$BABA @ $160 in pre-market
Who would have thought that discouraging speculative pump-and-dumps in the market could redirect flows into a scarce and reasonably valued asset that is *also* riding a generational tailwind, pushing it to multi-year highs?
Who would have thought that discouraging speculative pump-and-dumps in the market could redirect flows into a scarce and reasonably valued asset that is *also* riding a generational tailwind, pushing it to multi-year highs?
Pivot Point Capital
Still pressing the short on $CRCL too. This short is harder to manage as the shareholder base doesn’t care about earnings (hence why it’s not a large short position for me), but fundamentals-wise I think it’s a definite short. Just need to not get blown out…
Added to $CRCL short $127 - $130
Besides the other threats to the core biz mentioned above, if we’re going into a rate cut cycle, this feels like a natural hedge
Besides the other threats to the core biz mentioned above, if we’re going into a rate cut cycle, this feels like a natural hedge
Pivot Point Capital
Re-added back to $U with the Jackson Hole dip @ $35.3-$35.7 On second thoughts, the bar is set very low for $U now vs someone more established like $APP. Given that we’re in the early phases of Vector (the beta stage with restricted access for the masses)…
Trimmed 15% of $U @ $47, +31% from entry
Pivot Point Capital
Jeffries raised $RDDT PT to $300, driving the name to new ATHs @ $257, +75% from our initial entry @ $147 “Sustainability of Rev growth is the key debate following an acceleration to multiyr highs and a recent stabilization in DAU trends. Our analysis of…
Trimmed 10% of $RDDT @ $264, +80% from initial entry
Pivot Point Capital
Longed $CRWV and $APLD @ $101.31 & $14.95 respectively, as a proxy bet on AI infrastructure services, driven by $NBIS <> $MSFT $18bn deal announcement. $ORCL strong revenue projections last night also validates the demand for AI infrastructure. CRWV has…
Trimmed 15% of $APLD @ $19.6, +31% from entry
Interesting OpenAI chart showing what customers use ChatGPT for. Seems like 70-80% of these tasks could just as well be handled by an intelligent EA.
Reminds me of how early we are in this journey; if AGI is truly achievable, then we've barely begun to tap into what AI can unlock.
It's very exciting to imagine that the boundaries of human knowledge could expand at a cadence we’ve never seen before, over the next few decades.
Think of the medical breakthroughs waiting to be discovered, the uncharted territories in space and sea ready to be explored, the new school of thoughts yet to be written, even new planets we may one day call home.
There's no better time to dream a little bigger, to adopt a Panglossian outlook on the future.
Reminds me of how early we are in this journey; if AGI is truly achievable, then we've barely begun to tap into what AI can unlock.
It's very exciting to imagine that the boundaries of human knowledge could expand at a cadence we’ve never seen before, over the next few decades.
Think of the medical breakthroughs waiting to be discovered, the uncharted territories in space and sea ready to be explored, the new school of thoughts yet to be written, even new planets we may one day call home.
There's no better time to dream a little bigger, to adopt a Panglossian outlook on the future.
❤1
Pivot Point Capital
Added to $CRCL short $127 - $130 Besides the other threats to the core biz mentioned above, if we’re going into a rate cut cycle, this feels like a natural hedge
Kind of poetic that one of my favourite longs is putting the squeeze on two of my favourite shorts 😮💨
$GOOG $COIN $CRCL
$GOOG $COIN $CRCL
Started new long in $DLO @ $14.87. Short pitch below
DLO acts as a payment bridge between global merchants and consumers in emerging markets, unifying all local payment complexities and methods into a single API and platform for pay-ins and pay-outs.
Their value-add lies in simplifying the fragmented payments industry of emerging markets — the dozens of currencies, payment methods (vouchers, digital wallets, bank transfers, BNPL, cards), and varying regulatory/tax compliance requirements. Given the fragmentation of these markets, I believe the switching cost for global merchants like MSFT, NFLX, and Shein, once they engage DLO, is very high.
DLO is led by CEO Pedro Arnt, who was previously the CFO of $MELI. During his 13 years as CFO, MELI grew from ~$3b to $75b in market cap. This helps in my conviction.
