How to Fish player count: the daily peak is half the record
Contents
Five days after its record, the concurrent player peak for How to Fish is roughly half what it was. That is a measured figure, not an estimate — but it does not follow that the game is emptying out: on the same day, different channels returned numbers 70% apart.
Five readings in one day
Every number below was read on 31 August 2026, at different moments.
| Channel | Moment | Players |
|---|---|---|
| Steam Web API (official) | morning reading | 163,914 |
| Steam Web API (official) | afternoon reading | 101,746 |
| Steam Web API (official) | 20:48 Kyiv time | 96,569 |
| SteamCharts | playing, 20:49 Kyiv time | 99,719 |
| Steam community hub | In-Game | 103,040 |
The lowest and highest readings differ by 67,345 players — 69.7% of the smaller one. The intra-day amplitude is therefore about 1.7×.
More telling still: the official Steam Web API returned 96,569 at 20:48, and SteamCharts returned 99,719 a minute later. That is 3,150 players (3.3%) between two channels that both claim to show what is happening right now.
What actually fell
Only like-for-like figures can be compared. Peak against peak can be.
| Metric | Value | Source |
|---|---|---|
| Players right now | 96,569 | Steam Web API |
| Peak in the last 24 hours | 189,311 | SteamCharts |
| Average over the last 30 days | 150,921 | SteamCharts |
| All-time peak | 373,887 | SteamCharts |
| 373,887. The record was set on 26 August, so that is a **49.4% drop in | ||
| five days**. Both figures measure the same thing, the maximum of simultaneous | ||
| players, which is what makes the comparison valid. |
But the current 96,569 against the 24-hour peak of 189,311 works out to 51.0% — almost the same gap. In other words, the overnight trough inside a single day is as deep as the fall from the record over five days. That is exactly why a single player-count reading proves nothing: it is explained equally well by decline and by a late hour.
An independent signal: the review rate
Player count is not the only thing that can be counted. Steam’s review histogram gives daily counts of new reviews, which is a proxy for new buyers.
Daily reviews peaked on 23 August at 4,535. By 30 August the figure was 3,237, a 28.6% fall over seven days.
Two independent sources show a decline, but of different depth: the player peak is down 49.4% in five days, the flow of new reviews down 28.6% in seven. That is not a contradiction — they measure different things. Concurrent players depend on whether existing owners come back; the review flow depends on how many people are buying for the first time.
Context: the game is 11 days old
How to Fish was released on 20 August 2026, so it is eleven days old at the time of these readings. The SteamCharts average for the last 30 days — 150,921 — formally covers a period longer than the game has existed.
What we are not claiming
We did not sample an hourly curve. We have three official readings across one day, not a graph, so we do not know the real daily minimum or when it occurs.
We are not claiming the fall will continue. Five days after a launch peak is too little to separate a normal post-launch settling from a decline. We record the state on 31 August and do not extrapolate.
We did not establish how SteamCharts computes a 30-day average for an 11-day-old game — whether it divides by 30 or by 11. So 150,921 should be read as a SteamCharts figure, not as an average we verified ourselves.
We do not equate concurrent players with owners. A fall in concurrency does not mean refunds; that is a separate quantity and has to be measured separately.
Finally, the record itself is unresolved: the game’s news feed carried 373,971, while SteamCharts on 31 August shows 373,887. The difference is 84 players. We do not know which is more accurate, and we are not quietly picking one.
Frequently asked questions
How many people are playing How to Fish right now?
Our latest reading is 96,569 concurrent players at 20:48 Kyiv time on 31 August 2026, from the official Steam Web API.
Is the player count falling?
The 24-hour peak is 189,311 against the record of 373,887 set on 26 August, a 49.4% drop in five days. But the intra-day swing alone is close to 2×, so a single reading proves nothing.
Why do different sites show different numbers?
Because they are readings taken at different moments. The official API returned 96,569 at 20:48 and SteamCharts 99,719 a minute later — 3,150 players apart in adjacent minutes.
What is the all-time player record?
SteamCharts shows 373,887 as of 31 August. The game's news feed carried 373,971 — an unresolved difference of 84 players.
Does a falling player count mean people are refunding the game?
No. Concurrency and refunds are different quantities. The flow of new reviews over the same period fell by 28.6%, less than the drop in the player peak.