How Legends Deck Card Ratings Work — The Full Math
Every Legends Deck card rating starts from real Statcast data — no scout opinions, no narrative, no hot-streak halo. This is the full breakdown of how hitter and pitcher ratings are calculated, why they sometimes surprise you, and how to read a card to know whether it's a buy or a sell.
Card ratings in most sports games are opinions. A small team of rating "experts" watches tape, reads scouting reports, and assigns numbers. The numbers are biased by narrative — hot streaks, big-market bias, positional scarcity myths, last-year's MVP halo.
Legends Deck does it differently. Every rating starts from real, publicly available Statcast data — no scout opinions, no narrative adjustments, no hot-streak halo. This post walks through exactly how that data becomes a rating so you can evaluate cards before you spend a coin.
The Core Pipeline
Every attribute rating on every card follows the same four-step pipeline:
1. Rank against the league. A player's Statcast metric (barrel rate, sprint speed, OAA, fastball velocity, etc.) is converted to a percentile rank against the qualified active-MLB population.
2. Map percentile to a 0–99 scale. That percentile runs through a fixed curve that stretches the elite tier — so a 90th-percentile skill reads as a 93 and a 50th-percentile skill reads as a 74, not a flat 50-to-99 line. The curve is the same for every player and every metric; it never changes card to card.
3. Regress small samples toward the position mean. A raw percentile from 60 plate appearances is noisy, so each component is regressed toward its position's average in proportion to sample size. A 60-PA hot streak can't mint a 99.
4. Blend into an overall. The component ratings combine with position-specific weights — a shortstop's fielding counts more than a DH's — and anchor to expected run production (xwOBA) so the overall tracks actual offensive value, not just the sum of its parts.
The percentile rank is computed against the entire active MLB population, refreshed nightly, with a minimum plate-appearance or batter-faced threshold to avoid small-sample noise.
Hitter Attributes
| Attribute | Input metric | Why |
|---|---|---|
| Contact | xBA (expected batting avg) | Rewards quality of contact over luck |
| Power | Barrel% per BBE | Best predictor of sustainable power |
| Vision | Whiff% (inverted) | Lower whiff = better bat-to-ball |
| Discipline | Chase% (inverted) | Lower chase = better zone recognition |
| Speed | Sprint speed (ft/sec) | Direct measure |
| Fielding | Outs Above Average | Modern defensive standard |
| Arm | Arm strength (mph) or Throwing velo | Statcast-measured |
| Clutch | RE24 performance rate | Leverage-weighted |
Positional versions of these ratings apply:
- Catchers also get a Framing rating from framing runs above average
- Shortstops/2B/3B get position-specific OAA splits
Pitcher Attributes
| Attribute | Input metric | Why |
|---|---|---|
| Velocity | Fastball velo (mph) | Direct Statcast measurement |
| Control | Zone% + walk rate | Composite |
| Movement | Induced vertical break + horizontal break | Per pitch |
| Deception | Release extension + delivery consistency | Perceived velo |
| Stamina | Pitches per start + decline curve | How late they hold velo |
| Each pitch | xwOBA allowed on that pitch | Actual results |
Each pitch type (4-seam, slider, changeup, curve, etc.) gets its own sub-rating based on xwOBA-allowed percentile.
Why Some Ratings Surprise You
Luis Arraez is not elite in Legends Deck. Despite hitting .314, his Barrel% is 15th percentile and his Hard-Hit% is bottom 5%. The game correctly rates him as a contact specialist, not a star. His Contact rating is 85+, but his Power rating is in the 30s.
James Wood's card looks better than his stats. His 95th-percentile exit velocity and 90th-percentile barrel rate are elite — even if his actual HR total lags. The rating sees underlying skill, not lucky outcomes.
Spencer Strider cards bounce around. His velocity ratings track his actual velo trajectory. When his fastball sat 98 in 2023, his Velocity rating was 99. When it dipped to 95 post-surgery, his rating dropped accordingly. Live rates, live ratings.
What This Means for the Marketplace
Because ratings are deterministic functions of real stats, card prices should be deterministic functions of real Statcast percentiles. They mostly are — but market inefficiencies show up when:
- Narrative drives price. A hitter with an 89 overall but mediocre barrel rate trades at 8K. The "fair" price based on expected sim performance is 5K.
- Rookie cards lag ratings. Young cards trade below the rating-implied price because the market discounts uncertainty. Once a rookie has 200+ PA of stable performance, his card jumps.
- Daily refresh lag. Ratings update at 3 AM Pacific. Breakout weeks show up in ratings before the market fully prices them.
These inefficiencies are the bread and butter of players who finish in the top 1% of the leaderboard. They're reading cards against underlying Statcast percentiles and finding gaps.
How to Read a Card Before You Buy
1. Look at Barrel%, not Power rating. If Power = 85 but Barrel% is 65th percentile, the rating is inflated by other factors (park, luck). Be cautious.
2. Check xwOBA vs. wOBA gap. If xwOBA > actual wOBA by 30+ points, the card is undervalued — regression coming.
3. Mind the handedness park split. A lefty Red Sox hitter has a much different true park factor than a righty. See park factor.
4. Age-adjust velocity. Pitchers over 32 lose 0.5–1.0 mph annually. Factor this into your Velocity-rating trust.
5. Watch 14-day rolling metrics. The market reacts to 30-day numbers. You can front-run it by checking 14-day splits on Baseball Savant.
Why This Matters
The reason Legends Deck has a healthy marketplace and competitive leaderboard is that ratings are honest. Every player, regardless of budget, is betting on the same underlying math. There's no insider info, no "trust us, Player X is elite" — just public Statcast data, run through the same calibrated pipeline for every card.