The main bear case is that DLO has seen significant take rate compression in recent quarters. Bears view this as evidence of DLO’s lack of moat, as well as a source of operating deleverage.
However, my view is that the take rate compression is natural when the business scales, as DLO provides volume discounts for higher TPV processed. As long as TPV growth outpaces take rate compression, net gross profit dollars will increase meaningfully even if gross margin compresses.
Moreover, given that this is an extremely asset-light business, the natural operating leverage inherent in the model may cushion or even neutralise the effects of take rate compression if gross profit dollar scales faster that expenses.
Management is guiding for midpoint 45% TPV growth in FY25e. While they have been capturing share in LATAM, their current wallet share is still only ~4–6% of LATAM and AMEA e-commerce TAM. Hence, I’m not too worried about TPV growth slowing down.
Valuation also looks reasonable (something that’s hard to find in this environment!), with FY26e P/FCF standing around 18-22x for a 25-30% growth compounder.
I like the stock, and think the r/r is higher than $SE — so moved 75% of my $SE size over; I don’t want too many correlated bets in emerging markets.
DLO acts as a payment bridge between global merchants and consumers in emerging markets, unifying all local payment complexities and methods into a single API and platform for pay-ins and pay-outs.
Their value-add lies in simplifying the fragmented payments industry of emerging markets — the dozens of currencies, payment methods (vouchers, digital wallets, bank transfers, BNPL, cards), and varying regulatory/tax compliance requirements. Given the fragmentation of these markets, I believe the switching cost for global merchants like MSFT, NFLX, and Shein, once they engage DLO, is very high.
DLO is led by CEO Pedro Arnt, who was previously the CFO of $MELI. During his 13 years as CFO, MELI grew from ~$3b to $75b in market cap. This helps in my conviction.
The main bear case is that DLO has seen significant take rate compression in recent quarters. Bears view this as evidence of DLO’s lack of moat, as well as a source of operating deleverage.
However, my view is that the take rate compression is natural when the business scales, as DLO provides volume discounts for higher TPV processed. As long as TPV growth outpaces take rate compression, net gross profit dollars will increase meaningfully even if gross margin compresses.
Moreover, given that this is an extremely asset-light business, the natural operating leverage inherent in the model may cushion or even neutralise the effects of take rate compression if gross profit dollar scales faster that expenses.
Management is guiding for midpoint 45% TPV growth in FY25e. While they have been capturing share in LATAM, their current wallet share is still only ~4–6% of LATAM and AMEA e-commerce TAM. Hence, I’m not too worried about TPV growth slowing down.
Valuation also looks reasonable (something that’s hard to find in this environment!), with FY26e P/FCF standing around 18-22x for a 25-30% growth compounder.
I like the stock, and think the r/r is higher than $SE — so moved 75% of my $SE size over; I don’t want too many correlated bets in emerging markets.
Closed $COIN short to size up $CRCL short again @ $142.5
Not back to full position yet as conserving ammo to ladder up if needed; wary of a short squeeze as it seems like IBKR is out of borrowable shares for $CRCL. Some of my short orders are still pending
Not back to full position yet as conserving ammo to ladder up if needed; wary of a short squeeze as it seems like IBKR is out of borrowable shares for $CRCL. Some of my short orders are still pending
Pivot Point Capital
Kind of poetic that one of my favourite longs is putting the squeeze on two of my favourite shorts 😮💨 $GOOG $COIN $CRCL
The poet has scripted another wild turn for the portfolio. One of my favourite longs is lifting another of my favourite longs — all is forgiven now $GOOG!
Interestingly, the market seems to view GOOG as a kingmaker now; maybe there should then be value ascribed to that
Interestingly, the market seems to view GOOG as a kingmaker now; maybe there should then be value ascribed to that
Pivot Point Capital
Trimmed 10% of $RDDT @ $264, +80% from initial entry
“Unlike its earlier $60M licensing deal, Reddit is pushing for deeper integration and dynamic pricing tied to how essential its data becomes for AI models.”
