Frank Ronson🧠

@CedarStResearch And it didn’t even address the reserves either

GPT created a DCF model that produced an >11,000% return for $UBER don’t panic just yet

@dcfgod Crypto figuring out why fiduciary duties and conflict of interest restrictions were invented

@RyeNotBerben Why didn’t you keep funding it?

I had this tiny roth ira i funded wen they first started (never made another contribution unfortunately) - first 16 yrs was about flat while spy up 160%. I decided to put on front burner and see if i could really compound it. AMA.

@dcfgod This is actually one of the true cases where crypto is better — Heritage fees are nuts. Buyer & seller both going to lose vs whatever platform you seeded

@CedarStResearch Yeah, I was trying to throw people off so they don’t know it’s $XZO

@PoGoFrankR I can assure you it’s not a mining company lol

Not a bad day. My core trading account, but not all my direct holdings as I use two other platforms. Name blacked out is a company I haven't commented on yet, so going to keep it blanked out right now. $MAX $WRB $ORI $PATH $AIG $GOOG $ARX $... $HG $PLMR $HCI $AJG $TSM $BETR $ASIC by position size

@compound248 “They” said the same thing in in the early 2010s too because of ZIRP. I look forward to another decade of 15%/year

have fun staying poor Apollo casually predicting ZERO returns for the S&P 500 over the coming DECADE.

@Icebergy It’s the only token where there’s someone buying

why are people bullish eth all of a sudden?

@Floebertus “Big 4 Auditor” is like saying “Chase is our commercial banker”. Meaningless from a risk perspective, just means they pay a lot of fees

Intellego... I saw the warning signs, and initially didn't trust them. Mistake is that I changed my opinion and put in 1% because: - Henkel collaboration - Large insider buys - Massive guidance - Big 4 auditor Lesson: Don't take insider buy/sells too serious...

We received calls from shareholders livid at @BaillieGifford after learning they dumped 1/3rd of their SpaceX shares held in EWI and USA two months ago at a level massively lower than yesterday’s announcement of a planned $1.5T IPO. Baillie put their interests ahead of their clients yet again. They calculated they needed to sell to keep the private holdings percentage from going up in their ill fated EWI/USA merger. They didn’t consult their largest investor by far and likely didn’t consult ANY investor before this terrible sale of their crown jewel. I was on record saying we would keep SpaceX. Is the UK press going to finally hold them accountable?

you can deadass just VPN to the UK and report people for 'UK Illegal Content' us americans have the opportunity to do the funniest thing

@EricBalchunas Isn’t this result also present in stocks? Have a feeling this is more and accounting quirk than some massively unexploited alpha.

We looked at this last year and found most of the gains are in fact after hours. Doesn't mean the ETFs aren't having impact. Some of this is positioning bc of the ETFs etc or derivatives based on flows etc etc. But yeah, bitcoin After Dark ETF could put up better reutrns, we'll see tho.

BITCOIN AFTER DARK: new filing for an ETF that will only hold bitcoin at night, buying it when the US market closes and selling it when it opens.

@Johnisatoolfan @Blinklebloop Lmao, I’m not paying $240 to read a blog post from some random South African dude about some Japanese microcap who only got verified 2 weeks ago. You gain an audience by proving value. So far he’s done nothing but breach the X ToS.

@PoGoFrankR @Blinklebloop Yea he should give up all his months of work for free. That makes business-sense. Your tweet is essentially saying “I’m a cheapskate leech”

Here it is, by far the best risk-reward I have ever written up (in my 10 year + career). - Near monopoly in its niche. - EV/EBIT <10 - AI Expose and growing its ~50% operating margin segment fast. - Clear room for more growth (and busy growing double digits), capex spending situation close to TSMC gigafabs. The best part is that its undervalued for obvious reasons! There is a clear misunderstanding of this business. It’s got a moat that would make Buffett drool but unfortunately he wouldn’t buy because it’s too small (which is another reason it’s undiscovered). It has management that have a firm grasp of governance and capital allocation who have performed extremely well and have earned a exceptional ROIC. link is in my bio

@dcfgod Charizard Strategy (tm)

@kakashiii111 I bought a couple for a new gaming PC so please make sure that’s factored in

