How to log felt volatility from a demo session

A log is observation practice, not prediction

An earlier article covered why volatility grades sit on studio-specific rulers. What a reader can do about that is exactly one thing — put the label and the actual on-screen impression side by side. Session logs are the material for that comparison.

Be clear about the purpose first. A session log says nothing about the next session: one session is one sample, and no sample describes the next result. What a log trains is observation, not prediction — the ability to put into words how long the quiet stretches felt and how rarely the feature arrived.

Three items are enough

First, felt volatility — a value you assign yourself, 1 to 5. This is your ruler, not the studio's. Rating a game 4 when the studio label says medium is not a contradiction; it is the discovery of a gap between label and impression.

Second, feature-entry frequency. A rough rare / normal / frequent is enough. There is no point counting exactly: sessions of different lengths make counts incomparable anyway, and the goal is to keep the direction of the impression.

Third, session length. A ten-minute impression and an hour-long one carry different weight, and only a recorded length lets you judge later how much to trust the note.

What you can skip is money and outcomes. A demo runs on a play balance, so amounts mean nothing, and a win or loss is the result of a single sample — it adds nothing to observation practice.

Reading the log back

Once notes accumulate, two comparisons open up. One is against the label — placing your felt value next to the studio's published grade or hit rate. Where they diverge, the session was usually short, or the game's payout distribution simply does not compress into one grade word.

The other is between notes. When several people leave several notes on the same game, a tendency in the screen impression becomes carefully discussable — something no single sample can support. That is exactly why the session-report form in this site's community takes these three items — felt volatility, feature entry, demo time — as its choices.

The line to hold is the same in either comparison: a log can say "this is what happened", never "this is what will happen".