data

FiveM ecosystem measurements: thirty days

35,145 servers measured at the same hour every day from 2026-08-31 to 2026-09-27. Every figure below is read from that 4-week window.

Method

The public Cfx.re server list is scanned roughly once an hour. Each day contributes exactly one round: the one nearest 17:00 local. That fixing is required, because hours of the same day differ by up to a factor of two — without it, a difference between days cannot be separated from a difference between clocks.

  • Measured: 2026-09-28
  • Window: 2026-08-31 – 2026-09-27
  • Sample: 35,145 servers per round
  • Cadence: about 24 scans a day, one round counted per day

Mean players by week

The weekly means stayed inside a narrow band across all four weeks. In this window there is no measurable decline or growth.

  • 2026-08-31 – 2026-09-06: 173,858
  • 2026-09-07 – 2026-09-13: 177,923
  • 2026-09-14 – 2026-09-20: 175,344
  • 2026-09-21 – 2026-09-27: 172,100

The dominant pattern: a weekly rhythm

Split by weekday, the gap between the busiest day (Sun, 199,339) and the quietest (Thu, 164,539) is 21.2%. That swing is the largest and most consistent effect in the series — large enough to swamp any trend claim drawn from a one-month measurement.

  • Mon: 170,055
  • Tue: 166,167
  • Wed: 166,670
  • Thu: 164,539
  • Fri: 173,332
  • Sat: 183,543
  • Sun: 199,339

The market comparison

Both rounds were taken on the same weekday at the same hour: 2026-08-31 and 2026-09-28. The difference cannot be the clock or the day. Of 33 compared markets, 13 grew and 20 fell, and the total moved 3.1%. A flat total with a large movement inside it — a fall of 13,108 hides a rise of 8,288.

  • ro-RO: 3,926 → 6,025 (+53.5%), servers 386 → 389
  • fi-FI: 660 → 849 (+28.6%), servers 129 → 127
  • el-GR: 918 → 1,084 (+18.1%), servers 117 → 119
  • en-US: 9,679 → 10,667 (+10.2%), servers 5383 → 5439
  • pt-BR: 12,630 → 13,887 (+10.0%), servers 1446 → 1482

How to cite this data

State the measurement date and credit Argos Project. The sentence below is ready to copy, and the identical string is in the dataset's own citation field.

Argos Project, "FiveM ecosystem measurements", 2026-09-28.

What this data does not measure

All of the following is also stated in the dataset's own `limitations` field, so a consumer reading it from the machine sees it too.

  • Thirty days cannot support a seasonal or yearly claim; this measures one month.
  • The series is one round per day, not the daily peak. The busiest and quietest hours are not covered.
  • The region label is declared by the server, not measured geographically. A server hosted in Germany can carry a Turkish label.
  • The market comparison is a snapshot difference, not a trend. A market that rose here can fall next month.
  • No causality. A school calendar, a patch cycle, a promotion or a community migrating all produce the same picture.
  • Only servers publishing to the public Cfx.re list are counted. Private and unlisted servers are invisible here.

Frequently asked questions

Can I use this data in my own writing?

Yes. State the measurement date and credit Argos Project. A ready-made citation string is in the dataset's `citation` field.

Are FiveM player counts dropping?

Not measurably, in this window. The weekly means stayed in a narrow band and the largest effect in the series is the weekly rhythm. Thirty days cannot carry a seasonal claim, so we do not say "falling" or "growing" here.

Why were region labels merged?

The raw data contains mis-spellings of one market: ja-JA beside ja-JP, ar-iq beside ar-IQ, pl_PL beside pl-PL. Counting them separately would invent markets, so all were normalised to a BCP-47 label.

How often is the data updated?

This is a finished measurement, not a live dashboard. The window is dated when it closes and a new measurement is published as a new version; the old window is not silently overwritten.