On a serious note, the news appears positively incremental in three ways: (1) this deal looks much more evolved than its previous flat-fee licensing agreements. “Dynamic pricing based on how essential the data is..” seems particularly incremental when Reddit is the most quoted source for LLMs and makes up almost 40% of LLM content. (2) It revives the AI content/data licensing for the 99% margin narrative (potentially with better pricing!) which had faded amongst investors last quarter when there was no real pickup in licensing revenue. If you revisit our $RDDT post-mortem coverage, that was our small nitpick with the otherwise stellar Q2 results, though we were ultimately fine with it as we expected licensing contracts to be lumpy rather than linear. (3) A large part of Reddit’s bear case was its heavy dependence on Google for traffic, which could decrease due to the rollout of AI summaries and AI mode, along with the gradual decline of search traffic. If this evolved deal/partnership materialises, Reddit could potentially secure an agreement with Google for deeper integration into Google’s AI products, thereby prioritising traffic to Reddit over other sites and also creating a longer-term flywheel for AI content (just my speculation).
On a serious note, the news appears positively incremental in three ways: (1) this deal looks much more evolved than its previous flat-fee licensing agreements. “Dynamic pricing based on how essential the data is..” seems particularly incremental when Reddit is the most quoted source for LLMs and makes up almost 40% of LLM content. (2) It revives the AI content/data licensing for the 99% margin narrative (potentially with better pricing!) which had faded amongst investors last quarter when there was no real pickup in licensing revenue. If you revisit our $RDDT post-mortem coverage, that was our small nitpick with the otherwise stellar Q2 results, though we were ultimately fine with it as we expected licensing contracts to be lumpy rather than linear. (3) A large part of Reddit’s bear case was its heavy dependence on Google for traffic, which could decrease due to the rollout of AI summaries and AI mode, along with the gradual decline of search traffic. If this evolved deal/partnership materialises, Reddit could potentially secure an agreement with Google for deeper integration into Google’s AI products, thereby prioritising traffic to Reddit over other sites and also creating a longer-term flywheel for AI content (just my speculation).
Putting All Your Eggs In One Basket, And Then Watching That Basket Very Carefully
Something I wish I had learned more about before leaving an institution to manage my own capital is execution —specifically sizing.
Hence, I’m tracking my trade execution improvement by analyzing my slugging ratio and batting average.
Slugging is how much larger your winners are versus your losers; batting is your average win rate. Simple concepts, but it helps me to compartmentalise the learning process.
Basically, when you have a winning idea, do you have the conviction to load the boat? If not, it doesn't matter how thorough your research is or how many experts you consult — if your best trade only nets your entire book an inconsequential gain. Similarly, if a trade turns against you, can you recognize it quickly and take a minimal loss, or do you cling to your champion idea with mental gymnastics?
Not all ideas will be winners, and that is ok! What matters is having the humility to tell the difference, knowing when an idea needs more work, and not over-sizing in those cases.
Over the past year, my batting average has been poor due to the macro volatility, which caused my mental capital to suffer (e.g., cutting would-be winners too soon just because they turned red due to lack of conviction, or spraying over-sized bets on loose ideas).
My slugging ratio was my saving grace, allowing me to generate a positive YTD% even with <50% win-rate. Yet, attempting to raise my win rate would likely mean sacrificing my slugging ratio — by taking profits too early to secure frequent but smaller wins.
So would I trade my slugging for a better batting? Ultimately it depends on what type of investor you are; personally I prefer home runs and am willing to sacrifice consistency, hence probably not.
This is also why winning trades in mega-caps like BABA and GOOG are so attractive to me. These mega-cap trades are easier to research and more correlated to market benchmarks (boosting batting), while being more predictable and less volatile (improving slugging by allowing you to easily ride the trend) — all without having to worry about liquidity constraints or spreads. You can have your cake and eat it too with these trades!
Sharing as I think too many people focus excessively on batting average and pay too little attention to slugging ratio, when the reverse should be true.
So that our winners really count, and losers only graze us.
Something I wish I had learned more about before leaving an institution to manage my own capital is execution —specifically sizing.
Hence, I’m tracking my trade execution improvement by analyzing my slugging ratio and batting average.
Slugging is how much larger your winners are versus your losers; batting is your average win rate. Simple concepts, but it helps me to compartmentalise the learning process.
