Bullpen.News From Edwin Dorsey
Get Daily Recap Feedback

Sunday, September 13, 2026

Back to today

From X

@MultiplesCap

Multiples Capital on X

Not an AI expert by any stretch, but wonder if US labs actually coordinate to "pace the frontier," it could be a stealth negative for Nvidia’s top-line growth rate. Here’s why: Flattening the Frontier Race: If scaling massive base models is intentionally delayed for alignment, the panic to hoard the newest, most expensive contiguous GPU clusters cools down. Shift to Post-Training & RL: Progress shifts to squeezing intelligence out of existing models via synthetic data and self-play. This relies on inference-class rollout fleets rather than cutting-edge pre-training mega-clusters. The Verification Layer: Evaluating, sandboxing, and grading all that synthetic RL data requires massive CPU compute .... benefitting $INTC Less brute-force pre-training = extended hardware refresh cycles and a normalized upgrade curve for $NVDA. Markets much smarter on this than I, will wait into see how the stock prices react to this next week!
@Fierce__beast

Fierce_beast on X

For $CASY, let's say it grows earnings roughly 6% on average for the next 20 years, then 3% terminally. Maybe its 9% for the next 5 years, but gradually moves toward something like maturity of 3%, average let's say is 6%. What multiple would you currently pay on that earnings stream? Personally, i would pay no more than 25x for this. Add other considerations to the business, maybe you think it will grow faster or longer, adjust your multiple as you prefer.

Investor Letters

All letters

Lululemon: The 15-Minute Thesis

Lululemon’s North American sales, margins and legging demand are falling as Alo and Vuori gain share, leaving its brand moat and depressed…

LULU $11B
Hated Moats 2 clicks

The Five Deals That Made Apollo

Apollo Global Management’s five defining deals, from Executive Life to Intel, built a $1 trillion asset manager spanning insurance and…

APO $74B
Net Interest 1 click

In Conversation: NVIDIA

NVIDIA sees inference, agentic workloads and robotics broadening AI compute demand beyond hyperscalers, while high-speed memory shortages…

NVDA $5.5T
Aurelion Research 1 click
ACN · long

Accenture PLC

Accenture has been repriced as an AI loser despite its expected near-term benefit from AI adoption.

ACN $111B
Value Investors Club 1 click

Deep Dive: Amazon (Updated)

Amazon’s AWS grew 28%, with a $364 billion backlog, supporting the case that up to $200 billion of annual AI capex can earn solid returns.

AMZN $2.7T
Rebound Capital

MercadoLibre: Valuation Model

MercadoLibre’s sum-of-the-parts model values commerce and payments on FCFF, credit on equity returns, producing a $1,980 base-case target…

MELI $93B
Hated Moats

Meta On the Offense!

Meta’s Muse Spark 1.1 targets agentic enterprise work, backed by employee workflow data and 3,000 engineers building reinforcement-learning…

META $1.7T
MBI Deep Dives 1 click

U-Haul Bought at the Top

U-Haul’s pandemic-era fleet purchases at inflated prices now drive higher depreciation and disposal losses as truck and van prices…

UHAL $13B
scuttleblurb 2 clicks

Deep Dive: Palantir ($PLTR)

Palantir’s forward-deployed engineering model underpins its AI software growth, but AWS, Microsoft and AI labs are copying it at scale as…

PLTR $440B
Rebound Capital

Blue Chips

Project Kilby, a planned Texas behind-the-meter data center twice Project Stargate’s size, would reshape Permian gas flows and boost demand…

GEV $249B
Doomberg

Publications we follow

Every publication on this site, hand-picked.

108 publications

Newsletters 91

YouTube 4

Blogs 12

Seeking Alpha 29 of 159

Nominate a publication

X accounts we follow

Every account on this site, hand-picked.

68 accounts
Nominate an account

Investor letters we follow

78 firms

Sectors

Ordered by how often our editor reaches for them.