A new five-star gets revealed on Friday. By Saturday there are three tier lists ranking her, two of them disagree, and a forum thread forty replies deep is already arguing about whether she is a must-pull or a trap. Nobody in that thread has the character. The banner is not even live yet.
We run tier lists on this site. We think they are useful. So take this as a confession as much as a critique: the version of a tier list that goes up the moment a character is announced is, almost by construction, wrong. And people read it more intensely than any tier list that comes later, when it might actually be right.
Why the first ranking is structurally broken
It is not that the people writing day-one rankings are careless. It is that the information they need does not exist yet.
The kit is theory, not practice. A character's abilities on paper and a character's abilities in your hands are different animals. Numbers that look terrifying in a skill description turn out to need a specific team, a specific rotation, or an investment level most players will not reach for months. Kits that read as mediocre turn out to bend the game around a mechanic nobody clocked in the reveal.
The math is running on estimates. Day-one damage calculations use leaked or best-guess multipliers, assumed buffs, and idealized rotations against a training dummy. Real content has add waves, downtime, positioning, and enemies that do not stand still. A theoretical ceiling is not a floor, and most players live near the floor.
The meta it is ranked against is frozen. A character is only strong or weak relative to a moment. The units they compete with, the endgame that is live, the supports that exist: all of it shifts. Today's "overrated" pull is next patch's centerpiece the instant the right teammate arrives. The ranking was not wrong so much as early.
Reveal hype is a distortion field. The loudest takes travel fastest, and "must-pull" and "skip" both outperform "it depends" in a headline. The first ranking you see is filtered for confidence, not accuracy.
So why does anyone read it?
Because the day-one tier list is not really answering "how good is this character." It is answering a different question, one that is genuinely urgent: should I spend my resources right now?
That question has a real deadline. Banners expire. Pity is a finite, hard-won currency, and the fear of stranding it on the wrong unit is the most reliable emotion in the entire hobby. The day-one ranking is a decision aid under time pressure, and time pressure is exactly when people reach for a single letter grade instead of a paragraph of nuance. S-tier means pull, F-tier means save, and both feel like permission to stop agonizing.
There is a social function too. The tier list is a shared map. Even when it is wrong, it gives the community a common reference to argue against, which is half of what makes a new patch fun. "The tier list says S but I think she's overrated" is a better conversation than staring at a kit alone. The ranking is the campfire everyone gathers around, not the final word anyone actually believes.
How to read a day-one tier list without getting burned
You do not have to ignore early rankings. You have to read them for what they are.
- Treat placement as a hypothesis, not a verdict. A day-one S-tier means "this looks strong on paper." Wait a week and the same list, re-rated by people who now own the unit, is worth ten times as much.
- Read the reasoning, skip the letter. Any ranking that only gives you a grade is telling you what to think. A good one tells you why, and the "why" survives even when the exact placement does not. If a list will not explain itself, it is entertainment, not analysis.
- Separate "strong" from "for me." The best character in the game is a bad pull if you already have three units that do her job, and a niche unit is a great pull if she fixes the exact hole in your account. Tier lists rank power in a vacuum. Only you know your roster.
- Watch what the list does after release. The tell of a trustworthy source is not being right on day one. It is being willing to move a unit up or down once the evidence arrives, and saying so out loud. A ranking that never changes was never measuring anything.
- Give it the two-week test. If a character is genuinely account-defining, that will still be obvious in two weeks, and you will usually get a rerun. Very few pulls are actually now-or-never. The panic is louder than the stakes.
What we are actually promising
Here is the standard we try to hold ourselves to, and the one worth demanding from anyone ranking characters for you. A tier list should tell you when it was last updated, what patch and content it is rated against, and how confident it is. It should change when the game changes. And it should be honest that a placement made before anyone has played a character is a forecast, not a fact.
That is why the rankings on this site carry a version and a refresh date, and why we would rather move a unit than pretend we called it perfectly the first time. The ZZZ, Honkai: Star Rail, and Genshin tier lists all get re-rated as a patch settles, not frozen at reveal.
The day-one tier list is a weather forecast for a storm that has not arrived. Useful for deciding whether to bring an umbrella. A terrible thing to bet your house on. Read it, argue with it, enjoy the chaos of reveal week, and then let the character actually release before you decide the ranking was right.
Deciding on this patch's banners? The honest "should you pull" calls live on each game's banner schedule, updated as the meta settles rather than at reveal.
