CurryOption

Over the past couple of days, AI frontier models have been going at it again: Gemini 3.8 FLASH/MUSE/Fable. Performance benchmarks look solid; next up is chatGPT 5.7 or 6 (Astra). I don't think chatGPT will widen the gap, though model capabilities may start to diverge—while chatGPT could lead in reasoning, Fable may maintain its edge in coding. For true collaborative workflows, however, chatGPT Codex remains the most practical choice for now, given its more realistic usage quotas.

[Relationship between Stock Price and Volatility Spread] If we use a bi-weekly timeframe, plotting the return of $Rocket Lab(RKLB.US) stock price on the y-axis against the volatility spread on the x-axis, we can visualize how the volatility spread changes during periods of significant price drops or rallies over two weeks.

I was just looking at the top ten US stocks with the highest short interest. I suddenly saw a familiar name, and only just now noticed the staggering data.$Ondas(ONDS.US) has issued a total of 570.55M shares, with as many as 533.41M shares circulating in the market for trading. In other words, most of the shares are publicly float, yet simultaneously (right now) as many as 234.97M shares are shorted.

Please ask Grok to look up information online about the cooling performance of DGX Spark. Although the chip and memory are identical, cooling is crucial for long-term use. Every brand has its own cooling strategy, so there will be significant differences in temperature comparisons. What surprised me was that Acer topped the list? I haven't considered buying a laptop or PC from Acer for over ten years... This cooling ranking really caught me off guard:

Before I knew it, I spent so much money on the Google Store that I became a Platinum member...

This afternoon at 2:30, while everyone's phones were ringing with the mobile network outage drill warning messages, I was sitting in the Yunlin County Government conference room, explaining the principles, architecture, and functions of NTsocial/Meshtastic to everyone. Halfway through my presentation on stage, all our phones started sounding the national-level alert tone, creating a very immersive atmosphere!

Model student $Apple(AAPL.US) is about to release its earnings report. Looking at the volatility spread between long-term and short-term, there is a consistent light buying pressure for put options. In terms of traded options, based on new positions, put options are also the main focus.
Even though the stock price has hit a new high, there is not a hint of bullish arrogance.

The two products from the past couple of days have put me in deep thought.

I bought the MacBook a little over a month ago, and this morning I used its built-in speakers to play music for the first time. Wow! Is this really a laptop? The sound quality is simply on par with a bedside speaker. How did they manage that? Producing such a speaker-like effect in such a thin chassis.

I originally thought TSMC ($Taiwan Semiconductor(TSM.US)) could be as reassuring as in the past, so I haven't looked at TSMC for quite a while, even on the eve of the earnings report.........

I've found a better way to work: I only use "5.6 sol ultra-high reasoning" during the planning phase, and use "5.6 Terra ultra-high reasoning" for regular work. Discuss and outline the architecture and workflow with 5.6 sol, and save the discussion results directly as a markdown file. Then, open 5.6 terra and ask it to follow that markdown to complete the specific coding and testing tasks. If it's a long-term development project, you don't really need to set the speed to 1.5x acceleration. Because even with 1.5x acceleration, even 5.6 terra will use up its 5-hour quota. If it runs out, and you don't immediately pull out a magic little card to pay for more quota, but just sit there waiting, it's pointless. Instead, it's better to set it to normal speed and let 5.6 Terra output steadily and consistently. After using it continuously for over 24 hours, I'm very sure that with ultra-high reasoning enabled, 5.6 Terra's performance is significantly better than the previous generation chatGPT 5.5. As for the previous generation GPT 5.5, I honestly don't think it has any issues for work anymore; frankly, it's already sufficient. The main significance of 5.6 for me is during research and discussions; it's a more intelligent collaborator, and its spoken Chinese expression has also improved somewhat. In purely scientific discussions, it's a more reliable partner than 5.5.

I just noticed by chance that Microsoft had officially fallen below $356. It's currently rebounding. If it can break through $401 in the next few days and return to the earlier boxed oscillation range from mid-April to the end of May, it will be relatively safer. The worst fear is that it just touches $401 and falls back down. If this pattern emerges, I think it will be even harder to hold the $356 level on the next drop. Enough talk, I need to quickly check the $Microsoft(MSFT.US) options data. I'll post an analysis article in forty minutes.

It's been a long time since I wrote about the "Options Talk" series. Today, I came up with a topic. I haven't started researching it yet, but I'll share my thoughts with you first. Perhaps before I start, the feedback from everyone can help me complete a more exciting mini-research project.

〖Is the implied volatility of options with 14 days to expiration consistent with the actual stock price volatility that occurs in the following 14 days? Underestimated? Overestimated?〗

People who participate in options trading have a preconceived notion in their minds about the distribution of stock prices 14 days later. This notion determines how they judge the price of that option.

