Offshore
Photo
Wasteland Capital
$AEO closed today at +128% since the late July entry.
Many quick flipped this at +30-50% on the back of the media attention, instead of trusting the fundamentals.
In cases like this, you have to let the fundamental case play out. Let it cook. Compounding always wins. https://t.co/Y1Qy0xbtpT
tweet
$AEO closed today at +128% since the late July entry.
Many quick flipped this at +30-50% on the back of the media attention, instead of trusting the fundamentals.
In cases like this, you have to let the fundamental case play out. Let it cook. Compounding always wins. https://t.co/Y1Qy0xbtpT
Sydney Sweeney doing the fall campaign at $AEO.
Consensus EPS next year is $1.23 (Feb-27), which implies 8.5x P/E at the $10.50 price. Stock’s beaten down due to poor (-3%) Q1 SSS & tariff-induced margin pressure.
I think they can do $1.60 in EPS, which would be $24 @ 15x P/E. https://t.co/SiNGjPetF8 tweet
Offshore
Photo
Fiscal.ai
Shares of Snowflake are down 22% over the last 5 years.
Meanwhile, revenue is up 659%.
Valuation matters.
$SNOW https://t.co/WNJ787sO29
tweet
Shares of Snowflake are down 22% over the last 5 years.
Meanwhile, revenue is up 659%.
Valuation matters.
$SNOW https://t.co/WNJ787sO29
tweet
Offshore
Photo
Dimitry Nakhla | Babylon Capital®
RT @DimitryNakhla: MercadoLibre is trading near its lowest multiple in the past 2 years (P/E & FCF Yield) 💸
Here’s $MELI projected CAGR through 2028 assuming EPS hits $106.00 (-10% below consensus):
36x P/E: $3816💵 | ~20% CAGR
34x P/E: $3604💵 | ~18% CAGR
32x P/E: $3392💵 | ~16% CAGR
30x P/E: $3180💵 | ~14% CAGR
28x P/E: $2968💵 | ~11% CAGR
___
Even assuming EPS comes in 10% below consensus and a 28% multiple contraction (ending at 30x), $MELI still offers compelling return potential
tweet
RT @DimitryNakhla: MercadoLibre is trading near its lowest multiple in the past 2 years (P/E & FCF Yield) 💸
Here’s $MELI projected CAGR through 2028 assuming EPS hits $106.00 (-10% below consensus):
36x P/E: $3816💵 | ~20% CAGR
34x P/E: $3604💵 | ~18% CAGR
32x P/E: $3392💵 | ~16% CAGR
30x P/E: $3180💵 | ~14% CAGR
28x P/E: $2968💵 | ~11% CAGR
___
Even assuming EPS comes in 10% below consensus and a 28% multiple contraction (ending at 30x), $MELI still offers compelling return potential
tweet
Offshore
Photo
WealthyReadings
As expected.
You sometimes have to miss one to be protected for the next mistakes.
Still hurts to miss a 20% move.
Congrats to shareholders!
$PATH
tweet
As expected.
You sometimes have to miss one to be protected for the next mistakes.
Still hurts to miss a 20% move.
Congrats to shareholders!
$PATH
I am intentionally walking away from one of the best setups on the market.
$PATH looks just ready to explode post-earnings 🚀
But it doesn't fit my system. This is the hardest part in the markets: turning your back on some massive setups because it doesn't check all boxes.
But systems are here to protect us from ourselves. tweet
Offshore
Photo
App Economy Insights
$CRM Salesforce has renamed all of its segments with the word 'Agentforce' in front.
At this point, they might as well change the company name and ticker. https://t.co/AuehniMAKb
tweet
$CRM Salesforce has renamed all of its segments with the word 'Agentforce' in front.
At this point, they might as well change the company name and ticker. https://t.co/AuehniMAKb
tweet
Offshore
Photo
App Economy Insights
$CRM 'Platform, Slack & Other' is slowly becoming the largest segment at Salesforce. https://t.co/Ct8QQj6ObZ
tweet
$CRM 'Platform, Slack & Other' is slowly becoming the largest segment at Salesforce. https://t.co/Ct8QQj6ObZ
tweet
Offshore
Photo
EndGame Macro
Data Centers Are Growing Faster Than the Grid Can Keep Up
This is the realization that the original forecasts were way too conservative. The dashed line was the old BloombergNEF estimate from early 2025. The solid bars are the new one, and the gap between them is the story.
