How the How to Fish Steam score works: 94.9% and the label

31 August 2026 · 4 min read

GameMapa editorial · how we check facts
Checked: Claude (AI) — fact-checked against the Ukrainian original; not proofread by a native speaker

How to Fish carries the label Very Positive on Steam, next to a number — 94.9% positive out of 43,902 reviews. Where does each of those come from?

The number itself is simple

It is a plain ratio: 41,661 ÷ 43,902. No weighting by playtime, by date, or by how many people found a review helpful goes into that figure.

The label, though, is a threshold — and Valve does not publish it

Steam separately returns a numeric score code. For How to Fish it is 8. But there is no public table of “which percentage gets which label”. So we did something else: we measured the labels of 15 games and looked at where they switch.

What the table shows for certain:

• The switch from Very Positive to Overwhelmingly Positive sits between 94.90% and 95.57%. How to Fish (94.90) is not there yet; Far Far West (95.57) is.
• Review count alone does not decide the label: Counter-Strike 2 with 9.8 million reviews carries the same label as Funnel Runners with 3,286.

What the table does not show: where the Mixed / Very Positive boundary sits. In our sample the highest Mixed is 57.84% and the lowest Very Positive is 82.18%. That is a 24-point hole, and we measured no game inside it. So we cannot name that threshold, and we will not.

How to Fish is 0.7 points from the next tier

The practical read: 94.90% is right at the edge. The ceiling of its label is somewhere around 95.5%, so it is less than one percentage point short of the next step up.

The trend is working in the game’s favour. In the week of 20-26 August the positive share was 94.0%; in the week starting 27 August it was 95.3%. Fresh reviews are already better than the game’s own average, and the average is slowly climbing.

Two discrepancies we are not hiding

First. The per-language review counts sum to 42,871 while the total counter read 43,770 (a slightly earlier snapshot than 43,902). The gap is 899, or 2.1%.

Second. The daily values in the review histogram sum to 39,095 while the total counter reads 43,902. The gap is 4,807, or 10.9%.

The obvious explanation — purchases made outside Steam — does not fit: there are 5,197 of those, not 4,807. We could not explain this gap, and we are not inventing an explanation.

What we are not claiming

We do not know Valve’s official thresholds — we measured where they show up across 15 games, which is not the same thing. We also do not know whether anything besides the percentage feeds the label: there may be a minimum review count, but our sample did not reveal one.