Roblox Games Gaining the Most Players
Ranked by the share of its own player base each game added over the last 6.7 hours, among the 838 games holding at least 500 players at both ends.
A full day of history does not exist yet, so this chart measures the longest window the readings genuinely support and names it by the true median age of the baselines it used — 6.7 hours — rather than rounding it to a day. The window widens on its own as readings accumulate and settles at 24 hours once collection has been running that long. Short windows are noisier and far more exposed to the daily cycle, which is why the “vs tide” column matters more here than it will later.
- 1
15,631 → 73,139+367% vs tide+368%
- 2
1,175 → 3,813+224% vs tide+225%
- 3
5,005 → 16,172+222% vs tide+223%
- 4
1,104 → 3,092+179% vs tide+180%
- 5
1,339 → 3,536+163% vs tide+164%
- 6
2,291 → 5,943+158% vs tide+159%
- 7
661 → 1,474+122% vs tide+123%
- 8
789 → 1,720+117% vs tide+118%
- 9
15,048 → 31,989+112% vs tide+113%
- 10
864 → 1,814+109% vs tide+110%
- 11
1,711 → 3,579+108% vs tide+109%
- 12
713 → 1,489+108% vs tide+109%
- 13
732 → 1,489+102% vs tide+103%
- 14
2,251 → 4,143+83.1% vs tide+84.1%
- 15
915 → 1,678+82.4% vs tide+83.4%
- 16
1,152 → 2,077+79.3% vs tide+80.3%
- 17
2,324 → 4,096+75.3% vs tide+76.2%
- 18
23,610 → 41,521+74.9% vs tide+75.9%
- 19
7,142 → 12,426+73.0% vs tide+74.0%
- 20
3,229 → 5,566+71.4% vs tide+72.4%
- 21
4,543 → 7,793+70.6% vs tide+71.5%
- 22
1,283 → 2,193+70.0% vs tide+70.9%
- 23
42,073 → 71,277+68.4% vs tide+69.4%
- 24
1,843 → 3,109+67.7% vs tide+68.7%
- 25
15,886 → 26,737+67.3% vs tide+68.3%
- 26
924 → 1,543+66.0% vs tide+67.0%
- 27
5,655 → 9,385+65.0% vs tide+66.0%
- 28
2,157 → 3,577+64.9% vs tide+65.8%
- 29
3,117 → 5,114+63.1% vs tide+64.1%
- 30
1,693 → 2,767+62.5% vs tide+63.4%
- 31
1,267 → 2,062+61.8% vs tide+62.7%
- 32
40,910 → 66,535+61.7% vs tide+62.6%
- 33
3,094 → 4,949+59.0% vs tide+60.0%
- 34
6,647 → 10,623+58.8% vs tide+59.8%
- 35
112,868 → 180,276+58.8% vs tide+59.7%
- 36
8,422 → 13,332+57.3% vs tide+58.3%
- 37
573 → 901+56.3% vs tide+57.2%
- 38
7,239 → 11,375+56.2% vs tide+57.1%
- 39
539 → 840+54.9% vs tide+55.8%
- 40
1,176 → 1,832+54.8% vs tide+55.8%
- 41
577 → 898+54.7% vs tide+55.6%
- 42
2,187 → 3,397+54.4% vs tide+55.3%
- 43
675 → 1,029+51.5% vs tide+52.4%
- 44
1,363 → 2,075+51.3% vs tide+52.2%
- 45
4,932 → 7,502+51.1% vs tide+52.1%
- 46
1,092 → 1,657+50.8% vs tide+51.7%
- 47
785 → 1,190+50.6% vs tide+51.6%
- 48
1,715 → 2,591+50.1% vs tide+51.1%
- 49
2,292 → 3,458+49.9% vs tide+50.9%
- 50
5,761 → 8,661+49.4% vs tide+50.3%
Hourly player-count collection began , which is 10.3 hours ago across 5 completed runs. We have no readings from before that date and do not estimate any, so nothing here describes a period we did not measure. The window this chart uses widens automatically toward 24 hours as history builds.
Why the tide column exists
Roblox does not have a player base so much as a wave. Concurrency across the whole platform roughly doubles between its overnight trough and its evening peak, and almost every individual game rides that wave together. Over a short window, the single biggest determinant of whether a game is “up” is not the game at all — it is what time you looked.
