CfCapletFi

Vault simulation

Replay report

Directional stress replay from local DeepBook Predict testnet oracle history, calibrated to committed live fill receipts. It assumes near-zero PLP carry (the testnet reality) and so understates the real economics, so treat it as illustrative, not a performance claim. The live vault and its on-chain receipts are the primary evidence.

Replay cycles

751

2026-04-17 -> 2026-06-11

Hedge cost

2,726 bps

Cumulative premium spend over starting NAV.

Payout events

20 / 751

Hit rate 2.66%.

Net hedge cost

-5.705780 dUSDC

Premium spend minus true notional payouts.

Max drawdown, hedged

2,736 bps

Max drawdown, unhedged

3,953 bps

Baseline now carries the same downside the hedge covers, not carry-only.

Ending hedged NAV

34.069350 dUSDC

Derived from first live open receipt budget 1.172486 dUSDC at the 2.5% hedge policy cap.

Ending unhedged NAV

28.361564 dUSDC

Modeled with the same downside exposure the hedge offsets.

Vault strategy simulation

History replay.Live-fill pricing.

simulation on testnet indexer history, model premia calibrated to 42 live fills, not a forward promise.

Open execution ledger
Simulated hedged versus unhedged NAVDaily NAV series from the committed vault simulation JSON.
Hedged NAV Unhedged stress baseline Payout markerunhedged baseline takes the same downside the hedge covers, near-zero carry

Hedge spend and payout days

2026-06-04

3 payout marker(s)

Payouts
2.605351
Hedge spend
0.293792
Cycles
20

2026-06-05

3 payout marker(s)

Payouts
2.579605
Hedge spend
0.189629
Cycles
13

2026-05-27

2 payout marker(s)

Payouts
1.811514
Hedge spend
0.292541
Cycles
19

2026-06-02

2 payout marker(s)

Payouts
1.756474
Hedge spend
0.283713
Cycles
19

2026-06-03

2 payout marker(s)

Payouts
1.747529
Hedge spend
0.281429
Cycles
19

2026-04-21

1 payout marker(s)

Payouts
1.137588
Hedge spend
0.366716
Cycles
19

2026-04-29

1 payout marker(s)

Payouts
1.093443
Hedge spend
0.353086
Cycles
19

2026-05-04

1 payout marker(s)

Payouts
1.052354
Hedge spend
0.338951
Cycles
19

PLP carry, live-calibrated band

0.5x live observed PLP carry

Carry
-0.01 bps
End NAV
34.088629

1.0x live observed PLP carry

Carry
-0.02 bps
End NAV
34.069350

2.0x live observed PLP carry

Carry
-0.03 bps
End NAV
34.030825

PLP carry is derived from committed close receipts: navAfterClose minus prior NAV, plus realized hedge cost, minus any binary payout.

Methodology

  1. 1.Sort settled BTC oracle captures by expiry and replay non-overlapping vault cycles.
  2. 2.From each simulated open time, choose the next settled oracle whose expiry is at least 75 minutes away, mirroring the keeper's minimum-horizon rule.
  3. 3.Use the settlement price of the latest oracle whose expiry is at or before the simulated oracle activation as the spot-at-open proxy.
  4. 4.Snap the downside strike to 99% of that spot proxy, rounded down to the oracle strike grid.
  5. 5.Anchor premium to the median signed live fill ask (1.6959% of notional) from keeper/data/vault_fills.ndjson under the same 99% strike policy; near-expiry SVI is not used to price simulated opens.
  6. 6.Set quantity equal to the 2.5% hedge budget, matching the live keeper policy; an in-the-money cycle pays true notional when settlement is below strike.
  7. 7.Effective cycle spacing is 105.9 minutes across 751 simulated cycles, after applying the 75-minute minimum horizon over this captured oracle grid.
  8. 8.Run PLP carry as a live-observed band from committed close receipts only; if no closes exist, no positive carry is assumed.

Limitations

The spot proxy can be up to about 15 minutes stale, and can be much staler across gaps in the local capture, which can bias strikes versus true open spot.

Premium anchoring assumes the committed live fill ask persists across regimes; the calibration tightens as more cron receipts land.

Near-expiry SVI parameters are kept in the receipt table for transparency but are not used to price simulated at-open premia.

PLP carry is not invented from historical pool yield. It is derived only from committed close receipts; if close receipts are unavailable, no positive carry is assumed.

The unhedged baseline carries the same downside event loss that the hedge is sized to offset, so the drawdown comparison is a stress proxy, not a PLP-only performance finding.

This is a testnet indexer-history simulation, not a forward promise or APY projection.

