Steam charts
Live player counts for every game we track, ranked by how many people are in each one on Steam right now — who is trending, today’s peaks, hours played, and the servers behind each game. Counts come from Valve’s own endpoints, sampled every ten minutes and kept as a history you can look back through.
Where these Steam player counts come from
Concurrent players, as Steam counts them: everybody with the game open, wherever they are playing it. Valve publishes a live figure per game and an official top 100; we read both every ten minutes and keep the samples, so the history here starts when we started recording rather than going back to release day.
Peaks are Steam’s own 24-hour figures rather than the highest sample we happened to catch, which is why a peak can be higher than anything on the graph.
Hours played has no source at Valve — their charts publish a rank, a count and a peak and no playtime at all. It is our own samples added up: each reading stands for the ten minutes until the next one, so a day of them is player-hours. Hours observed, in other words, which is what every chart site of this kind is counting.
The window is one day here, and that is why the figure is smaller than the one beside the same game on sites that total thirty. Same arithmetic, thirty times the span — a game averaging 840,000 players is twenty million player-hours a day and six hundred million a month.
Players in a game against players on a server
These are not the same measurement and neither is a subset of the other. A game’s player count includes single-player and matchmaking; a server’s is a machine somebody runs, which we query directly every few minutes.
Read together they say something useful: a game high on this chart with few servers listed is one where community hosting has room, and a game with thousands of servers and a modest count is one where the players are spread thin. The catalog is the other half of that picture.