To give a little more context ahead of my post next week, here is some rough simple math: Let's assume everything Jensen says is true (including reading thousands of emails a day): Jensen said 6 million Blackwell GPUs have shipped. Since Blackwell is out, Nvidia reported 111B in revenue in GPU datacenters. If you do simple math, 6 million Blackwell GPUs within the reported 111B revenue of datacenters since Blackwell started to ship is not matching, because it represents only between 2.5 to 3.5 million Blackwell chips. Let's try to help Nvidia and add the over "10B in Blackwell revenue" they reported in 2024 Q4. That brings the total to over 121B in revenue. Over 10B represents around 250k to 330k chips, which in the best scenario brings us to almost 4 million Blackwell GPUs. It still does not match. Ok, let's try another calculation. Let's assume again that Jensen is always telling the truth and he means the total Blackwell chips shipped across all segments. That would imply that between 2.5 to 3.5 million Blackwell chips fit into the 111B in revenue, while the remaining 3.5 to 2.5 million are for gaming and other segments. Because we believe Jensen and are trying to find where my math is not matching, I will assume he means all segments, which would be 20 percent of all GPUs shipped, while the other 80 percent go to datacenters. So it means that even if we count Nvidia’s Q4 10B in Blackwell revenue and assume 80 percent goes to datacenters, we still have a gap of 500k to 800k GPUs. It still does not match. Ok, let's try another calculation or a new physics law since Nvidia broke one last year according to Jensen, and he is always telling the truth. Let's assume I am bad at even rough and simple math and that the 6 million shipped Blackwell GPUs are correctly calculated in the numbers. So we have 6 million Blackwell GPUs shipped since January 2025. Roughly 65 to 70 percent of these GPUs go to datacenters in the US. To power 65 to 70 percent of that 6 million GPU fleet (roughly 3.9 to 4.2 million GPUs), you would need approximately 8.5 to 11 gigawatts (GW) of datacenter capacity. (FYI, This amount of power is roughly equivalent to the total electricity generation capacity of Singapore or about ten standard nuclear reactors.) Between 2024 and 2025, the US has built 3.8 to 4.2 GW in 2024 and an estimated 4.5 to 5.0 GW in 2025, which is 8.5 GW combined (Let's ignore that due to power delays, only about half of that is physically delivered within the calendar year). So we have at best 8.5 GW, which is the rough minimum needed to power all the Blackwell chips in the US. The math is matching only if we assume that datacenters with Nvidia chips are the only ones being built in the US (excluding AMD, TPUs, and any other datacenters without Nvidia) and that only Blackwell has been installed in datacenters built in 2024, and that they were waiting for Blackwell chips in 2025 to operate. Now some energy data: According to the Federal Energy Regulatory Commission (FERC), the US installed nearly 26 GW of new generation capacity between January and August 2025, slightly up from the approximately 23 GW installed over the same period last year. From the 26 GW, 3 GW are wind, 3.7 GW natural gas, and 20 GW solar, and 0 GW nuclear. I will believe Jensen, because in this post we are just believing Jensen’s words that in the next 6 to 7 years we are going to see a bunch of small nuclear reactors and that we will all be power generators, just like somebody’s farm. So I will assume he means starting from January 2026, because so far there were deployed 0 GW of nuclear in the US in 2025. So it means that if all datacenters that have been built, deployed, and planned to finish by the end of this year are truly completed on time, then by numbers only (simple math, I do not know the actual allocation) all natural gas built this year has been allocated to datacenter consumption, including about 80 percent of wind or around 25 percent of all solar energy. I hope this post gives some context ahead of my post next week about where all the GPUs that do not find homes in datacenters are going.

@Bonhoeffer_KDS @ReneSellmann Are there some good examples of this that aren't private/single co operations? e.g. Who turned $1m into $100m since 1975? or $10m into $1bn?

@ReneSellmann The real beater is multi-generational wealth compounding @ 10%. If you look @ 50-100 years @ those rates it is incredible. Developing the family discipline to do this is key.

This is why time in the market beats almost everything. 15% doesn’t look spectacular in year 1. It looks ridiculous in year 30.

@orrdavid Thought that started with "Outright Blackmail"

Been doing an hours long chatGPT session about risk management as a prime broker. I really think I could run the best broker in the world.

@ImperiumPaper Wasn’t that like 2 weeks ago? Timeline was a guy who got rekt on “senior tranche” of loans to motorcycle co in Uganda or some shit

I can’t wait for the day this kind of private credit finds its way to the blockchain and the reaction from investors is to scratch their heads and ask “Wtf is DIP? And my unsecured loan is still good, right?”

@TheCryptoNexus “Ad Optimization R&D”: no one blinks an eye “AI”: must be a bubble

I truly do not understand how anyone thinks AI isn’t a massive bubble… listen to the things ppl come up with to justify long term investments into AI *at current vals* Apparently the fact that it all just burns massive piles of money constantly doesn’t matter New paradigm!! Sure it’s not profitable or even close to it now but just watch, in 5 years it’ll be so crazy man! Fascinating