Rocketman

It is August 21, 2025. I have entitled this video, Past Resistance = Future Support. As I purchased many stocks from the yearly lows of 2025, I anticipate, from a probabilistic perspective, that this new retracement will not last, but alas! It shall be a 3% to 5% pullback. I investigate, from a technical perspective, the chart built on the principles and experience I have gained over the years of watching the markets: from the smallest (general) timeframes to the biggest. Stocks have been pent up for months, coiling, building up weekly pressure. It is time to grow. Time to invest. Time to be patient over the course of one or two quarters. Expectation: A Halloween or Thanksgiving rally. Targets will be hit by December 2025 to January 2026.

Technicals: After a daily a-b-c to long-term horizontal trendline support, when we zoom out, we can see Apple's 2009 Long-Term Trendline Bounce. Monthly Chart:

Technicals: When the chart hits just right, something are left best unexplained. Look at this month's chart's last reddish color down price bar in Apple's stock (AAPL). Do you see a collapse in it's stock? Or does it just look like it took a brief pause and giving you a chance to buy more "apples" at a discounted price. 🙂 Monthly: Weekly: Equal legs + 50% fib pullback + cloud support+ uHd Daily: ExDiv1 at equal legs ".....accelerating growth in the money supply is historically correlated with broadening stock performance. As smaller companies have easier and less expensive access to capital, they can invest more in their own growth initiatives. That leads to stronger returns investors typically expect from smaller companies in normal economic environments and more S&P 500 constituents outperforming the overall index. One of the easiest ways to invest in that trend reversal is to buy an equal-weight index fund like the Invesco S&P 500 Equal Weight ETF. The equal-weight index balances every component of the S&P 500 equally. This means the amount you'll invest in the biggest mega-cap stocks is the same as the smallest members of the index. Each quarter, the index's managers rebalance it, and new constituents are added, while others leave." ( fool.com/investing/2025/03/09/us-money-supply-growth-is-accelerating-it-could-si/) I think that is the safest bet for where to set aside and allocate your money every month towards if you don't want to worry about being a market wizard and a savant. However, Apple and Nvidia are still good businesses to invest in.

Tech: Dip to horizontal support + 50% fib + TL Normal drop and expect a rally or sideways market