Basically, when you have a winning idea, do you have the conviction to load the boat? If not, it doesn't matter how thorough your research is or how many experts you consult — if your best trade only nets your entire book an inconsequential gain. Similarly, if a trade turns against you, can you recognize it quickly and take a minimal loss, or do you cling to your champion idea with mental gymnastics?
Not all ideas will be winners, and that is ok! What matters is having the humility to tell the difference, knowing when an idea needs more work, and not over-sizing in those cases.
Over the past year, my batting average has been poor due to the macro volatility, which caused my mental capital to suffer (e.g., cutting would-be winners too soon just because they turned red due to lack of conviction, or spraying over-sized bets on loose ideas).
My slugging ratio was my saving grace, allowing me to generate a positive YTD% even with <50% win-rate. Yet, attempting to raise my win rate would likely mean sacrificing my slugging ratio — by taking profits too early to secure frequent but smaller wins.
So would I trade my slugging for a better batting? Ultimately it depends on what type of investor you are; personally I prefer home runs and am willing to sacrifice consistency, hence probably not.
This is also why winning trades in mega-caps like BABA and GOOG are so attractive to me. These mega-cap trades are easier to research and more correlated to market benchmarks (boosting batting), while being more predictable and less volatile (improving slugging by allowing you to easily ride the trend) — all without having to worry about liquidity constraints or spreads. You can have your cake and eat it too with these trades!
Sharing as I think too many people focus excessively on batting average and pay too little attention to slugging ratio, when the reverse should be true.
So that our winners really count, and losers only graze us.
👍8🔥4
Pivot Point Capital
Confirmed that USG is taking a 10% stake in Intel. It’s not new money but a conversion of previous awarded but undistributed grants. I can see an $ASML situation where customers come together to invest and make $INTC workable. Whats a billy$ to NVDA, to…
GG. $INTC +32% to $33 in pre-market, breaking out of its year-long $18-$25 range.
What’s a billy$ to NVDA? Now what’s a billy$ to AAPL? To AMD? To Broadcom? To Qualcomm? Nothing.
What’s a billy$ vote of confidence to $INTC..? Everything. The flywheel has started to turn..
What’s a billy$ to NVDA? Now what’s a billy$ to AAPL? To AMD? To Broadcom? To Qualcomm? Nothing.
What’s a billy$ vote of confidence to $INTC..? Everything. The flywheel has started to turn..
🔥6
Pivot Point Capital
GG. $INTC +32% to $33 in pre-market, breaking out of its year-long $18-$25 range. What’s a billy$ to NVDA? Now what’s a billy$ to AAPL? To AMD? To Broadcom? To Qualcomm? Nothing. What’s a billy$ vote of confidence to $INTC..? Everything. The flywheel has…
Added more $INTC long @ $29.57 - $32
Pivot Point Capital
Added more $INTC long @ $29.57 - $32
Listened to the $INTC <> $NVDA pressor live yesterday; penning down my thoughts. Price retraced slightly after Jensen emphasised that the announcement was about the product collaboration and not Intel’s foundry. I’m still bullish — I think there’s too much focus on the foundry angle (for now at least!) and too little on Nvidia’s $5bn equity stake. Not going to dive into product collab specs since there are already plenty of good takes out there. To me, that isn’t the main point anyway.
1) Analysts obsessed over the foundry question — three consecutive rephrased qns in a row on whether NVDA would use INTC’s foundry, despite both CEOs repeatedly dodging. Of course there were no answers: 1) the 14A fab isn’t even built, so how can Nvidia commit without seeing PDKs and yields? 2) Both Nvidia and Intel are still TSMC customers. Publicly implying they’re moving to Intel could risk angering TSMC, which controls supply at the leading edge. Hence, until Intel can show both reliable yield at scale and a node cadence on par with TSMC (harder to prove), I don’t expect any customers to commit publicly. This reflects the chokehold TSMC has on everything AI and modern electronics. It also highlights why Intel’s revival is so strategically important.
