Scoring Methodology

Categories and redistribution

The eventual category plan is Performance 45%, Age/Upside 15%, Defense 15%, Value 10%, and Recognition Gap 15%. Only Performance and Age/Upside have reliable inputs today, so active weights redistribute proportionally to 75% Performance and 25% Age/Upside. Missing categories are N/A, never zero.

Normalization and eligibility

Each metric becomes an empirical percentile within the eligible current-season MLB role pool. ERA, WHIP, BB%, K% for hitters, and age reverse direction because lower is better. Available metric weights automatically renormalize if MLB omits a value.

Players below these thresholds are excluded from the primary ranking. Starters and relievers are normalized separately.

Age, missing data, and limitations

Age/Upside is the inverse age percentile within the player's eligible role pool: younger players score higher. Defense is N/A because basic fielding percentage is not a sufficient measure of defensive contribution. Value and Recognition Gap remain N/A until verified salary and awards sources are connected. Gem Score v1 therefore identifies young current-season performers—not yet true financial bargains or overlooked award candidates.

Similarity Score

Career Trajectory standardizes each available metric difference by the historical candidate pool's standard deviation. Per-metric closeness is max(0, 1 − normalized difference ÷ 3), and the displayed score is the weighted mean. At least five comparable metrics are required. Strongest similarities, biggest differences, and summaries come from those same differences. It is a similarity measurement, not a forecast.

Sources and status

Identity and standard statistics come from the official MLB Stats API. OPS+, ERA+, WAR, FIP, OAA, DRS, salary, and awards are not used. DiamondDNA scores are original analytical tools, not official MLB statistics.