Since June 28 we've been banking a daily snapshot of the live player count for every game our network tracks, Adopt Me included. Sixteen days in, I can make Adopt Me look like it's collapsing (392,110 on June 28, 284,066 today — down 28 percent) or look fine (347,799 on July 1, 372,268 on July 12 — up 7 percent) without touching a single number. Both are honest endpoints from the same ledger. When our first look at this dataset reported Adopt Me down 18 percent, that was a third pair of endpoints from the same series. Endpoint arithmetic on a noisy series tells you about the endpoints, not the game.
So here's the series itself, all ten readings: 392k, 363k, 187k, 348k, then a gap, 386k, 358k, 323k, 361k, another gap, 372k, and today's 284k. The median reading is about 359,000 concurrent players. Ignore the 187k for now — I'll come back to it, because it's the most instructive number in the set.
The steadier metric: share
Raw concurrents swing with the hour and the day of the week, and our snapshot fires when the daily build runs — anywhere from noon to 10pm UTC across this window. That wobble poisons any two-point comparison. But it poisons every game equally, because the whole platform breathes together: kids get home from school everywhere at once. So the steadier question is: what fraction of all the players across our ~104 tracked games are inside Adopt Me at that moment?
Answer: a remarkably tight band. Outside the first day (11.5 percent) and today (7.5 percent), Adopt Me held between 8.8 and 10 percent of our tracked network on every single reading. Roughly one in ten concurrent players across a hundred of Roblox's biggest games is standing in Adopt Me at any given moment. Games that are dying do not do that. They drift down a share chart week after week, and Adopt Me's share line is mostly flat.
The warning sign that survives the caveats
Here's the part I'd take seriously if I were holding a large pet inventory. Over this same window, the sixteen games in our 50k+ cohort posted a median gain of about 16 percent — school's out, and the summer tide is lifting nearly everything. Murder Mystery 2 is up 49 percent. Pet Simulator 99, which bolted a World Cup event onto its title, is up 84 percent. Dress To Impress is up 73 percent. Brookhaven, already the biggest thing we track, added another 22 percent. Blox Fruits, 15.
Adopt Me is one of exactly three games in that cohort that went the other way. And in the standings that ordinary players actually see, it slipped from the second-biggest game we track to the fourth, passed by Murder Mystery 2 and 99 Nights in the Forest. You can watch that table move day to day on our sister site's trending board, which runs on this same ledger.
That's the honest headline. Not "Adopt Me is dying" — "Adopt Me is standing still in a season when its neighborhood is sprinting." A flat share of a growing pie is fine. Being outgrown by every peer during the easiest growth window of the year is the kind of thing that's invisible in any single day's number and unmistakable in a ledger.
Why I don't fully trust my own thermometer
Now, that 187k reading from June 30. The network total that day was normal — every other game read ordinary numbers — so it wasn't a platform outage or a bad crawl. Either Adopt Me had a genuinely terrible few hours, or Roblox's API served us one weird figure for one game. I can't distinguish those, and that's the point: any single snapshot, including today's low 284k (a Monday evening, against a Sunday-evening 386k eight days earlier), deserves the same suspicion. Ten readings with two gaps, at a floating hour, is a short and imperfect window. The frozen RAP feed taught us to interrogate our own instruments before we accuse a game of dying, and I intend to keep that habit.
Dying versus smaller
Whatever Adopt Me has lost from its 2020-era peak — and by every account it was once far larger, though we weren't measuring then, so I won't put a number on it — the absolute figures today are absurd by any standard except its own history. Our window caught it gaining about 141 million visits in fifteen days, roughly 9.4 million a day, on the way to 44.1 billion lifetime. The votes players cast during our window ran 96.5 percent positive — 32,244 new likes against 1,186 new dislikes. A third of a million people are typically inside it whenever we look. The codes page still exists because people still search for it, which is its own lagging indicator of life.
I wrote earlier this month about whether Adopt Me's trading economy is dead, and the answer there was "the top is healthy, the middle is thinning." The population data now gives that a second, independent leg: the player base isn't draining, but it isn't growing while everything around it grows, and market softness in the mid-tier is exactly what you'd expect from a stable population that mints fewer new traders each month than its rivals do.
So: is Adopt Me dying? Measured over our window — no. It's enormous, stable in share, and beloved by the people in it. But "not dying" and "winning" have never been further apart, and the ledger will keep score either way. Ask me again at a hundred days.