We audited our own website — and deleted what was not true

Every game's website drifts. A feature is planned, described, then changed or postponed — and the description stays. Over months, a homepage can end up promising a game that does not quite exist. In September 2026 we decided to find out how far ours had drifted.

The method

We took every factual claim on the public site — the homepage, the guide, the wiki, the FAQ — and checked each one against the game itself. Not against our memory of the game, or our plans: against what actually happens when you play. Over seventy claims went through that check.

What was not true

Most claims held up. Several did not, and we want to be specific about them, because vague apologies are worth very little:

• The login streak. The site said missing a day would only slowly reduce your streak. In reality, missing a day resets it to one.
• Bank loans. One page described loans as interest-free. They are not: they cost 2% per period, or 3% above ten million, over 26 weeks — and the wiki said so correctly in another article.
• Stadium growth. The site described a ladder of stadium sizes that no club in the game had actually climbed, because stadium upgrades were not working at the time.
• Formations. The site mentioned eight formations. The tactics screen offers five.
• AI clubs trading among themselves. The site described a busy market between AI clubs that was not happening in practice.
• Advertising. The site said it was supported by advertising, at a time when it showed none.

Testing the match engine's promises

Some claims were about tactics — for example, that a high defensive line, intense pressing or a faster tempo produced specific effects. Those are harder to check by reading, so we tested them by playing: 15,000 simulated matches in controlled pairs, changing one setting at a time. Three mechanisms described on the homepage produced no measurable effect at all, or even the opposite of what was claimed. We removed them from the site, and we tell the full story of that experiment in 15,000 matches to answer one question.

What we did about it

Every untrue claim was either corrected or removed. Where a feature was being rebuilt — stadium and facility upgrades, for example — the site now says exactly that, without numbers or promises. A wiki article about set pieces, which described options the game did not offer, was rewritten from scratch.

We also added an automatic check to our publishing routine, so that a specific kind of promise we had to remove cannot creep back onto the homepage unnoticed.

The rule

The rule we follow now is short:

If we have not measured it or seen it in the game, we do not say it.

It is the same rule this blog follows. When a post describes a number, it is a number we measured. When something is planned, we say it is planned. And when we get something wrong — it will happen again — we correct the post and note the change.

Why publish this at all?

Because trust is the only thing a small independent game really has. You cannot see inside our code, so you have to take our word for what the game does. We would rather tell you about the things we got wrong than have you find them yourself.

Next in the diary: 15,000 matches to answer one question — do tactics matter?