If an options participant believes that stock price volatility will be high, and there is a high probability of breaking through the strike price in 14 days, then they are willing to accept a high market price for the option.

If an options participant believes that stock price volatility will be low, and the probability of the stock price breaking through the strike price in 14 days is very low, multiplying the small probability value by the breakthrough price results in a very small value, then they will price the option at a very low price.

Options traders' judgments about prices often contain an emotional premium. This leads to the actual future 14-day stock price volatility perhaps always being lower than the implied volatility. The portion where "implied volatility" is higher than "actual future 14-day stock price volatility," the excess part, is the emotional premium that investors themselves assign.

What is this premium? Is it positive or negative?

Which stocks have particularly high emotional premiums? Which ones are particularly low?

When are emotional premiums particularly high? When are they particularly low?

#Volatility Risk Premium (VRP)

$Marvell Tech(MRVL.US) recently had a piece of negative news that I believe will have no substantial impact. On 06/11, the company announced a CFO transition: the current CFO Willem Meintjes will step down. The latest update on 06/15: Meintjes filed Form 144, planning to sell approximately 207,000 shares (nearly 48% of his holdings), amounting to about $60.1M.

$Intel(INTC.US) taught me an important lesson this year.

$AST SpaceMobile(ASTS.US) Since the Blue Origin rocket explosion, stocks in the space industry have been acting strangely. $AST SpaceMobile(ASTS.US) stock price has been highly volatile, and in the short term, it's also been affected by the SpaceX (SPCX) IPO, showing a clear phenomenon of "fund rotation + sell the news." Last Friday, the stock price fell 15.53%, and the current price is $82.41.

The most clear positive catalyst recently is that AST SpaceMobile has confirmed that BlueBird 8, 9, and 10 will be launched on a SpaceX Falcon 9 from Cape Canaveral on June 17.

But whether a company can operate well isn't solely about technological strength.

The top-left chart shows the implied volatility over the past two years. In this chart, we can clearly see the significantly declined stock price while volatility remains high. This is a relatively positive aspect I see in today's data, but how positive this is must be judged with the help of the Put/Call volatility spread.

The top-right chart shows the recent trend of the volatility spread. This chart looks quite unfavorable because, despite such a large drop in stock price, the main trading pattern is still chasing to buy call options. The volatility spread has been in a deeply negative state most of the time. Very few market participants are buying Puts to hedge against the price drop; everyone is still in the fervent speculative atmosphere of buying on every dip.

For the stock price to rise sharply, it naturally needs speculative frenzy to boost it. However, during the process of price correction and consolidation, if it doesn't trigger panic and spur a frenzy of chasing to buy put options, it's very unfavorable for the price correction and consolidation. In plain terms: the deleveraging of the long positions hasn't been completed yet.

Looking at the bottom-right chart, the correlation coefficient analysis between stock return and volatility, as expected, the current correlation coefficient still maintains a significantly positive correlation. The stock price has fallen so much, but compared to February and April this year, we cannot determine that the price correction is complete based on the existing data.

The bottom-left chart is the option position change chart from last Friday's trading. There is still a more noticeable increase in call option positions, but put option positions have also started to increase, which is a good sign. I think the stock price shouldn't rush to rebound subsequently. It would be a better state if it could consolidate weakly and let the bullish trading sentiment cool down.

Summary: The stock price has already fallen so much, but the analysis data surprisingly suggests "it hasn't fallen enough yet."

I'm on a flight back to Taiwan.
On the way there, I had to drive as soon as I landed, so I didn't drink at all. On the way back, I'm taking a car to Hsinchu, so I'm drinking freely on the plane with peace of mind.
I'm feeling quite relaxed, after all, I've completed a mission, taking my parents on a big loop around the northernmost part of Japan's main island all by myself.

We are so dedicated to trading and investing because of love.
We hope to reach a state where we live in complete freedom. If you understand yourself deeply enough, you'll probably find that you don't want to make money because you love money. Fundamentally, you want to make money because you love yourself. Making money is just a means, not an end.
You want to avoid grievances, you want to live the way you desire, so you yearn to earn enough money.
You, we, make money because of love.

At the same time, we hope our family can also enter this state. We hope the people we love can also live as they like.
So, we invest our efforts, we dare to endure.

First $Broadcom(AVGO.US) then $Oracle(ORCL.US), this wave of pullback probably won't end early.
All things that are particularly meaningful are never easy. I understand your difficulties, and I'm certain these difficulties will accompany us for life.
But, it's okay, because with love, whether it's love for ourselves or love for our family, we can withstand disappointment after disappointment.
By the time you read this message, I'll already be back in Taiwan. I've returned to the team, let's fight side by side again!

$AMD(AMD.US) (Advanced Micro Devices) suddenly dropped 10.86% last Friday. Does such a significant single-day decline create any particular pattern in the options trading data?