2) The real under-discussed piece is Nvidia’s $5bn equity stake. Jensen brushed it off with corporate-speak about being excited for the product collab, but I don’t see Nvidia freely handing out equity investments to its hundreds of partners. Clearly not a passing thought.
3) Jensen said the product collab has been a year in the making, so it started under Gelsinger. I can believe that the joint SoC work began then, but was the equity stake also in the works one year ago? Or was it more recent, spurred by the Trump administration or recent events?
4) Worth noting, Jensen seemed particularly excited to tell Lutnick about the investment, and Lutnick emphasized being happy that American companies are working together. Which do you think Lutnick appreciates more — the product collab, or Nvidia’s $5bn equity infusion? Even if the Trump admin wasn’t directly involved, the stake feels positioned to curry favor (while also helping Nvidia secure production capacity if 14A actually hits).
5) ASML, SNPS, and other equipment makers’ share prices after the announcement show that the market is clearly assigning a higher probability to 14A happening now, even with both CEOs dodging the foundry narrative live.
6) Do I think Intel Foundry will succeed? I honestly don’t know. Intel is WAY behind; even Jensen described TSMC’s tech as “like magic.” But the stakes are REALLY high. Should all of AI be dependent on a monopoly supplier? If not, what other alternatives do we really have, besides shooting a shot at INTC..? Samsung Foundry is also suffering from severe yield issues, and have fallen far behind TSMC. Also, if there’s an operator that can successfully execute this turnaround, LBT is probably it.
Valuation is reasonable. As previously shown with napkin maths, Intel’s fabless segment generated about $13bn EBIT in 2024. Our previous expectation was for EBIT to trend down given AMD’s competition, but with this NVDA collab they might be able to at least defend share. Keeping EBIT at ~$13bn and slapping on a 12x multiple gives $156bn EV. Back out ~$37bn net debt/minorities = $119bn EQV, or ~$27.5/share. At $30.5 today, that means ~$3/share is what you’re paying for the foundry optionality. For comparison, TSMC’s market cap is ~$1.78tn. If Intel can catch up to 5% of TSMC share, that’s ~$89bn of optionality value, or ~$20.5/share. Combining Fabless + Foundry = ~$47.5/share. Simple 80/20 framework, but you get the idea. You can then tweak assumptions however you like to see if the price makes sense.
1) Analysts obsessed over the foundry question — three consecutive rephrased qns in a row on whether NVDA would use INTC’s foundry, despite both CEOs repeatedly dodging. Of course there were no answers: 1) the 14A fab isn’t even built, so how can Nvidia commit without seeing PDKs and yields? 2) Both Nvidia and Intel are still TSMC customers. Publicly implying they’re moving to Intel could risk angering TSMC, which controls supply at the leading edge. Hence, until Intel can show both reliable yield at scale and a node cadence on par with TSMC (harder to prove), I don’t expect any customers to commit publicly. This reflects the chokehold TSMC has on everything AI and modern electronics. It also highlights why Intel’s revival is so strategically important.
2) The real under-discussed piece is Nvidia’s $5bn equity stake. Jensen brushed it off with corporate-speak about being excited for the product collab, but I don’t see Nvidia freely handing out equity investments to its hundreds of partners. Clearly not a passing thought.
3) Jensen said the product collab has been a year in the making, so it started under Gelsinger. I can believe that the joint SoC work began then, but was the equity stake also in the works one year ago? Or was it more recent, spurred by the Trump administration or recent events?
4) Worth noting, Jensen seemed particularly excited to tell Lutnick about the investment, and Lutnick emphasized being happy that American companies are working together. Which do you think Lutnick appreciates more — the product collab, or Nvidia’s $5bn equity infusion? Even if the Trump admin wasn’t directly involved, the stake feels positioned to curry favor (while also helping Nvidia secure production capacity if 14A actually hits).
5) ASML, SNPS, and other equipment makers’ share prices after the announcement show that the market is clearly assigning a higher probability to 14A happening now, even with both CEOs dodging the foundry narrative live.
6) Do I think Intel Foundry will succeed? I honestly don’t know. Intel is WAY behind; even Jensen described TSMC’s tech as “like magic.” But the stakes are REALLY high. Should all of AI be dependent on a monopoly supplier? If not, what other alternatives do we really have, besides shooting a shot at INTC..? Samsung Foundry is also suffering from severe yield issues, and have fallen far behind TSMC. Also, if there’s an operator that can successfully execute this turnaround, LBT is probably it.
