Skip to content

Revenue Clearing Rationale

mister-mayflower edited this page Nov 14, 2026 · 4 revisions

On a public blockchain, accepted commitments and funding reveal the current pot and outstanding shares. With a fixed deadline, a bidder who waits gets to see more of that activity and gives others less time to react. Future bids, funding, and transaction inclusion remain uncertain.

Sniping exploits the fixed close. If a token buys the same number of shares throughout the auction, a bidder can watch the pot and competing shares grow and decide at the last moment whether to participate. An early bidder commits while those values are still changing and exposes their bid for others to react to.

A stink bid is a large apparent commitment that makes the reward per share look unattractive to other bidders. If it is free to cancel near the deadline, the bidder can recover the principal after others have decided to sit out or commit less. The bid influenced their choices despite carrying little cost for the bidder who posted it.

Make the bid cost something

In revenue clearing, a deposit puts actual clearing tokens in escrow. The deposit increases visible competition and creates a claim on the pot if its shares survive. A withdrawal gives up the shares backed by the withdrawn principal and burns a fraction of that principal. Any clearing tokens still committed at finalization are burned in full.

Burning the principal keeps the payment for shares separate from the funded reward asset. Bidders supply their own valuations when bidding, and the pot pays out in the asset it received. Revenue Clearing

Why the entry bonus is linear

Let $u=t/T$ be the fraction of the epoch that has elapsed. A new deposit earns the following ideal number of shares per clearing token.

$$w(u)=1+b(1-u),\qquad 0\le b\le2.$$

Entering early earns extra shares in exchange for committing before the final pot and competition are known. Waiting lets a bidder observe more data or avoid committing entirely if the auction turns out to be unattractive. The linear curve gives up the same amount of potential weight for each second of waiting. This makes the tradeoff predictable even when bids and funding arrive in bursts. With $b=1$, one token buys two shares at the start and 1.5 halfway through. A deposit retains its own shares as the entry bonus falls for later deposits.

The slope $b$ sets the size of that incentive. At zero, every deposit has unit weight. At the maximum slope of two, a deposit at the start receives three shares per token. The baseline of one near the deadline leaves late bids able to compete.

Why the exit penalty can be quadratic

Early in the epoch, other bidders have time to react when new funding or competition prompts someone to revise a bid. Later, a bidder has had longer to observe how others responded to their displayed commitment. Canceling close to the deadline can leave others little time to revise decisions made under that apparent demand.

The ideal penalty on the withdrawn principal is

$$p(u)=u^k,\qquad k\in\{1,2,3,4\}.$$

With $k=2$, the ideal penalty rises by 6.25 percentage points over the first quarter of the epoch and by 43.75 points over the last quarter, up to the deadline limit. This gives bidders room to adjust early as the pot and other bids take shape. It makes the option to keep a bid visible until others have acted on it expensive to exercise. The curve uses elapsed epoch time as a proxy for that opportunity. It applies equally to a fresh lot and an old lot withdrawn at the same time.

An exponent of one gives a constant rate of increase. Larger exponents lower the penalty between the endpoints and concentrate more of the rise near the deadline. A channel can reward early entry strongly and still give bidders room to revise. At $b=2$ and $k=4$, a deposit at the start earns three shares per token and a halfway withdrawal has an ideal penalty of 6.25%. Time Weight

What still depends on the auction

  • Late bids can still earn more shares by committing more tokens. Inclusion and ordering still determine which actions land before the deadline.
  • A sole bidder with positive shares receives the whole funded pot. Economic competition depends on other bidders showing up and committing tokens.
  • The funded reward and surviving shares determine the payout. Promised funding carries no claimable balance, and remaining principal burns even when the funded pot is empty.

Every claim uses the same final pot and share total. Bidders can calculate the effect of an accepted action from public state, while the eventual payout depends on the rest of the auction. Formal Specification

Clone this wiki locally