How to read a projection: floor, ceiling, and why the range matters

When most people look at a projection, they see one number: 41.2 points, say, and move on. That number is genuinely useful, but it's only the middle of a much wider story. Every projection Benchlytics produces is really a distribution, a range of outcomes with a most-likely value in the middle and meaningful probability spread out on either side.

Learning to read that range is the single biggest upgrade you can make to your research process. It turns 'who has the higher number' into a more honest question: 'who gives me the outcome I actually need this week?'

Floor, ceiling, and median

We describe each projection with three reference points. The median is the balanced midpoint. The floor is a conservative outcome, roughly what a player produces on a quiet night when things don't break their way. The ceiling is the optimistic end, the version of the game where usage, matchup and game flow all line up.

Two players can share an identical median and be completely different assets. One might be a metronome, a tight range clustered around the median. The other might be volatile, with a low floor and a sky-high ceiling. Neither is 'better' in the abstract; they're better in different situations.

Match the shape to the situation

If you're building a lineup where you're comfortable and just need dependable production, tighter ranges and higher floors reduce the chance of a single roster spot sinking your night. If you're chasing a big result and need players to outperform, higher ceilings matter more than safe floors, even at the cost of a lower median.

This is why we surface consistency indicators alongside the raw number. A boom/bust tag isn't a warning; it's context. It tells you what kind of week you're signing up for when you lock that player in.

The honest caveat

Projections are estimates, not predictions. Sports are noisy, and even a perfect model is wrong on any given night more often than people expect, that's the nature of variance, not a flaw. The value isn't in being right about one player on one night; it's in making better-informed decisions repeatedly over a season. Use the range as one input into your own judgment, never as a promise of an outcome.