Ratings & scoring

Six judgments per player

Roster entries carry a separate internal value for each of the six skills. A steal takes the value for the current attribute only. Box-stat clues help identify a player’s role; they do not directly calculate the pick’s rating. A player’s reputation as a scorer cannot answer every round.

Jumpshot

Shooting touch and range. Handles help turn that shot into effective shot creation.

Handles

Ball control and creating room. A famous scorer is not automatically the best handler on a roster.

Finishing

Converting chances near the rim. Athleticism helps that finishing become rim pressure.

Playmaking

Creating opportunities and reading the floor. It contributes to offense and directly to OVR.

Defense

Stopping opponents. Athleticism supports the build’s defensive score.

Athleticism

Physical tools that support rim pressure, defense and the simulation’s fatigue modifier.

How a pick earns its grade

The game counts how many players on the landed roster have a strictly higher rating in that skill. Tied ratings share a rank. It converts that rank into a percentile using the full roster size, then assigns a letter grade. The top rating always earns A+; the A+ band starts at the 95th percentile. A starts at 90%, A− at 80%, and lower bands descend through F.

For example, taking the best handler on a weak roster earns A+ even if the stolen value is only 74. That 74 still feeds into your build. Grades use the whole landed roster, including players you have already used, rather than recalculating the standard around the remaining selectable names.

Best Steal and The Reach order your picks by grade points, with the skill value breaking ties. They describe the strongest and weakest decisions within that run.

How six skills become one OVR

1. Jumpshot combines with handles to form shot creation. Finishing combines with athleticism to form rim pressure.
2. Shot creation, rim pressure and playmaking combine into offense. Defense also receives support from athleticism.
3. Qualifying skill combinations and decade combinations add small synergy bonuses. Shake & Bake rewards strong shooting and handles; Time Machine rewards picks spanning at least four decades.
4. A curve reduces gains at the high end before the final target weights combine offense, defense and playmaking.

For Best Player, the final mix is 52% offense, 36% defense and 12% playmaking, rounded to an integer and limited to 40–97. This is not a simple average of the six ratings. The current six-steal game cannot produce 99 OVR: the ceiling is explicitly 97, and diminishing returns make even that end of the scale difficult. Older shared builds may use earlier rules.

A position changes what matters

Point guards emphasize creation and playmaking. Shooting guards lean into shot creation. Small forwards favor a balanced two-way build. Power forwards and centers put more weight on rim pressure and defense. The position also changes the roster pool and boss ladder, so a center challenge is a different draft problem, not just a new label on the result.

OVR is not a guaranteed boss result

The gauntlet uses your build’s power against each boss, with a bounded seeded variance and matchup probabilities. Large power advantages are protected; close matchups can go either way. Budget flaws can change a matchup or introduce injury risk without lowering the displayed OVR. The simulation records scores, events and where the run ended.

The build code carries the inputs for replay. The same code and game rules reproduce the same result; new runs or attempts can change the simulation. A higher OVR is useful, but the result log explains what actually happened.

Two competitions, two tie-breakers

Daily ranks by OVR, then bosses beaten. Head to Head compares OVR, then the total points from your six grades; if both match, the duel is a tie. The Daily server derives results from a submitted build instead of trusting a score typed by the browser. This is validation, not a claim that abuse is impossible.