A Save Predicts a Stream
Streamers went live with the games they saved at 26x the rate of the games they skipped. We tried nine ways to break the result.
StreamGist Blog · Topic
Methodology
How GistScore, Save Rate, and Stream Fit are computed, validated, and used.
4 posts
← Back to all postsAbout Methodology
Methodology posts explain the math, the inputs, and the validation behind the signals the rest of the blog cites. GistScore, Save Rate, Stream Fit, viewer concentration, and the risk and outlook tags all have specific definitions, and each one's documented here.
The intent's straightforward. If you're reading a Weekly Pulse or Game Spotlight piece and want to verify how a number was derived, a Methodology post explains what went in and what got excluded. Posts here also document the tradeoffs we made and the things the signals don't measure. GistScore doesn't predict ad revenue. Save Rate doesn't measure long-term retention. Read these when you want to check our reasoning.
4 posts
Posts in Methodology
New Site Is Live. More Data Coming to the Blog
The new StreamGist site is live, and the blog is leaning harder on the daily data behind hundreds of Twitch categories.
Light Up The Town Scores a 96 GistScore — The Best Discovery Opportunity This Week
Light Up The Town scores a 96 GistScore — the best discovery opportunity this week. Plus Resident Evil 2, RV There Yet?, and Trackmania data.
Stream Fit Rejection: A New Standard for Understanding Streamer Choice
A concise introduction to Stream Fit Rejection that defines skips, saves, and neutrals and explains how tracking rejection data transforms streaming analytics, sharpens the read on creator intent, and aligns game recommendations with true streamer fit.