Somewhere in the last decade, hours became a specification. Marketing quotes them, reviews mention them, and players compare them as though the number described value.
Completion statistics suggest otherwise. Most players do not finish most games, and the drop-off is steepest in exactly the titles that advertise the largest hour counts.
Where the number comes from
A reasonable place, originally. A game costing sixty units of currency and lasting four hours felt like poor value next to one lasting forty, and publishers responded to that comparison.
The problem is that the cheapest way to add hours is repetition. Collectibles, fetch quests, an enlarged map with the same activities spread further apart. That inflates the number without adding anything.
What players actually do
Abandon games in the middle, in enormous numbers, and not because the games are bad. They stop because the game stopped introducing anything and they noticed.
The pattern is consistent: engagement holds while new mechanics, locations, or ideas keep arriving, and falls off sharply once the game has shown everything it has and is simply asking for more of the same.
The shift underway
Shorter, denser games have found both critical and commercial success, and the six-to-ten hour range has stopped being treated as insufficient.
Subscription services accelerated this. When trying a game costs nothing extra, the first hour has to earn the second, and a slow opening that a purchaser would push through because they paid for it simply results in an uninstall.
What is worth measuring instead
Whether the last hour is as good as the first. A game that maintains quality for eight hours is better than one that maintains it for twelve and then continues for thirty.
Completion rate is the honest metric, and it is the one publishers have and do not publish.