Premium anchor: signed live fill receipts

TxAskSVI fairSpreadSurface events
2Ty9Fc...RvK2EL4.0597%3.4909%1,629 bpsSVI E4Ktnz...paFJ31Price 5x1ZZc...oVHk15
CNhNWJ...rJwCgm4.2559%3.9227%849 bpsSVI 6NCneR...bN4431Price 8W2PXs...Rp4A15
933KQE...7hGuLF3.2588%2.8508%1,431 bpsSVI 9YGoud...WZHx31Price L8AMC5...LS7515
HzxJ79...XuQ3eh2.7171%2.2381%2,140 bpsSVI AwtG9M...weyr17Price AKFFhm...1PHxj8
4cy8eb...c4mU8S2.7370%2.2669%2,074 bpsSVI BYwTF9...HetW31Price 5vxF3H...Wr8c15
fmt9YT...axLrVx2.3106%1.8107%2,761 bpsSVI 2Yr2Ub...pnca17Price 2Yr2Ub...pnca16
GzWnqt...Xftx5L2.1683%1.6510%3,134 bpsSVI 6KBYV5...nLtH31Price AG7K1y...sqLA15
EEQVe2...AVHG5K2.2310%1.5743%4,171 bpsSVI Hn9oW9...8J6b31Price 41imyT...xtjv15
CfQkHQ...mXZrjM3.5157%3.0157%1,658 bpsSVI DcGSwW...z1SL29Price 6VmrU3...hQNd14
GPQYXy...TJp7Ut8.4742%7.8231%832 bpsSVI 5SjZYS...fLhs29Price HFWT8P...VAbk14
67Mi3A...18j28t1.3777%0.8587%6,044 bpsSVI F8rza7...yCJ233Price 9drkVq...Sbmf16
HvsAU4...mfh7ij1.9409%1.4410%3,469 bpsSVI HkaeWS...LHqm33Price aimVcV...d2xt16
3mHLEm...DMYwnB1.8283%1.3311%3,735 bpsSVI 5y4j3q...Djqu33Price 2S8ArR...UUaD16
7umcRr...KaE6uT1.0838%0.6095%7,783 bpsSVI 6kKAAG...sWYP19Price ARwx6g...xxzgt9
FK9xKN...L2XrSi1.2537%0.7536%6,637 bpsSVI BhMWqt...1LcT33Price FcuQdM...83om16
DXDv6L...fvSQkU1.0131%0.5132%9,742 bpsSVI Dtk4rx...jQU619Price FYuJFR...jQc1b9
7bLDfK...uBt88a1.0051%0.5045%9,925 bpsSVI CXWSFj...aWiZ31Price CXWSFj...aWiZ30
3X7PjH...RWazqk1.5568%1.0517%4,803 bpsSVI HZxNLn...NREj17Price 4375k7...EH6ZG8
69GT1t...pBYyrd1.6566%1.1567%4,322 bpsSVI 39bWWZ...fUXW33Price 39bWWZ...fUXW32
B1cmJp...qstWPr2.2168%1.7644%2,564 bpsSVI GNH8U4...GWrW19Price GNH8U4...GWrW18
ETHgZn...ZhWE1h2.4868%1.9868%2,517 bpsSVI 7x7S69...adTe33Price 7L6Jzw...Tzgt16
8YK3vE...5G1Lqg2.9263%2.4265%2,060 bpsSVI 2zXC8W...AupY19Price 2zXC8W...AupY18
8KPz5k...1i7vNK3.6166%3.1245%1,575 bpsSVI EAP2NC...V8ym33Price GgjS7F...cRQf16
8U7mda...vQEURo6.2411%5.7197%912 bpsSVI 595ZaC...2iqe33Price BkfWB4...MkJE16
tgpxzC...558KLU1.6944%1.1949%4,181 bpsSVI Cn9jrM...UjxV19Price Cn9jrM...UjxV18
HQyJzP...UKBihd1.2769%0.7790%6,391 bpsSVI C6GRCP...kwxi31Price BwcJsU...MnJG15
HdEcPy...PX9T1F1.2806%0.7806%6,405 bpsSVI EWePAk...237N31Price Auav8x...as7J15
6CWSLh...K46Y7P1.0246%0.5246%9,529 bpsSVI 8c3X3F...JBM917Price EVzbLp...AqmFw8
6PbzHo...zuh1dX2.2378%1.6907%3,236 bpsSVI Hah38V...QC4317Price Hah38V...QC4316
ErGHRE...NBkQw81.5907%1.1032%4,419 bpsSVI AVGt9Q...Dynm31Price 9KwbUP...hqLY15
ANpQxw...m38ALu1.2948%0.7960%6,266 bpsSVI 5RDWx1...hkVz31Price 5RDWx1...hkVz30
nimcEb...6Zv7wk1.4265%0.9265%5,397 bpsSVI GPAzJH...ipQG31Price 7yNPY7...zuh515
7GF4yC...8XCjPo1.7866%1.3602%3,136 bpsSVI C1wF7Q...vB7C31Price AfBjjF...RMoT15
5BEfQo...6eQWb11.6328%1.0469%5,596 bpsSVI UoYdsp...4NBQ31Price Epi22z...12nr15
5TErgV...tWTWDq1.6974%1.1915%4,246 bpsSVI GhgZKN...ome617Price FCv5Vp...LpPE48
4fAuxt...thG68T1.7081%1.2095%4,122 bpsSVI CGRN4d...vE7229Price 6Wg7wq...6YPF14
3sxyMn...BFeWz61.2535%0.7395%6,951 bpsSVI LN6YB7...4mcD31Price 9KVYqa...j6iy15
WbdEY9...D654wv1.1631%0.6632%7,537 bpsSVI 6PZXjx...KkgV17Price HKHJWU...zsnaK8
DBEiJT...MW3JxG1.4776%0.9776%5,114 bpsSVI Avxqkc...jkn831Price 2GxgFv...AfDV15
9ZdcUt...9Kk9Sw1.2861%0.7822%6,443 bpsSVI Af8Fui...MmNT31Price sC3xvi...BdaZ15
GN6uNG...Pnv55Z1.0270%0.5271%9,486 bpsSVI 9JmAcT...wx9e31Price GCubRz...XZk515
4Ftqqo...xprcB71.3353%0.8441%5,818 bpsSVI FTiAF9...h5Bf31Price DqKtQv...wiBU15

Sim premium rate: 1.6959% of notional, anchored to 42 committed live open fill(s). SVI fair is shown for receipt transparency, not used to price this replay.

Contracts and raw links

Committed sim JSON/sim/vault_sim.jsonExecution ledger API/api/vault/ledgerDataset reproducibility notehttps://github.com/dolepee/capletfi/blob/main/keeper/data/DATASET.mdLive fills filehttps://github.com/dolepee/capletfi/blob/main/keeper/data/vault_fills.ndjson