| ERCOT and PJM load growth arrives at or above management's guided 4-6% / 2-3% annual bands through 2030 | ||
| Forward power curves begin to reprice structural load rather than near-term weather and cash prices | ||
| The retail-plus-generation combination continues to offset weather shocks rather than compound them | ||
| Data-centre load converts from interconnect-queue position into contracted volume at Vistra specifically | ||
| The buyback continues at scale and management leans into weakness as stated | ||
| The $2.0-2.5bn of cash available for allocation is real and recurring, not a one-off |
Our probability-weighted expected return is +13.7% over 12 months against a street implied return of +57.4% (consensus target 220.9 versus 140.34 at 2026-08-18, 22 analysts, 20 buy / 2 hold). We do not disagree with the | high |
The revision tape is UNAVAILABLE, not neutral. estimates_query.py returns n_up 0, n_down 0, direction flat, breadth 0.0, median_move_pct null and velocity 0.0 for both EPS and revenue over 365 days, and the outliers func | high |
No options-implied anchor exists — no chain entry for VST, and desk_options.json holds ATM-only at one near-dated expiry across the system. The market's own probability for both my bear and bull magnitudes is therefore u | high |
Internal tension in management's own testimony, and it is the crux of the thesis. The same management team states a structurally improved multi-year demand backdrop with quantified load bands (VST/VST_2026Q2.json:6, VST/ | high |
I scored growth 7 on named, quantified, multi-year drivers, but the consensus curve I am using as BASE decelerates hard: +20.8% revenue FY2026, +8.9% FY2027, +4.9% FY2028, -4.5% FY2029. Either the street does not believe | medium |
ROE of 18.5% against ROIC of 3.3% (FY2025 key-metrics) is a ~15pp spread that says the equity return is largely leverage- and buyback-driven rather than asset-level. That argues for a lower quality score than the franchi | medium |
The pack retrieved 59 chunks but only 16 (27%) are on-ticker, and all 16 are VST's own transcripts — a single issuer by identity, so they are one source class, not sixteen independent ones. The 43 off-ticker chunks are A | medium |
| archived | hash | bytes |
|---|---|---|
| 2026-08-19T09:44:36 | 6c6173111849 | 43,996 |