LZLZL/Prediction markets/Ecosystem
FREEJUDGE IT B · RebatesEcosystem

The reward pool
how much is paid out each day

2026-08-21 · check this before copying anyone

The venue really does pay out, and not a small amount. But "money exists" and "you can get it" are different claims — it is split by score, so more participants means less each, and it is concentrated in places you may not be trading.

1Size of the pool

ItemMeasured
Boards carrying a live daily rate8,631
Nominal poolabout $172,177 / day
Actually reaching the "small player" tierabout $77,000 / day
Wallets in that tierabout 5,500
Of those, collecting on three consecutive daysabout 3,800

That tier earns roughly $2 to $150 a day. Spread $77,000 over 5,500 wallets and the mean is about $14/day — but the distribution is severely right-skewed, so the median is far below the mean.

⚠ Nominal pool ≠ paid out

$172,177/day is the sum of every board's daily rate — a nominal ceiling. A board only pays its pool out if someone posts quotes of sufficient quality; where quoting is thin, that money does not materialise from nowhere.

So do not size your share off the nominal pool. Read actual on-chain payouts — they are batched transfers at a fixed daily time, fully enumerable.

2How it is paid

ItemDetail
Frequencydaily
Timeabout 00:45 UTC
Formbatched transfers, roughly 400 per batch
Verifiabilityfully on-chain — anyone can enumerate it

The last row is the valuable property: reward distribution is completely public. You need not trust anyone's claims — count it yourself: who received, how much, how many days running. That is exactly what my census did, enumerating recipients for three consecutive days.

⚠ Two traps when enumerating

① Most public nodes will not serve you. Pulling chain logs in bulk gets refused by most free providers, and the one that worked imposed hard limits (address filter required, a few dozen blocks per call). See eight API traps.

② One address contaminates the statistics. It receives $600k–920k per day inside the batches while having no corresponding trading record on the leaderboard. Any per-capita, median or distribution figure must exclude it. (I only note that it distorts the basis — no identity attribution is made.)

3Where the money actually is

The decision-relevant part: rewards are not spread evenly across categories.

Checking live crypto up/down boards one by one: 1h and 15m return empty, 5m returns only a placeholder config (daily rate 0.001). So market making up/down earns essentially none of this — that lane's only subsidy is the fill-triggered maker rebate.

⚠ "Has rewards" must be checked per board, never inferred

And not with the most convenient field: an empty clobRewards array does not imply no pool. The authoritative source is the CLOB sampling-markets endpoint.

Likewise metadata fields like minimum size and maximum spread exist as part of a config template — their presence does not mean money is paid.

4What collecting it costs

Rewards score quote quality, and the formula specifically pays for tightness: closer to mid scores higher, with quadratic decay, and zero beyond the board's maximum spread.

So it is not free money — scoring forces you tight, and tight gets swept. Measured cost of stepping back:

Quoting atRewardFillsBad-fill $
join best×1×1×1
−1¢×0.59×0.33×0.53
−2¢×0.26×0.16×0.29

Detail in LP rewards.

5How to use these numbers

QuestionAnswer
Can this pool support a person?depends on your share, not the pool total
How do I estimate my share?find wallets running your approach and read what they received on-chain
What is the ceiling here?see the real ceiling on each path
Will the subsidy persist?No guarantee. Rates, splits and categories all change

Row two is the practical one: do not estimate — count. On-chain distribution is public; find a few wallets of the same shape and read their actual receipts. That beats any model.

6One line to keep

Something like $170k of nominal pool exists daily, but it is split by score, unevenly distributed, and outright zero in some categories. The only reliable way to know whether you can reach it is to count what comparable wallets actually received on-chain — which requires trusting nobody.

EvidenceCheck it yourself

Pool full pull of the sampling-markets endpoint: 8,631 boards with a live daily rate, nominal about $172,177/day.
Small-player tier about $77,000/day to roughly 5,500 wallets earning $2–150/day, about 3,800 of them collecting on three consecutive days. From enumerating on-chain recipients for three days.
Payout form about 00:45Z, batched transfers of roughly 400, enumerable on-chain.
Contaminating address receives $600k–920k/day in the batches with no corresponding leaderboard trading; must be excluded from statistics. No identity attribution.
Zero on up/down measured returns for live 1h / 15m / 5m boards.
Cost of stepping back every recorded day × every reward board: reward ×1 / ×0.59 / ×0.26.
Checked 2026-07 to 2026-08. Reward rules change frequently — verify before use.

NextWhere to go

B · REBATES
LP rewards: the scoring formula
F · METHOD
The real ceiling on each path
F · METHOD
Leaderboards are lagging indicators
G · ENGINEERING
Traps when enumerating chain data
This is an educational and research record. It is not investment advice, promises no returns, and offers no personalised trading recommendations. Rules and API behaviour are per the official documentation; this page states when it was checked and both can change without notice. Prediction markets are restricted or unavailable in some jurisdictions — confirm your own before taking part.