I'll state the conclusion first: the existing structure remains unchanged.

The top-left chart shows the two-year overall implied volatility. This chart indicates that after the last earnings report, the baseline level of implied volatility was elevated by one notch and did not fall back to the pre-earnings baseline. Post-earnings, the trading sentiment entered a new atmosphere (or narrative).

Following last Friday's sharp decline, the current volatility level is still fluctuating within the range of the new baseline.

The top-right chart shows the recent Put/Call volatility spread. After last Friday's drop, the volatility spread jumped upwards, but the magnitude of the jump is not large and remains within the typical fluctuation range.

After Friday's decline, neither the volatility level nor the volatility spread has shown a new pattern. It seems the current stock price movement is still considered a 'normal energy release' for $AMD(AMD.US) by market participants.

The bottom-left chart shows the single-day position change. It reveals some long positions closing out, but more notably, put options were opened across a wide range of strike prices. These trades aren't clearly about who is exiting or entering, as under these general position changes, the volatility spread change is minimal, only shifting slightly towards puts, with no clear directional bias emerging.

The bottom-right chart shows the correlation coefficient between stock price returns and volatility. Last Friday's drop also caused the correlation coefficient to fall from a positive correlation above 0.5. It would be more reassuring if it could touch zero, because after May, there was a local stock price low point, which was accompanied by the correlation coefficient briefly touching zero. If it doesn't even touch zero, it doesn't resemble a complete stock price pullback pattern.

Summary: While $AMD(AMD.US)'s stock price is pulling back, volatility hasn't changed much. Position changes show increased buying interest in puts and some call options being closed out, but these are not significant. The structure presented by the options trading data remains the same as before the sharp price drop. According to correlation analysis, it is indeed in a consolidation phase of the price pullback. If the subsequent correlation coefficient can drop to negative or at least touch zero, then we can start watching for the potential emergence of a local bottom in the stock price.

$Marvell Tech(MRVL.US)'s stock price surged after 06/02. Although it experienced a significant pullback in the most recent trading day, volatility remains at historically high levels. The current short-term volatility exceeds 120%, which is sure to attract many options traders to sell options.

$Micron Tech(MU.US) Micron is the data pattern I've seen in the past two days that is relatively reassuring.

After $Broadcom(AVGO.US)'s earnings report, a full trading day has passed. Now, we're trying to analyze the current situation using options trading data. Just one trading day after the earnings, the overall implied volatility has almost returned to its pre-earnings initial value. The top-left chart shows that the current implied volatility has fallen to the upper edge of the pre-earnings benchmark. There will definitely be a wave of high and low fluctuations in volatility going forward. The key point we need to observe is whether the lowest point of the benchmark shows signs of gradually rising higher. If it does, that would be an early sign of the stock price preparing to reach new highs. The top-right chart shows the Put/Call volatility spread trend. I thought the shocking lesson of a single-day drop of over 12% after earnings could push the volatility spread back above the zero line in one day, but it didn't. In fact, it remains significantly below the zero line. In plain language: the single-day 12.59% drop didn't deliver a shocking lesson to market participants; they're still not scared enough! The bottom-right chart is a slightly more positive message. We see that the sharply falling stock price has already pushed the maximum pain point below some of the options. For the mid-June monthly options, it has now become possible for the maximum pain point price to catch up with the stock price. The maximum pain point for the 06/18 options is currently near $375. I believe this can be seen as the first support level where the stock price might potentially stop falling in the near term. The bottom-left chart is also not too optimistic. Even after yesterday's big drop, the correlation coefficient between stock returns and volatility remains at a strongly positive value. While not 100% accurate, this phenomenon usually appears around the peak of a stock price. Conclusion: $Broadcom(AVGO.US) currently shows no signs of stopping its decline; it hasn't fallen enough yet!

$QQQ seems to have fired the first shot after $Broadcom(AVGO.US)'s earnings report the day before yesterday, and then the market sentiment turned sour. During yesterday's session, the declines of many individual stocks were somewhat contained, but $Broadcom(AVGO.US), which ignited the fuse, still fell by 12.59%.
This decline is probably not something that can be resolved by a one-day pullback; at least $Broadcom(AVGO.US) hasn't shown any signs of stopping the decline yet.

If we look at the trading data of the broader market ($QQQ). The top-left chart shows the overall implied volatility over the past two years, which is currently in a neither-here-nor-there position. Whether it's going back to the low point at the beginning of the year or the high point in April, we can always deduce some ideas. Unfortunately, the current volatility is at a moderate 20% level.

The top-right chart shows the implied volatility over the past two years. The current volatility spread is also very close to past benchmark values, and like most sectors, it is dominated by the volatility of put options.
Quite pessimistic. According to the two charts above, if we are currently in a downtrend, we can say with certainty that we are still in the early stages, and there is still some distance to go before reaching near the bottom.

