Essay
Why I Won't Round a 3 Up to a 4
July 31, 2026 · by Elliot Rhodes
Somewhere along the way, star ratings on the internet quietly stopped meaning what the numbers actually say. A 3 out of 5 should mean average, genuinely fine, worth knowing about but not a priority, and instead it’s become shorthand for something closer to disappointing or even bad, because everyone rounds up out of some unspoken politeness contract nobody actually signed. I don’t do that here, and I get pushback on it more than any other single thing about this site.
I think the habit comes from the same place most soft feedback comes from anywhere, nobody wants to be the person who tells a friend their favorite thing is just okay. I understand the impulse. I also spend my working life doing the opposite of that impulse professionally, walking through someone’s biggest purchase and writing down exactly what’s wrong with it in language that doesn’t protect anyone’s feelings, including the seller’s, including sometimes the buyer’s own excitement about a house I know has real problems. That job doesn’t work if I round a cracked foundation up to a hairline settling crack because the buyer already picked out paint colors.
Games and books get the same treatment here for the same reason. A genuinely average game that I’d play again but wouldn’t go out of my way to recommend gets a 3, not a 4 padded up because I don’t want to seem harsh about something a lot of people clearly love. A book that lost me in the back third gets marked down even if the front two thirds were genuinely great, because the whole experience is what I’m rating, not the best moment in it.
The honest cost of doing it this way is that my average rating across this whole site probably reads lower than most review blogs, and I’ve had people assume that means I’m just a tough grader or don’t enjoy things as much as other reviewers. I don’t think that’s true. I think it means most other numbers floating around online have quietly drifted upward until a 3 stopped meaning average and started meaning bad, and a 4 stopped meaning good and started meaning fine. Somebody has to keep the scale honest or the whole thing stops being useful information and starts being a popularity contest with extra steps.
Dale thinks I’m needlessly stingy about this. Maren thinks it’s the most consistent thing about my personality, for better or worse. I think a rating that always goes up when you’re not sure isn’t really a rating at all, it’s just enthusiasm wearing a number, and I’d rather my numbers actually mean something when someone’s deciding whether to spend their own money and time.