Data center power demand doesn’t rise a little. It ramps like a new industrial sector being built in real time. By the mid 2030s, we’re talking about power needs that rival the output of dozens of major power plants. And this isn’t like air conditioning or heating, where usage spikes and drops. Data centers are more like factories that barely sleep…high, steady, always on load.
That means the grid can’t just produce more energy over a year. It needs to deliver instantaneous power reliably, 24/7, to highly concentrated clusters. That’s the kind of demand that strains transformers, substations, transmission lines, and local generation, not just total supply.
Why PJM is the punchline
Zerohedge’s “Good luck PJM” line isn’t random. PJM covers Northern Virginia, basically the epicenter of global server infrastructure. Once a region becomes the hub for data centers, everything reinforces itself: latency advantages, fiber density, construction talent, existing supply chains.
But data centers can be built fast. Transmission upgrades can’t. New substations, transformer capacity, new gas lines, and regional planning take years. In some cases, nearly a decade. That gap is the real issue. It’s not that PJM can’t power these things, it’s that the load is growing faster than the infrastructure that feeds it.
When you get that mismatch, the pressure doesn’t show up evenly. It shows up as interconnection queues stretching into the 2030s, local bottlenecks, and political fights about who pays for upgrades and who gets priority access.
My View
The problem isn’t that AI will break the grid. The problem is that AI growth moves at tech speed and the grid moves at utility speed. The first stress points won’t be rolling blackouts, they’ll be permitting delays, transformer shortages, and regions quietly telling data center developers not here and not yet.
Over time, this demand reshapes the map. New hubs form where power is cheap, easy to build, or politically welcomed, places like ERCOT, MISO, and parts of the Southeast. Meanwhile, regions like PJM will be forced into hard choices about prioritizing reliability for households versus supporting industrial scale cloud demand.
This chart is predicting constraint. And the regions that figure out how to deliver reliable, scalable power the fastest will end up owning the next wave of AI growth.
tweet
Data Centers Are Growing Faster Than the Grid Can Keep Up
This is the realization that the original forecasts were way too conservative. The dashed line was the old BloombergNEF estimate from early 2025. The solid bars are the new one, and the gap between them is the story.
Data center power demand doesn’t rise a little. It ramps like a new industrial sector being built in real time. By the mid 2030s, we’re talking about power needs that rival the output of dozens of major power plants. And this isn’t like air conditioning or heating, where usage spikes and drops. Data centers are more like factories that barely sleep…high, steady, always on load.
That means the grid can’t just produce more energy over a year. It needs to deliver instantaneous power reliably, 24/7, to highly concentrated clusters. That’s the kind of demand that strains transformers, substations, transmission lines, and local generation, not just total supply.
Why PJM is the punchline
Zerohedge’s “Good luck PJM” line isn’t random. PJM covers Northern Virginia, basically the epicenter of global server infrastructure. Once a region becomes the hub for data centers, everything reinforces itself: latency advantages, fiber density, construction talent, existing supply chains.
But data centers can be built fast. Transmission upgrades can’t. New substations, transformer capacity, new gas lines, and regional planning take years. In some cases, nearly a decade. That gap is the real issue. It’s not that PJM can’t power these things, it’s that the load is growing faster than the infrastructure that feeds it.
When you get that mismatch, the pressure doesn’t show up evenly. It shows up as interconnection queues stretching into the 2030s, local bottlenecks, and political fights about who pays for upgrades and who gets priority access.
My View
The problem isn’t that AI will break the grid. The problem is that AI growth moves at tech speed and the grid moves at utility speed. The first stress points won’t be rolling blackouts, they’ll be permitting delays, transformer shortages, and regions quietly telling data center developers not here and not yet.
Over time, this demand reshapes the map. New hubs form where power is cheap, easy to build, or politically welcomed, places like ERCOT, MISO, and parts of the Southeast. Meanwhile, regions like PJM will be forced into hard choices about prioritizing reliability for households versus supporting industrial scale cloud demand.
This chart is predicting constraint. And the regions that figure out how to deliver reliable, scalable power the fastest will end up owning the next wave of AI growth.
Good luck PJM https://t.co/3I0nN7Hh85 tweet