The first score that makes game masters better on its own.
Every owner knows who their good GMs are. None of them can prove it, show it, or scale it — so coaching is vibes, and quality lives and dies with whoever's on shift. GMQS turns real gameplay data — escape rate, time remaining, and the quality of every clue intervention — into one score, plotted on a matrix every guide can see.
No more hard conversations.
The score was born so managers could coach without write-ups and without “you're on your phone too much” arguments. Now the manager just points at the board: “Why do you think your dot is over here in the Ghost while John and Jessica's are up near the Zone?” The data does the talking. The conversation isn't about the phone anymore — it's about the score. And because every guide sees the board, improvement becomes self-directed and peer-driven before a manager says a word.
Where does your team land?
Two basic numbers you already track. The same scoring engine that runs two real venues plots your operation on the industry's leading standard for game-master performance.
Loss depth: two identical losses, opposite problems
A spreadsheet counts a loss as a loss. But a team that got 95% of the way through and a team that quit at minute 30 with time still on the clock are not the same failure — and coaching them the same way is how you lose a good game master.
● Nearly finished
The GM cared and missed the timing. That's a Heartbreaker — close, engaged, one strategic nudge from a different outcome. Fix the timing, not the person.
● Stalled early
The team was gone long before the clock was. That's a Ghost — presence or group-sizing, not craft. A different conversation entirely.
So NailBiter measures how far teams actually get when they lose, and colors it into the dot you're already looking at — the ring stays the guide's tier, the core becomes loss depth. Same chart. Same mental model. A whole diagnostic thread that was invisible before.
Loss depth is deliberately diagnostic, not scored. So is clue quality at the individual level. The governing rule: score only what a guide cannot fake, and only what measures outcome rather than method. Fold a diagnostic into the score and you corrupt the very data that made it useful.
Let the data do the confronting
This is the part owners underestimate. Young managers find confrontation hard, and staff get defensive under judgment. The score removes the judgment entirely.
“I saw you on your phone.”
A personal accusation.
“Your teams quit with 12 minutes left three times this week — walk me through those games.”
A question about games.
Every problem this system cares about has a data signature. Once the signature is on the wall, nobody has to catch anybody — the manager's job shifts from policing behavior to helping someone move their dot.
And it runs in both directions: the score defends a present game master as loudly as it exposes an absent one.
That's what turns it from surveillance into a scoreboard — and it's why staff stop resisting it.
The conversation, in five steps
- Show the chart — the zone, never the character.
- Ask about specific games: “walk me through these three.”
- Run the diagnostics together — room mix, group size, concurrency.
- Agree on one move: the zone's single coaching line, not a lecture.
- Next week, look at the arrow together. Coach the trend, not the snapshot.
Or put another way — coach like you game master. Size them up before you speak. Strike early, on the first drifting game rather than the month-old pattern. Ask, don't tell, so the fix feels like theirs. And show up loaded: diagnostics already run, one specific move ready.
The Clue Ladder
Every clue is typed on an intervention ladder: question → affirm → nudge → walkthrough → answer. The doctrine is built into the tooling: ask, don't tell; strike early; a small nudge now beats a rescue at the wire.
Send a clue. See what it costs.
12:00 elapsed. The team has been silent at the map-and-compass puzzle for 3 minutes. The youngest player is holding the compass upside down. What do you send?