Goalie Fantasy Score, explained

Why we use 0.6 × saves − 1.0 × goals against, how it compares to DraftKings and FanDuel scoring, and where the formula falls short.

Goalies don't fit the standard prop model. A skater's stats are clean — goals, assists, shots, hits. A goalie's primary stats are defensive and they're correlated in confusing ways. More saves can mean a great game (high shot volume, made the stops) or a terrible game (defense collapsed, goalie was under siege all night).

So we needed a single number that combines saves and goals against in a sensible way. The result is the Lyzos Goalie Fantasy Score:

Why these weights?

0.6 per save

This is roughly the credit a goalie gets for stopping a shot in DraftKings-style scoring (which uses 0.7 per save). We use 0.6 because we're not modeling DK fantasy points exactly — we're modeling per-game performance value, and slightly under-weighting saves keeps us from rewarding goalies who simply face a high volume of weak shots.

1.0 per goal against

One goal against is roughly equivalent to losing 1.7 saves of value. The negative weight is high enough to penalize bad nights but not so high that a single fluky goal sinks an otherwise-solid performance.

How it compares to DraftKings

DK's standard goalie scoring is 0.7 per save, +3 for a win, +6 for a shutout, −1 per goal against. Our formula skips the win and shutout bonuses for two reasons:

• We're not predicting DK points. We're scoring per-game performance for prop analysis, not fantasy contests.
• Win/shutout aren't reliably in the data. Per-game stat objects don't always include final result fields, and we want a formula that works on every game without exceptions.

If you're comparing our score to a DK-style projection, expect ours to be a few points lower on average. The shape of the distribution is similar, just shifted.

How it compares to FanDuel

FanDuel uses 0.2 per save, +6 for a win, +9 for a shutout, no penalty for goals against. That's a fundamentally different model — it rewards winning more than performance. Our formula is closer to a "true talent" measure than to either fantasy site's scoring.

Where the formula falls short

Honest disclosure: this isn't a perfect goalie evaluation tool. Two notable gaps:

It ignores shot quality.

A save off a slot wrister and a save off a bad-angle wrap-around count the same. Top analytics sites use Expected Goals (xG) to weight shots by probability of scoring. We don't yet — the data isn't reliably available in our current pipeline. It's on the roadmap.

It penalizes goalies behind bad teams.

A goalie who faces 40 shots a night and lets in 4 looks "worse" by our formula than one who faces 22 and lets in 2, even if the first goalie was actually playing better. The save-volume credit partially offsets this, but not entirely. For prop bets specifically, this is a feature more than a bug — you want to favor goalies in shot-volume situations on save props anyway.

Practical use

The Goalie Fantasy Score works best as a relative measure across recent games for the same goalie. Is this goalie trending up or down? Are they performing above or below their personal baseline? Those questions are what the score answers.

For absolute comparisons across goalies on a given night, combine the score with the volume context: which goalie is facing more shots tonight? Which is at home? Which has more rest? The Lyzos Score on a goalie prop combines all of that.

Try a goalie prop in the analyzer to see the Fantasy Score in action. Open Lyzos →