Leaderboard Tiers vs Pro Rata Distribution

Every competition has to answer one question: given a final leaderboard and a pool of money, who gets what. Two answers dominate, and they produce completely different contests from identical trading.

Pro rata: everyone who qualifies, in proportion

Pro rata splits the main pool by share of counted points. Your reward is the pool multiplied by your points divided by the total points of the qualifying field.

Three properties follow.

It pays a wide field. Every trader above the eligibility filters gets something. Nobody trades for two weeks and finishes with zero because they came fourth.

Its variance is low. Your outcome is a smooth function of your effort. Doubling your volume roughly doubles your reward, and the derivative is stable — which is what makes the projection column on a live leaderboard genuinely useful for deciding whether to keep trading.

It needs a floor. Without a qualifying minimum, a pool split across a large field shreds into thousands of dust payments that cost more to distribute than they are worth. This is why pro-rata competitions almost always carry a minimum volume to qualify — and the filter runs before the split, so traders below the floor do not dilute the pool for the traders above it. That makes a high floor good news for anyone who clears it.

Pro rata is the honest default for an open competition where anyone can register. It is also the structure that most rewards being large, because share of pool is share of volume and volume scales with capital. The daily cap is the counterweight.

Leaderboard tiers: fixed shares to fixed positions

Tiers assign a percentage to each finishing position: 1st takes 50%, 2nd 30%, 3rd 20%, and nothing below. The split is decided when the competition is created and does not move.

The high-variance property is the whole design. Tiers exist to create a race worth watching — a leaderboard where positions matter and the top of the table is contested. Pro rata produces a fairer distribution and a duller contest.

The cost is that marginal volume is worth nothing unless it moves your rank. A trader sitting fifth in a three-paying-position competition gains exactly zero from another million dollars of volume unless that million overtakes third. That discontinuity is why an average entrant in a large tiered field has an expected value far below what the headline pool suggests — the arithmetic is worked through in are trading competitions worth it.

What happens as the field grows

This is the axis on which the two structures diverge most sharply.

Pro rata degrades gracefully. More entrants means a smaller share each, but the shape of the payout is unchanged and everyone who qualifies still gets paid. Doubling the field roughly halves each reward. Unpleasant, but continuous.

Tiers degrade discontinuously. A three-position tier array pays three people whether the field is 10 or 10,000. At 10 entrants, 30% of the field cashes and the competition feels winnable. At 1,000, 0.3% cashes and it is a lottery with worse odds than the actual lottery. Nothing about the configuration changed; only the field did.

Which is why the sensible rule is: tiers for small or gated fields, pro rata for open ones. A tiered pool on a permissionless competition that anyone can join is a structure whose fairness depends on nobody showing up.

The reverse failure also exists. A tier array with more paying positions than there are qualifying traders simply leaves the extra positions unfilled — a five-position split in a three-trader field pays three.

Percentage-of-field buckets

There is a third shape, available on entry-fee PnL challenges, that fixes the field-size problem directly: instead of naming positions, you name slices. "The top 10% share 60% of the pool, the top 50% share the remaining 40%."

The winner count then resolves against the final eligible field rather than being fixed at creation. The top p% of n entrants is max(1, floor(n × p / 100)) — floored so everyone counted as a winner genuinely sits inside the stated slice, with the minimum of one so a bucket in a tiny field still crowns someone rather than stranding its share.

Inside a bucket, the money is not split equally. It decays down the ranks like a poker tournament payout table: each position carries a weight of rank raised to the power of −0.8, normalised across the bucket. That exponent gives roughly a 1.74× step from second place to first — top-heavy at the head, with a long paying tail so the last seat inside a wide bucket still receives a visible amount rather than dust. A flat split inside a bucket would give eighth place the same cheque as first and remove any reason to keep fighting once safely inside.

Two corrections keep it honest. Weights use the global rank rather than restarting at each bucket boundary, and a final pass flattens any boundary where a lower rank would otherwise out-pay a higher one into an equal-payout band. Because that flattening moves money across bucket lines, a bucket's actual share can differ from its stated share at a given field size — so any surface showing per-bucket percentages next to a live field has to read the resolved number rather than echo the configuration. Telling traders 20% while settlement pays 29% is the failure this avoids.

Buckets that resolve to zero members in a small field fold their share upward into the winners above them, proportionally to what those winners already hold, so the pool always sums to what the page advertised.

Tie handling

Ties are where a payout structure quietly breaks if nobody thought about them.

Pro rata has no tie problem. Equal points means equal share by construction. Nothing special happens.

Tiers pool and split. When several traders share a rank, the tier percentages covering the positions that block occupies are added together and divided equally among them. Two traders tied for first in a 50/30/20 split share 80% of the pool — 40% each — and the next trader down takes the 20% at third position rather than being promoted.

That is the fair reading: the tied traders collectively occupy positions one and two, so they collectively receive what positions one and two pay. No coin flip, no arbitrary tiebreak, and the pool still sums to 100%.

Ties are common on volume leaderboards where equal scores compress onto a shared rank, and rare at settlement, where ranks are assigned densely and are tie-free by construction.

Which one should a creator pick

Pro rata when the competition is open, the field is unpredictable, and the goal is broad participation — most volume campaigns. Pair it with a qualifying minimum sized so the smallest reward is worth distributing.

Tiers when the field is small, gated, or self-selecting, and the goal is a contest with a visible winner — duels, invitationals, and small challenges. Keep the number of paying positions proportionate to the field you actually expect.

Percentage-of-field buckets when you want tier-like concentration but cannot predict the turnout. They are the structure that behaves sensibly at both 8 entrants and 800.

Which one should a trader pick

If you are not confident of finishing in a paying position, pro rata is the only structure where your effort maps to a reward at all. If you are, tiers pay dramatically more for the same volume. Read the distribution mode before the prize pool — it decides which of those two traders you are.

Browse live competitions, or see how projected rewards are calculated for the step-by-step math behind both modes.