Power-Cycle v1.1 — Initial Deploy (2026-05-18)
Founder approval: 2026-05-18 ~8:12am ET via iMessage ("I think we can paper trade it."). Orders submitted: 2026-05-18 08:16:56 ET (~1h 13m pre-open). Execution window: Opening of regular trading 09:30 ET, 2026-05-18 (Monday).
Allocation rationale (post-KTOS-drop)
Per founder pushback 2026-05-18 8:07am ET — KTOS was DOWN 19-34% YTD, contradicting the sub-agent's "AI-power optionality unpriced" framing. The FTT-Boom-Superpower connection is real (Kratos' Florida Turbine Technologies designs Boom's Symphony engine core), but the optionality isn't material enough to KTOS earnings (~5-8% revenue) to overcome whatever's dragging the stock. Dropped from basket; 0.75R redistributed.
Final allocation, R-weighted by signal strength:
| Ticker | R | Notional | Signal stack |
|---|---|---|---|
| GEV | 1.5R | $7,500 | Backlog $163B +67%YoY; Q1 electrification orders > all of 2025; only Western SMR with construction underway (BWRX-300 2029); Tiger Global holds. ONLY basket name up YTD per founder observation. |
| TLN | 1.0R | $5,000 | First-mover Amazon nuclear-data-center deal (Susquehanna 960MW PPA); already re-rated; structural-tailwind ride |
| VST | 0.75R | $3,750 | 2-manager smart-money (Tepper + Druckenmiller — only basket name passing smart-money-mirror's 2+ threshold); 16.4x fwd P/E vs CEG 24.7x; new 2.2GW Meta PPA |
| CEG | 0.5R | $2,500 | Flagship nuclear pure-play but premium-priced 24.7x P/E; half-weight of VST given cheaper alternative for similar exposure |
| CCJ | 0.25R | $1,250 | Uranium upstream pull-through; zero smart-money corroboration; weakest fundamental link; sized smallest |
| Total | 4.0R | $20,000 | Full bucket cap deployed |
Account state — pre-trade
| Field | Value |
|---|---|
| Account status | ACTIVE |
| Cash | $100,000.00 |
| Buying power | $186,810.26 (after smart-money-mirror v1 orders that morning) |
| Portfolio value | $100,000.00 |
| Open positions | 0 (smart-money-mirror orders pending fill) |
| Trading blocked | False |
Validation passed: status ACTIVE, buying_power >> $20k threshold, no overlap with existing positions.
Orders submitted (5 of 5 accepted)
All orders: side=BUY, type=MARKET, time_in_force=DAY, notional=
Notional fractional orders fill at the first available market price on the next regular-hours open (2026-05-18 09:30 ET).
| # | Ticker | Notional | R | Order ID | Status |
|---|---|---|---|---|---|
| 1 | GEV | $7,500.00 | 1.50R | b94a64b1-12ee-445f-9c93-ccdd09effa78 | PENDING_NEW |
| 2 | TLN | $5,000.00 | 1.00R | ad29359c-1124-4c44-9c0d-955565ef61e1 | PENDING_NEW |
| 3 | VST | $3,750.00 | 0.75R | 79a7a271-6f25-49fb-800f-18e8e7c973c2 | PENDING_NEW |
| 4 | CEG | $2,500.00 | 0.50R | a102accf-0cc8-4301-a6e3-b18f701b2895 | PENDING_NEW |
| 5 | CCJ | $1,250.00 | 0.25R | cc9e7564-3a5a-4ea1-8ed1-b36ba12db701 | PENDING_NEW |
Total submitted: $20,000.00 / 4.0R / 5 of 5 orders
Cohort attribution
Each position row written to ~/rdco-vault/01-projects/investing/positions/_attribution-ledger.csv with strategy=power-cycle-v1.1. Quarterly P&L rollup can split Alpaca's combined portfolio by strategy.
Next actions / monitoring
- Fill check — pull positions + orders via Alpaca after 09:30 ET open today. Expect all 5 fractional DAY orders to FILL at the opening cross or shortly after. Log actual fill prices + slippage vs the Friday-close reference. Update CSV ledger with entry_price + actual_notional.
- Same fill-check cron as smart-money-mirror — 09:45 ET cron should be extended to cover both strategies' fills.
- Quarterly anchor review — 4 anchors per v1.1 spec: (a) ERCOT interconnection queue, (b) hyperscaler PPA cadence, (c) GEV order velocity, (d) smart-money 13F deltas on VST/GEV. Surface to founder only on flip.
- ARK backfill in flight — ARK Invest 13F backfill dispatched same morning; if ARK holds GEV or VST that adds a 3rd-manager confirmation (currently single-manager Tiger Global on GEV, 2-manager Tepper/Druckenmiller on VST).
Errors / rejected orders
None. All 5 orders accepted with PENDING_NEW status.
Risks called out at deploy
- N=1 cycle for the actual thesis test. The 2024-AI-power cycle is the only one where this specific thesis applies; analog cycles (1999-dotcom, 2007-commodity, 2010-shale, 2017-Texas-energy) are robustness checks. Leave-one-out drops V2 to +23% — same AI-supercycle dependence as memory v1.1.
- GEV's outsize sizing (1.5R = 37.5% of bucket) reflects the strongest signal stack but concentrates outcome risk on one name. If GEV underperforms while the rest of the basket holds, bucket P&L is dominated by GEV.
- Smart-money signal on VST + GEV is <12mo old. Could reverse on next quarterly 13F (mid-Aug).
Kill switch
Founder channel "halt power" or "pause investing" = immediate close ENTIRE bucket.
Related
- [[../../theses/2026-05-18-power-cycle-v1.1.md]] — full thesis spec
- [[../../backtests/2026-05-18-power-cycle-v1-v2-honest-rerun.md]] — backtest-v2 honest rerun
- [[../../anchors/power-layer/2026-05-18-pressure-test-and-winners-survey.md]] — pressure-test that surfaced basket adds
- [[../_attribution-ledger.csv]] — cross-cutting position ledger
- [[2026-05-18-deploy-result.json]] — raw API response