How the Balladex compare verdict is decided

A comparison puts two players' shared stats side by side and counts who leads more categories. The stats and awards are real. The verdict is arithmetic on them, not an opinion.

By Beau Thoutt · Used on Compare players

Inputs to the Compare verdict

Each player's comparison metrics: labeled stats such as points per game or career passing yards, each with a direction (higher is better, or lower is better for stats like interceptions thrown).
Each player's season table, used to put both players on a career basis when that is possible.

How each category is scored

Only metrics both players share, matched by label, are compared.
When both players' career values can be derived from their season tables, both switch to career figures and the label gains "(career)". If either cannot, both keep their stored figures, so the two sides are always on the same basis.
Each category goes to the player who is better in its stated direction. Equal values are a tie.

How the verdict sentence is written

Equal category counts: the players split the categories evenly.
One player leads every decided category: a sweep, noting any ties.
Otherwise: the leader holds the edge, leading in N of M categories, and the other player's winning categories are named. Curated marquee pairs add a note that this is the marquee debate.

Which player pairs get a comparison page

Player pool. NFL: current depth-chart starters, the top 10 per position by per-game production, and decorated retired legends. NBA: the top 150 active players by production with at least 20 games this season, and the 60 highest-stature retired players.
A pair must be in the same league and position, with both players in the pool, sharing at least three metrics and each having at least one season row.
The pair must also be curated (a marquee debate) or one of a player's three nearest peers by current production. Legends are paired by career stature, and actives pair with actives.
Each player has at most 6 indexed pairs, and organic pairs stop at 600 per league. Curated pairs are exempt from both caps.
Pairs already indexed are kept while both players still qualify, and are removed only at season end, so comparison URLs don't change week to week.

Known limitations

Every category counts the same. Winning in rebounds counts as much as winning in points.
Career totals favor longer careers, and per-game figures from different eras are compared without adjustment.
The verdict covers only the shared metrics. Defense, intangibles and the playoffs count only if a shared metric captures them.
Any two players can still be compared with the compare tool, but only pairs that meet the rules above get their own indexed page.

Compare verdict: frequently asked questions

How does Balladex decide who wins a head-to-head comparison?

Each stat both players share is a category. The player who is better in more categories, counting lower-is-better stats like interceptions in the right direction, holds the edge. The verdict is arithmetic on real stats, not an opinion.

Why are some stats labeled (career)?

When both players' career figures can be derived from their season tables, both sides switch to career values so the comparison is like-for-like. If either player's cannot be derived, both keep their stored figures.

Why doesn't every pair of players have a comparison page?

Pages exist only for pairs in the same league and position, drawn from current starters, top producers and decorated legends, that share at least three stats, and that are a marquee debate or nearest peers by production. Each player has at most six pages.

Other Balladex models

Balladex Impact index — A 0–99 composite of an NBA player's skill ratings plus a capped bonus for major honors.
MVP odds — NBA and NFL MVP ladders: a production-based score per contender, turned into probabilities across the field.
Title and Super Bowl odds — NBA title odds from a team rating (last season, roster talent, this season) run through the bracket; Super Bowl odds from point differential.
Power rankings and team grades — Teams ordered by point differential, with an A–F grade from win percentage.
Fantasy projections — Forward-looking fantasy points, value over replacement, tiers and a model draft round for NFL and NBA.
NFL production index — A cross-position NFL rating: each player's production as a percentile of his position, curved to 20–99.

Source. Transcribed from lib/compare.js (buildComparison, verdict); scripts/build-compare-index.mjs (pair rules) in the Balladex codebase. Found an error? Email hello@balladex.com. See the full methodology, including data sources and the corrections policy.