Worked examples

These examples work through the arithmetic for cover buyers, underwriters and stakers using the illustrative rates: 12% for the 15 min tier, 4% for the 1 h tier and 1% for the 3 h tier, per 30-day cover period. The protocol fee is 10% of premiums, split 5% to $STALL stakers and 5% to buy and burn; the other 90% is credited to the tier's vault.

Each example covers one 30-day cover period. It assumes any halt begins during the period and that the halt duration measured by the Sequencer Uptime Feed equals the stated duration. Amounts are in USDC and ignore Base transaction fees. Because the method for valuing staked $STALL has not been finalised, first-loss capital is given as an illustrative USDC value.

The simulator reproduces all five examples.

Example A: borrower, 2 h halt#

A borrower with a loan on Base buys 10,000 USDC of 1 h cover. The premium is 10,000 × 4% = 400 USDC. During the period, Base halts for 2 h.

The payout is the fixed payout size. It does not depend on what the halt actually cost the borrower, which could be more or less than 10,000 USDC.

Example B: borrower, 45 min halt#

Same cover as example A, but the halt lasts 45 min. That crosses the 15 min threshold only, so the 1 h tier is not triggered.

The position is not closed by the 45 min halt. It stays active for the rest of the period, and a later halt longer than 1 h in the same period would still trigger it.

Example C: market maker, 25 Jun 2026 halt#

Suppose a market maker had held 250,000 USDC of 15 min cover during the 25 Jun 2026 halt, reported on status.base.org (opens in a new tab) as lasting about 2 h. STALL was not live at the time; this example is hypothetical.

Of the 30,000 USDC premium, 27,000 USDC is credited to the 15 min tier vault, 1,500 USDC goes to stakers and 1,500 USDC to buy and burn.

With the 1 h tier instead, the premium would have been 250,000 × 4% = 10,000 USDC, and by reported duration this halt would also have triggered it. The 15 min tier costs more because it is triggered by more halts: by reported duration, the 43 min and 33 min incidents would have triggered the 15 min tier but not the 1 h tier.

Example D: tier at capacity, 43 min halt#

This example looks at a whole tier. The 15 min tier has 1,000,000 USDC of underwriter deposits and 100,000 USDC of first-loss capital, and has sold cover up to capacity. A 43 min halt then begins during the period.

As in the simulator, capacity here is measured against the 1,000,000 USDC of deposits; the premiums credited during the period are shown separately.

A 43 min halt crosses the 15 min threshold, so every cover position in the tier pays at once. It does not reach 1 h, so the 1 h and 3 h vaults pay nothing.

Underwriters lose 88.12% of their 1,000,000 USDC deposit, net of premiums. Stakers lose 93.4% of their 100,000 USDC first-loss capital, net of fees. After the halt, the vault holds 118,800 USDC.

Example E: tier in a quiet period#

The same tier as example D, at capacity, but no halt begins during the period. Every position expires and no payout is made.

These are figures for one 30-day period, not annual rates. Set against example D, the underwriters' loss in a single triggered period (881,200 USDC) is larger than seven quiet periods of premium income combined (7 × 118,800 = 831,600 USDC).

Reproducing the examples#

The simulator applies the same rates, fee split and loss waterfall to one tier and one 30-day period. Enter the payout size or cover outstanding, the trigger tier, the vault assets, the first-loss capital and a halt duration to reproduce any example above. For the rules behind the numbers, see How cover works, Underwriting vaults and Fees.