League Tiers Explained

If you stream on TikTok LIVE, the Creator League screen asks you the same thing every day: how many diamonds do I actually need? To climb, to hold your tier, to not slide back down. There’s no number posted anywhere — and the reason why turns out to be a surprisingly deep estimation problem.

Before the theory, though, meet the competition. These are the creators at the very top of their league right now — switch leagues in the sidebar and the board follows:

That’s who you’re ranked against. And where you land relative to them — not any fixed diamond total — is what decides whether you earn, keep, or lose a fragment each day. Here’s how that machine works, and how to read it.

How the league works

Creators are sorted into eighteen tiers — A1 at the top down to D5 (League A is the short one, just A1–A3; B, C, and D each run 1–5) — and every day you’re ranked inside your league by the diamonds you collect that day. Your daily rank drops you into a percentile band, and the band hands you a fragment change. Bank four fragments and you climb a league — lose four and you drop one.

The catch — and the whole reason this is hard — is that the reward isn’t tied to a diamond amount. It’s tied to your percentile. Each league plays its own schedule:

A few things fall out of this. The top leagues (A and B) add a brutal Top 1% tier worth a full +3 that the lower leagues don’t even offer. And the keep line gets stricter the higher you climb: in League A, slipping below the top 60% already bleeds a fragment (the bottom 40% lose one), while down in C you hold as long as you clear the top 80% — only the bottom fifth slips. In D2–D5 you can’t be demoted at all. The higher the league, the thinner your margin for an off day.

The real question hiding inside

Because the bands are percentiles, “how many diamonds for the top 10%?” is really asking: what daily diamond total lands at the 90th percentile of my league, today?

That’s not a property of you — it’s a property of the crowd you’re ranked against. Which is exactly why the number moves daily, differs by league, and is never published anywhere. To estimate it, we have to reconstruct the distribution of daily diamonds across a league. That’s where the data comes in — and where it runs out.

What we can see — and what we can’t

A leaderboard exposes the top ~99 of each league with their diamond counts. Plot them and the shape is unmistakable: a steep power-law drop at the very top, flattening as you go down.

You can read two things cleanly from that curve: the shape (how fast diamonds fall) and the scale (how loaded the top is, which lets you compare one league to another). What you can’t read is where the percentile lines fall — because in a league of thousands, the top 10% sits hundreds of ranks below anyone the leaderboard shows.

Change the assumed headcount and the cutoffs slide along the fixed curve. That single number is the one fact nobody hands you, and the whole absolute estimate hinges on it — which is why the calculator asks you to pin it from your own live screen rather than guess.

So what can we say without the headcount?

More than you’d think:

• A hard ceiling. Every cutoff is necessarily below the lowest diamond count we can see, so the bottom of the visible board is a guaranteed upper bound on all of them.
• Relative difficulty. Comparing leagues — “this one’s top-10% line is about 3× that one’s” — doesn’t depend on the headcount at all. It cancels out. Flip between leagues in the sidebar and you’re reading exactly this.
• The Top 1% is often visible. Because it’s so shallow (1-in-100), the +3 line for Leagues A and B frequently sits inside the top 99 — readable directly, no guessing.

And because the ranking resets daily, every day hands us a fresh board and a fresh real outcome (you finish in a band the app shows you). A couple of weeks of “I earned X and landed in the top Y%” is enough to back the headcount out empirically — which is exactly what the live capture behind this post is now logging, every half hour.

How to actually climb

Once you internalize that ranking is daily and percentile-based, the strategy writes itself:

• Clear the next band, then stop. Each day’s reward is a staircase: once you’ve passed a band’s cutoff, extra diamonds earn nothing until the next, far-higher cutoff. Fragments, meanwhile, add up across days — a +1 today and a +1 tomorrow is +2, same as one +2 day. So a diamond that would overshoot today is worth far more saved for a day it can clear a fresh band. Consistency beats hero days.

• Know your keep line. On days you can’t contend, the goal shifts to clearing the keep threshold (top 60% in A, up to top 80% in C/D1) so you don’t bleed a fragment. Defending is cheaper than re-earning.
• Exploit the safety net. In D2–D5 you can’t be demoted, so there’s zero downside to experimenting with schedule, format, and gift mechanics while you’re down there.

Where this leaves you

Two things stay honest in the numbers. Each cutoff is measured (read straight off the leaderboard) or estimated (extrapolated past the visible top 99) — and every absolute figure rides on the assumed headcount, the one number no leaderboard reveals.

But it compounds in your favor. We record all 18 boards every half hour, so the more days we bank, the more the curve averages out a noisy night or a one-off megagift — the shape, and the league-to-league comparisons, only get steadier. Pinning the headcount for real is the last mile: one outcome you can hand it (“I earned X and the app showed top Y%”).

So that’s the machine. Now go run your own numbers — pick a league and see the diamonds for every band, then pin it exactly with one number off your live screen:

Working draft — the model is built, the data is live and accumulating.