controlled). WAR surplus = era-neutral value (WAR minus payroll in win-equivalents). Roster surplus $M = dollar value of WAR over cost. Build channels (realized fWAR): HG = home-grown via amateur draft / international signing; Trade = acquired minus surrendered (incl. purchases/waivers); FA = signed minus walked; R5 = Rule-5 / minor-league-draft pickups. Credit follows the acquisition year — home-grown to the regime at the player's draft/signing (not whoever was in the chair when he peaked), trade/FA/R5 to the GM who made the move. Players acquired as prospects before debuting (e.g. Smoltz, Bagwell; Rule-5 picks like Uggla) are counted as trade/R5, not home-grown. All are franchise-level outcomes during the stint, not solely attributable to one executive.
Each point is a team-season: x = relative payroll (× league average), y = win %. Gray = every team 1985–present; orange = Roy Smith's teams. The dashed line is the win% that payroll predicts — points above it won more than their budget bought, and that vertical gap (in wins) is WAB.
Each point is a team-season: x = relative payroll, y = team WAR (talent, per 162 games). Gray = every team; orange = Roy Smith's teams. The dashed line is the talent payroll predicts — points above it built more WAR than their budget should buy, and that gap is TAB.
Why the Acquisition Leverage— every acquisition, graded in the year its wins arrived
Each point is one acquisition (FA/trade), placed in its arrival year: x = WAR it added that year, y = its fair share of the playoff-odds boost its season's acquisitions bought. Playoff odds are an S-curve in roster WAR (flat far from contention, steep near the threshold), so the same WAR moves the needle far more on a near-contender — that's where a GM spent. Color = leverage (green = high, red = low). The dashed line is this GM's career average of 0.1 pts per bought WAR; points above it beat their own average.
The grey dotted line is the win curve (playoff odds vs roster WAR), for reference. The ■ is the base — existing talent (home-grown + returning) already on hand before that year's moves. Each orange step adds one acquisition: across = its WAR, up = its Acquisition Leverage (its fair share of the playoff odds the class bought — the same number in the table). The steps stack from the base's odds up to the finished roster's, so they add up with no double-counting — one star never banks the whole crossing. Smallest WAR first; the curve is steepest near the ~45-WAR threshold, where bought wins are worth most. It opens zoomed in on the moves — double-click to zoom out to the whole win curve.