Comcast Corporation

The cable broadband plant is one of the finest infrastructure franchises in the country — a geographic near-monopoly where the marginal broadband subscriber is essentially pure margin and switching costs are stickier than most people appreciate. At current multiples, the market is pricing something approaching structural decay, which is too harsh given that subscriber losses, while real, remain a trickle rather than a flood. The buyback engine retiring a substantial share of the float annually is genuinely value-accretive compounding that patient holders benefit from while the market looks away. The business is navigating an uncomfortable transition: the heavy-investment phase may be cresting, and management's H2 2026 inflection narrative — promotional conversions to paid wireless, pricing rationalization, Epic Universe ramping — is plausible rather than implausible. Wireless at 9 million lines and growing is a legitimate bundle weapon if the free-line-to-paid conversion rate holds; theme parks with IP differentiation are the most interesting long-duration asset in the portfolio at a moment when the world is over-indexed on digital content. The conglomerate structure, with its media overhead and streaming losses, remains an anchor on returns that the core infrastructure business cannot fully escape. The single variable that restructures the entire thesis is broadband net subscriber losses compounding rather than plateauing. Fixed wireless access improves every year, AT&T fiber is overbuilding directly in Comcast's footprint, and the defensive response requires enormous capital that earns only adequate returns. If broadband penetration reverses structurally rather than temporarily softening, ARPU growth cannot compensate for volume decline — and the whole financial architecture built on that toll road collapses under a debt load that leaves equity holders no margin for error.