I share a letter I wrote to family regarding the fundamentals and technicals of Nvidia and why Nvidia continues to be a great buy. Fundamentals and Recent Events: I understand your concerns about my investment in the U.S. stock market, especially with the news circulating about DeepSeek and its potential impact on Nvidia. And I heard your concerns with rumors about Nvidia being in a bubble. Let me share my perspective on why I believe this situation presents a unique opportunity, and why for me, this selloff in Nvidia represents an amazing opportunity to load up more shares, as we take advantage of the confusion in the market. So, we all know the news. Recently, a new little kid on the block called DeepSeek (a recent Chinese AI startup) has emerged and hit the news headlines. It introduced its R1 model, an open-source AI application that delivers advanced reasoning capabilities at a fraction of the typical cost. So, the launch of DeepSeek’s chatbot claims to rival top Western models at a fraction of the training cost. This development This spooked investors led to a significant market reaction and a broader tech selloff, with Nvidia's stock experiencing a notable decline. Cisco's researchers point to the much lower budget of DeepSeek compared to rivals as a potential reason for these failings, saying its cheap development came at a "different cost: safety and security." DeepSeek claims its model took just $6 million to develop, while OpenAI's yet-to-be-released GPT-5 is reported to likely cost $500 million. PC Magazine was founded in 1982, and it is a well-known American technology magazine that provides reviews, news, and analysis of the latest hardware, software, and consumer electronics. According to PCMag, Cisco’s research team managed to "jailbreak" DeepSeek R1 model with a 100% attack success rate, using an automatic jailbreaking algorithm in conjunction with 50 prompts related to cybercrime, misinformation, illegal activities, and general harm. This means that DeepSeek, the new kid on the block, failed to stop a single harmful prompt. And I quote: "DeepSeek stacked up poorly compared to many of its competitors in this regard. OpenAI’s GPT-4o has a 14% success rate at blocking harmful jailbreak attempts, while Google’s Gemini 1.5 Pro sported a 35% success rate. Anthropic’s Claude 3.5 performed the second best out of the entire test group, blocking 64% of the attacks, while the preview version of OpenAI's o1 took the top spot, blocking 74% of attempts." Another problem in addition to this is that DeepSeek has been shown to have strong content restrictions—well, but only when it comes to China-related political content. So, this is not a product that scale or be useful for research and getting accurate information, since the information you can get from it is old history and current information is more or less a mix of state-driven propaganda, limited exposure to peer-reviewed research, and a fragmentary sense of historical objectivity. So, the result is a narrow understanding of complex issues, heavily influenced by government-promoted narratives. This myopic view is detrimental to making accurate investment decisions and for gaining an accurate perspective on the world and global issues. Despite this, Nvidia remains a dominant force in the AI hardware sector. The company's GPUs are integral to AI development, and even was revealed that DeepSeek's R1 model was trained using Nvidia's H800 series chips! So, Nvidia’s market position is in the AI ecosystem. Let us take a look at the recent drop in Nvidia: Nvidia had a sell from the beginning of in January 2025. Barron’s is a prestigious financial publication that focuses on investing, stock market analysis, and financial news. It is owned by Dow Jones & Company, which also publishes The Wall Street Journal. Barron’s has published an article regarding this called "Nvidia Stock Rises. Why DeepSeek AI Worries Are Overblown" by Elsa Ohlen (Jan. 28, 2025, 5:24 AM ET, Nvidia Stock Rises. Why DeepSeek AI Worries Are Overdone. - Barron's). According to this article in Barron’s, the recent selloff in Nvidia's stock, while significant, can be viewed as an overreaction. Analysts suggest that such market movements often present buying opportunities, especially for companies with strong fundamentals like Nvidia. Fears of an AI Slowdown Are Overblown! Some analysts, including Citi’s Christopher Danely, argue that concerns about DeepSeek disrupting Nvidia’s growth are exaggerated. Firstly, DeepSeek’s AI model is not built entirely from scratch. Instead, it improves upon existing AI models by using a technique called 'distillation'. What is meant by distillation? It essentially means taking a powerful AI model and making a smaller, more efficient version of it. However, this process still requires access to large-scale computing power and cloud infrastructure. So, DeepSeek still needs cloud computing and high-end chips; and these services rely heavily on Nvidia’s high-performance chips. AI models, even smaller ones, still require a lot of computing power. And this means that even with DeepSeek’s innovation, the demand for Nvidia’s hardware and Nvidia’s AI infrastructure will continue to grow, not shrink. DeepSeek relies on existing AI models using distillation techniques, meaning major cloud service providers and high-end computing remain essential—a sign that AI spending will continue. In the end, Nvidia benefits despite DeepSeek’s existence. Because it is Nvidia’s chips that power DeepSeek’s infrastructure it needs for AI to run. Tell DeepSeek to just buy for Nvidia chips. They’ll need it. Hahaha! Nvidia has seen sharp drops before. From a chartist’s perspective, the selloff in Nvidia is pretty common for Nvidia from its outset, and it is healthy for the market. From the year 2017 until 2025, Nvidia dropped beyond 18% or 20%; and in some cases beyond 30%, approximately 15 times. Two times, it fell beyond 50% in 2018 and 2022. . In fact, out of the 10 largest single-day losses in stock market history, Nvidia alone is responsible for 8 of them; more than any other company. This shows that Nvidia’s stock has historically been volatile in the short term, experiencing massive drops at times. But these declines have historically been followed by strong recoveries, and continued its long-term growth. This proves that these selloffs are actually great buying opportunities. So, just because Nvidia had major sells offs, those previous major selloffs did not indicate long-term decline, and spell the end of the world. Investors who previously bought during these dips often saw significant gains as Nvidia continued to dominate the AI and semiconductor industries. Companies like Meta and Amazon also saw $200 billion+ market value losses in a single day in 2022. In April 2024, META dropped 21%. My alerts were fired up and I made the purchase. From that time to today, META has risen from 420 to 690. It was a good, prudent, purchase. Many were calling it a bubble at that time. But history repeats itself in these US markets where they continue to innovative, grow stamina, have risk appetite and evolution to solve real problems and generate quality products and services for people. Below is a chart showing all those significant drops in Nvidia with black arrows. Weekly Chart: And based on my own analysis, Nvidia looked great last quarter and forward guidance for 2025 looks great also. Not as lofty, but still great. Here is a chart of what I see for the year 2025. The numbers of the left are estimated earnings for 2025. The numbers on the right are estimated sales for 2025. As long as we meet or exceed these calculations, we have nothing to worry about. I will monitor them as the quarterly come along. Question & Answers Session: Question 1 : "NVIDIA's valuation is only so high because of its crazy profit growth, but computing power is hitting a ceiling." ANSWER: Partially true, but computing power isn’t hitting a ceiling—it's evolving. •NVIDIA’s valuation is high because of its dominant role in AI infrastructure and its massive profit growth. •However, the idea that computing power is "hitting a ceiling" isn’t entirely accurate—AI efficiency is improving, but demand for higher computing power is still growing. •New AI models, like DeepSeek’s, show that training can be done more efficiently, but this doesn’t mean we’ve reached a limit on computing power. Conclusion: Computing power isn’t stopping—it’s shifting toward more efficient AI training and inference models. ________________________________________ Question 2: "This isn’t just about DeepSeek—it’s a question of open-source vs. closed-source AI." Answer: Good point—DeepSeek represents a bigger shift in AI development. •DeepSeek’s open-source model challenges NVIDIA’s dominance, since it reduces reliance on proprietary AI stacks like CUDA and NVIDIA GPUs. •Companies like Microsoft, Google, and Meta may start investing more in open-source models, which could shift the AI landscape over time. •However, DeepSeek alone won’t make this happen overnight—broad adoption of open-source AI and alternative hardware is required, and most AI companies still rely on NVIDIA. 🔹 Conclusion: Open-source AI is an important trend, but it hasn’t displaced NVIDIA yet. ________________________________________ Question 3: "DeepSeek is already working on supporting Huawei’s GPUs, and that’s happening fast. Within 1-2 years, Huawei could take NVIDIA’s spot in China at the very least." ANSWER: That is speculative—Huawei is improving, but replacing NVIDIA in 1-2 years is unlikely. •DeepSeek is optimizing for Huawei GPUs, which is expected given U.S. export restrictions on NVIDIA chips. •However, Huawei’s AI chips (like Ascend) are still behind NVIDIA’s in performance, and catching up will take time. •Past Chinese AI chip development has lagged by 3+ years compared to NVIDIA, and even if Huawei gains ground in China, NVIDIA’s global leadership remains secure for now. 🔹 Conclusion: Huawei may gain market share in China, but it’s not overtaking NVIDIA worldwide anytime soon. ________________________________________ Question 4: "NVIDIA’s insane profit margins won’t last forever." ANSWER: True—but NVIDIA’s margins are still strong for now. •NVIDIA’s high margins come from its dominance in AI chips and proprietary software (CUDA), which makes switching away from NVIDIA difficult. •DeepSeek shows that AI models can be trained more cheaply, which could put long-term pressure on NVIDIA’s pricing. •However, NVIDIA is still critical for AI training and inference, and demand for GPUs is growing, so their margins won’t collapse overnight. Conclusion: Margins may decline in the future, but NVIDIA remains highly profitable in the short term. ________________________________________ Question 5: "The AI hype bubble is something everyone knows about. DeepSeek is just the first to poke a hole in it—there will be plenty more doing the same soon." ANSWER: Overstated—AI spending is still growing, and NVIDIA isn’t collapsing. •Yes, there’s an AI hype cycle, and some stocks may be overvalued. •DeepSeek’s low-cost model surprised investors, but AI investment is still increasing. •Big tech companies (Microsoft, Google, Meta) are doubling down on AI spending in 2025-2026, and NVIDIA remains a major player. Conclusion: DeepSeek is an early warning of AI cost reductions, not the end of the AI boom. Technicals: Weekly: Engulfing Bullish candle at 61.8% Fib support Expectation: 200 to 250 by end of 2025