The bottom-left chart is the intraday position change chart. This chart is even more sensitive, showing active put option opening. But this is not necessarily bullish or bearish, because despite such obvious new put option positions, the volatility spread hasn't shown a significant jump (whether up or down).

The bottom-right chart shows an even worse characteristic.

How bad is it?

In the trading of the past year, the QQQ ETF has never had a positive correlation coefficient. In the past year, it was only in the last month that the stock price return and volatility showed a positive correlation for the first time.
The bottom of a stock price is more often associated with a highly strong negative correlation, while the top of a stock price is prone to trading overheating.
When the stock price rises and the call option volatility also continues to rise, this is a behavior of chasing prices to buy Calls. When the stock price falls slightly, the volatility of Calls also falls in sync. The bullish sentiment is shallow; it follows rises and flees from falls. This synchronized, sensitive state is the primary condition for incubating a significant downtrend.

Conclusion: Everyone, be careful! There are no signals of a price bottom, and some signals indicate that the main decline phase has not yet appeared.

Tonight's dinner was the most moving meal I've had since coming to Japan. I didn't do much research beforehand, just found a highly-rated eel restaurant near today's hotel. Upon arrival, I realized the shop was even older and more worn than I imagined. If you're not careful when flipping through the menu, the whole thing might fall apart.

Since eel cuisine prioritizes freshness, the owner only started preparing after we placed our order. The process involved many steps and was quite time-consuming. After getting the owner's permission, I filmed him working from the side.

After taking a few photos, I sat down to have a drink. While waiting, I inadvertently noticed a yellowed, mottled "Seikin-sho" (Diligence Award) hanging on the wall. Asking AI to analyze the content of the "Seikin-sho," I was surprised to find it was a certificate awarded fifty years ago. In other words, the owner is probably around eighty years old now.

For over fifty years, he has focused on doing one thing every day: eel rice. The existence of this shop and the bowl of eel rice before me are the most concrete manifestation of "shokunin kishitsu" (craftsman spirit).

Boris Cherny, senior engineer at the Anthropic team and creator/head of Claude Code, was interviewed by host Lenny Rachitsky. Lenny asked him, "Once AGI (Artificial General Intelligence) is fully realized one day and humans no longer need to write code for a living, what do you plan to do?" Without hesitation, Boris answered that his next goal is to return to Japan, seclude himself, and devote himself to fermentation techniques, aiming to create the most authentic, perfect miso soup in all of Japan, even the world.

After tonight's dinner, I think I have my own answer. If one day I achieve what I consider the "complete Holy Grail" in the field of trading, then my next job will be to come to Japan and learn to make authentic eel rice: "Three years to skewer, eight years to split, a lifetime to grill."

$Echostar(SATS.US) acquired SpaceX equity by "selling spectrum." In 2025, it reached a spectrum transaction with SpaceX (AWS-4H block, approximately 65 MHz), with SpaceX paying a large amount of cash plus up to $11 billion worth of SpaceX stock. This deal significantly improved EchoStar's balance sheet and resulted in SpaceX holdings accounting for an extremely high proportion of $Echostar(SATS.US)'s market value.

Since the rocket explosion incident of Blue Origin, the stock performance of space-related stocks has not been good, and we haven't looked at $Echostar(SATS.US)'s trading data for a while.

Overall implied volatility of $Echostar(SATS.US) over the past two years (top-left chart): Although the stock price has recently pulled back, implied volatility remains high. This high level is unlikely to come down before SpaceX's IPO. Comparing its historical record, the peaks of particularly high implied volatility did not correspond to stock price tops; they even had the potential to be starting points for rallies. From this perspective alone, it is positive.

Volatility is still present; stock price momentum is still present.

The top-right chart shows the recent trend of the Put/Call volatility spread. This result surprised me. We can see the volatility spread entering a deep negative range, with current call option buying momentum extremely strong. Interestingly, even though the stock price has significantly retreated, the momentum of long-side chasing is even stronger.

The bottom-left chart separately plots the volatility of options. It is very clear that the strong SpaceX IPO narrative persists, and for all options expiring within the next month, call option volatility is significantly higher than put option volatility. Long-side momentum is very strong and confident, ignoring the recent stock price decline.

The bottom-right chart shows that in the last trading session, there was a significant scale of call option closing positions. If the volatility spread continues to move more negative even while these positions are being closed, I speculate that the closing was led by buyers. Sellers who originally sold call options took profits and exited early, taking advantage of the significant recent stock price decline.

Conclusion: Although the $Echostar(SATS.US) trading landscape still has a strong $SpaceX IPO narrative component, even disregarding this event, $Echostar(SATS.US) trading remains dominated by call options.

There's no bad news heard about $Lumentum(LITE.US)'s fundamentals. The current stock price movement looks more like a pullback after a rally.