This is the failure mode of every naive growth chart, and it is worse the shorter the window. Measure from 12:00 to 16:00 UTC and virtually every game on Roblox is growing; measure from 22:00 to 02:00 and virtually every game is dying. A chart built on those raw numbers is a very expensive clock.
So each row shows two things. The badge on the right is the raw percentage change, which is a true fact about that game. The “vs tide” figure beside it is that change minus what the median qualifying game did over exactly the same window, which is the part specific to this game. When the tide is running hard, the two numbers diverge a lot, and the second one is the honest one.
We rank on the raw percentage rather than on the tide-adjusted figure, deliberately. The raw number is a simple measured fact and the ordering it produces is reproducible from the two readings printed on the row. The adjusted figure depends on a median across a set that changes composition as games cross the eligibility floor, which is a more useful number but a less verifiable one. Showing both, and ranking on the one you can check, seemed the right way round.
Growth that is real, and growth that is not
There are four common reasons a game appears on this chart, and only one of them is the interesting one.
A genuine update. A developer ships new content, existing players return and tell people, and the count steps up and stays up. This is what the chart is for, and it is the hardest to confirm without watching for several days.
An event. A limited-time mode, a collaboration, a concert. Real players, real spike, and a decay curve that starts the moment it ends. On a 24-hour window these look identical to a genuine update.
Attention from elsewhere. A large video or a trend pushes a wave of new players at a game that has not changed. Often the game cannot hold them, and the same game appears on the drops chart a few days later.
A measurement artefact. A missed collection run, or a Roblox-side problem that suppressed the baseline reading. Recovery from a suppressed reading is the most misleading pattern on any chart of this kind, because it looks exactly like explosive growth and is entirely an illusion. We record failed readings separately rather than storing them as zeroes, which prevents the worst version of this, but a count that Roblox reported as genuinely low during a problem on its side is indistinguishable from a real one.
Telling these apart is a matter of history, not cleverness. A game that is still here in a week was an update; one that is not was an event. That comparison becomes possible on this site as the readings accumulate, and the trending score is where it will surface, because its acceleration term is precisely the test of whether a rise is continuing or fading.
Questions about these charts
How far the game beat the median qualifying game over exactly the same window, in percentage points. Roblox concurrency rises and falls with the clock — as the American afternoon arrives, almost everything goes up together — so a raw percentage over a few hours largely measures the time of day. The tide figure strips that out: a game up 14% on a day when the median game is up 9% has really gained about 5 points of its own, and that is the number worth reading.
Because percentage growth on small numbers is noise wearing a big hat. A game going from 2 players to 20 is up 900%, and would sit permanently at the top of this chart while telling you nothing at all. There is a statistical floor too: concurrent player counts jitter by roughly the square root of the count, so a 20-player game varies by around 22% on its own with nothing happening, and a 50-player game by 14%. At 500 that jitter is down around 4.5%, which is low enough that a double-digit move is a real event. We require the floor at both ends of the window, which also stops a collapsed game re-entering the chart as a "gainer" while it bounces off the bottom.
Sometimes, and this chart is honest about not knowing which. Rapid growth means people are arriving, which usually means an update landed, a video did well, or an event started. It does not distinguish a genuinely improving game from one riding a spike that will decay within the week. The way to tell is to come back in a few days and see whether the game is still here — which is precisely what a longer history will let this page do automatically, and cannot do yet.
Because ranking by absolute players gained would produce the same five enormous games every single day. The biggest games on Roblox routinely swing by tens of thousands of players over a day purely on the daily cycle, which dwarfs anything a mid-sized game can do even when it triples. Percentage answers the more interesting question — what share of its own audience did this game just add — and the size floor stops that being gamed by tiny games. Both figures are on every row, so you can read either.
Yes, and there are three ways we know about and cannot yet exclude. A missed collection run makes the next reading look like a sudden step change rather than a gradual move. A Roblox-side outage that suppresses counts makes the recovery afterwards read as explosive growth. And a scheduled event — a two-hour in-game concert, a limited-time mode — produces a genuine but temporary spike that this chart cannot tell apart from durable growth. All three become detectable with weeks of history behind each game; none of them are detectable today.
Readings are taken once an hour and the page rebuilds every thirty minutes. Each row shows the two actual readings it compared, and the window named in the heading is the real distance between them, never a rounded-up label.
More Roblox charts
All Roblox charts → · Most played right now → · Trending games → · Biggest player drops →
Best Roblox games by genre → · Live Roblox player stats → · The full game database → · Working game codes →