Comments: For me, MicroStrategy is a purely technical and seems to track bitcoin more than its underlying product. It reach the year 2000's all-time highs. Technicals: Weekly Zoomed In: Weekly hammer with volume at the 50%-61.8% fib level Weekly Zoomed Out: All-time highs 5EMAs strategy RM KJ Trend Bounce ADX extra confirmation Bounce from horizontals support of the year 2000 all-time highs. Targets: 800ish by June 2025

Technicals: What appears to be an H&S looks like the hallmarks of a fake H&S by my standards. The market instead is going flat and positioning itself for a rise to 160,500 by the end of 2025 and 300,000 by 2026. This is also similar to Michael J. Saylor's trend of thought that Bitcoin will rise about 20% to 35% every year going forth. Daily: -Fake H&S -Extremes -uHd -Morning star -bounce from horizontal support Weekly: 5EMAs trend +ADX and MACD. Double wave points toward 163,074 Monthly: - price extends to 300,000 by June 2026 to May 2027

Crypto: I, personally, do not like crypto, but I like the chart of Dogecoin. I am playing it small while focusing on stocks. Technicals: -After a 4 to 5 weekly drop in Dogecoin (from May to June), Dogecoin also formed a weekly NR4 or NR7 bar. -This weekly bar is also a doji. This doji appeared right on top of weekly horizontal support from April 2022 and December 2022. -Weekly volume on the drops overall have been smaller that the rises; and the entire drop has shrinking volume overall. My indicators are showing extremes and hidden divergence at the same place. - After an a-b-c extension with have a Mathis wave of W-X-Y. -The "X" appears right in between 61.8%-78% fib. -uHd with indicator extremes

Fundamentals: Sales have accelerated in META and earnings are above 20% q/q. However, its three-year earnings are below 5%. But if this changes when earnings come out in April 2024, then that will renew META as a super stock for 2024. Weekly: uHd 50% fib and diagonal support with a-b-c extension downward into support. Kijun trend bounce target: 600-800

Fundamentals: From a speculative outlook, the fundamentals on Coinbase do not fully meet my criteria, but this IPO's balance sheet is projected to continue its acceleration in sales. Currently, this quarter's sales were at 52%; earnings over 142%. Technicals: Daily hammer at horizontal support with really high volume relative to other periods and days at or candles prior to this and around it. uHd in the histogram Based on what I perceive to be a bottom in the Coinbase valley, I look to add Coinbase's dividend paying ETF, CONY, to the portfolio.CONY is an ETF based on COI?N, paying a monthly dividend. I have invested in CONY based on the price pattern of COIN. Here is a daily chart of CONY: