3/20/25 :: VROCKSTAR :: 
NVDA
Patient, but path to $10 tn
- as we reflect on this crack tape, i'd like to share a few thoughts here, hopefully keeping it a bit punchy and we can duke it out/ debate in the comments.
- objectively, the level of HPC/ AI compute is accelerating up the S curve.
- nvidia's customers are the most price inelastic buyers you could want in a recession. microsoft. tesla. amazon. nation states *cough*. these aren't BNPL customers ordering burritos and a side of GPU chips thru doordash.
- NVIDIA's platform won. such that they don't screw the pooch and their 35k engineers all focused on a singular issue don't decide to go solve some unrelated market (/sarc bc we know Jensen's style), effectively there's no "rebuilding" NVIDIA. ASICs will eventually come, but they won't dominate. In the chip business, there's a reason each vertical tends toward 70... 80... 90% market share players.
- the real issue here, when i run the math (and i've built out my own model tn to wrap my head around this all), is really *where* we are on this S-curve. the mkt is concerned about a few things, and actually, the macro/ risk assets and long-duration i.e. discount rate seem to be more important than fundamentals. that's a good thing, BUT, in the short-term it can really dislocate price. and price tells a story. and that story can distract. it can avert your attention. it can make you nervous (on the converse, fomo). so it's good to have a clear idea of what's going on here, which is why i underwent this exercise.
- all-else equal, as beats, communication and sector dominance remain (and they don't need to be massive beats, just not misses that portend lower growth in the terminal), my estimates put NVDA's mkt cap close to 10 tn.
- but at shy of $3 tn today, that 3x LIKELY will take 2-3 years to play out and will largely be driven by the short term climb of the S-curve, and more immediately driven by, again, terminal rates.
- fair value today ex-beats but with lower terminal rates likely takes the stock toward $5 tn (and i'd expect this to be a reasonably year-end target) or a stock of $200. let's call that move 2/3 "macro" and 1/3 "fundamental"
- and the remainder of the move toward doubling likely happens over the following 2 years because more data will need to be collected by the market to assess this dominance, cash generation etc. etc.
- okay this isn't a punchy write up... at this pt. excuse me!
- so what's the downside? again, there's a lot we can and should duke out in comments to shorten the conclusion here, but i'd suggest something closer to $2 tn for a variety of reasons. that's nearly 30% downside or a stock in the $80s. do we get there? no clue. but in this environment, we've seen how deepseek headlines, blackwell overheating rumors (which btw remain), asic announcements, chinese "competition" etc. etc. all affect the bid. and i'd suggest that a 30% downside for a potential 70% upside into year-end remains a great risk-reward here at $120 today.
- my guess would be that long-term buyers accumulate at these levels and we probably get taken closer to the $130s... even $140s before this becomes more of a complicated equation.
- that being said, it's a clear buy, IMO, at this price, in a YE context and especially in a multi-year context given downside to upside potential and the work i've put in here. truly a one-of-a-kind asset that has actually held it's own against BTC in the last 10 years (THE ONLY of any real market cap)
TL;DR
- still a great buy at $120
- downside below $100 and it's obvious. buying that fear, possibly on leverage in the $80s.
- not using leverage ST in this environment
- YE target of $200
- unfortunately more of a macro punching bag ST but fundamentals remain the meat of the 2-3 year move and so far, don't see any flaws.
lmk what u think.
V
NVDAPatient, but path to $10 tn
- as we reflect on this crack tape, i'd like to share a few thoughts here, hopefully keeping it a bit punchy and we can duke it out/ debate in the comments.
- objectively, the level of HPC/ AI compute is accelerating up the S curve.
- nvidia's customers are the most price inelastic buyers you could want in a recession. microsoft. tesla. amazon. nation states *cough*. these aren't BNPL customers ordering burritos and a side of GPU chips thru doordash.
- NVIDIA's platform won. such that they don't screw the pooch and their 35k engineers all focused on a singular issue don't decide to go solve some unrelated market (/sarc bc we know Jensen's style), effectively there's no "rebuilding" NVIDIA. ASICs will eventually come, but they won't dominate. In the chip business, there's a reason each vertical tends toward 70... 80... 90% market share players.
- the real issue here, when i run the math (and i've built out my own model tn to wrap my head around this all), is really *where* we are on this S-curve. the mkt is concerned about a few things, and actually, the macro/ risk assets and long-duration i.e. discount rate seem to be more important than fundamentals. that's a good thing, BUT, in the short-term it can really dislocate price. and price tells a story. and that story can distract. it can avert your attention. it can make you nervous (on the converse, fomo). so it's good to have a clear idea of what's going on here, which is why i underwent this exercise.
- all-else equal, as beats, communication and sector dominance remain (and they don't need to be massive beats, just not misses that portend lower growth in the terminal), my estimates put NVDA's mkt cap close to 10 tn.
- but at shy of $3 tn today, that 3x LIKELY will take 2-3 years to play out and will largely be driven by the short term climb of the S-curve, and more immediately driven by, again, terminal rates.
- fair value today ex-beats but with lower terminal rates likely takes the stock toward $5 tn (and i'd expect this to be a reasonably year-end target) or a stock of $200. let's call that move 2/3 "macro" and 1/3 "fundamental"
- and the remainder of the move toward doubling likely happens over the following 2 years because more data will need to be collected by the market to assess this dominance, cash generation etc. etc.
- okay this isn't a punchy write up... at this pt. excuse me!
- so what's the downside? again, there's a lot we can and should duke out in comments to shorten the conclusion here, but i'd suggest something closer to $2 tn for a variety of reasons. that's nearly 30% downside or a stock in the $80s. do we get there? no clue. but in this environment, we've seen how deepseek headlines, blackwell overheating rumors (which btw remain), asic announcements, chinese "competition" etc. etc. all affect the bid. and i'd suggest that a 30% downside for a potential 70% upside into year-end remains a great risk-reward here at $120 today.
- my guess would be that long-term buyers accumulate at these levels and we probably get taken closer to the $130s... even $140s before this becomes more of a complicated equation.
- that being said, it's a clear buy, IMO, at this price, in a YE context and especially in a multi-year context given downside to upside potential and the work i've put in here. truly a one-of-a-kind asset that has actually held it's own against BTC in the last 10 years (THE ONLY of any real market cap)
TL;DR
- still a great buy at $120
- downside below $100 and it's obvious. buying that fear, possibly on leverage in the $80s.
- not using leverage ST in this environment
- YE target of $200
- unfortunately more of a macro punching bag ST but fundamentals remain the meat of the 2-3 year move and so far, don't see any flaws.
lmk what u think.
V
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.