Valuation is reasonable. As previously shown with napkin maths, Intel’s fabless segment generated about $13bn EBIT in 2024. Our previous expectation was for EBIT to trend down given AMD’s competition, but with this NVDA collab they might be able to at least defend share. Keeping EBIT at ~$13bn and slapping on a 12x multiple gives $156bn EV. Back out ~$37bn net debt/minorities = $119bn EQV, or ~$27.5/share. At $30.5 today, that means ~$3/share is what you’re paying for the foundry optionality. For comparison, TSMC’s market cap is ~$1.78tn. If Intel can catch up to 5% of TSMC share, that’s ~$89bn of optionality value, or ~$20.5/share. Combining Fabless + Foundry = ~$47.5/share. Simple 80/20 framework, but you get the idea. You can then tweak assumptions however you like to see if the price makes sense.
❤3👍1
Pivot Point Capital
Added more $INTC long @ $29.57 - $32
7) This event reminds me of Microsoft’s $150mn vote of confidence in AAPL in 1997. Apple was on the brink of collapse, and MSFT stepped in with capital and Office/IE integration. It was a lifeline for AAPL, and insurance for MSFT: keep the Mac alive for Microsoft’s software while also benefiting if Apple staged a comeback. Jobs had just returned as Apple’s CEO then, just like LBT has recently stepped into Intel’s CEO role. Nvidia’s move looks similar — guarantee GPU compatibility for enterprises that do not want to rework their legacy x86 infrastructure into ARM-based, while also securing upside in case Intel’s foundry revival actually succeeds. Back then, was Apple’s comeback obvious? Not at all. Is Intel’s turnaround obvious now? Also no. But history shows these bets sometimes pay off, especially in tech supply chains.
8) I’m all for taking profits, but I see euphoria from INTC bagholders rushing to exit yesterday for a 30–50% gain, right after the name broke out from a year-long range. Maybe it’s reflective of the pain Intel shareholders have endured over the past decade (condolences!), but isn’t the risk/reward off if you’re only playing the name for a 50% gain? Especially when there are plenty of other stocks that have done far better with far fewer uncertainties in the past year? Personally, I treat INTC as a binary long-term bet that rarely presents itself. Either it’s multiplying, or it’s back at $20; sizing appropriately for both outcomes should be key.
8) I’m all for taking profits, but I see euphoria from INTC bagholders rushing to exit yesterday for a 30–50% gain, right after the name broke out from a year-long range. Maybe it’s reflective of the pain Intel shareholders have endured over the past decade (condolences!), but isn’t the risk/reward off if you’re only playing the name for a 50% gain? Especially when there are plenty of other stocks that have done far better with far fewer uncertainties in the past year? Personally, I treat INTC as a binary long-term bet that rarely presents itself. Either it’s multiplying, or it’s back at $20; sizing appropriately for both outcomes should be key.
❤1
Pivot Point Capital
Confirmed that USG is taking a 10% stake in Intel. It’s not new money but a conversion of previous awarded but undistributed grants. I can see an $ASML situation where customers come together to invest and make $INTC workable. Whats a billy$ to NVDA, to…
"We know Trump would love to brag about how his “$10b investment deal” in INTC produced a 5x ROI for the U.S. taxpayers."
0.5x ROI, so far.
0.5x ROI, so far.
🤪1
Pivot Point Capital
“Unlike its earlier $60M licensing deal, Reddit is pushing for deeper integration and dynamic pricing tied to how essential its data becomes for AI models.” On a serious note, the news appears positively incremental in three ways: (1) this deal looks much…
Trimmed another 15% of $RDDT @ $275 yesterday, +87% gain from initial entry
Trimming as that was my original TP before the AI content deal w Google. Reddit was also my largest position, so feels apt to trim down the exposure after such a rally.
Trimming as that was my original TP before the AI content deal w Google. Reddit was also my largest position, so feels apt to trim down the exposure after such a rally.