陈小晖
陈小晖
夫唯不争，一位略懂投机之道的混子。 只用2倍杠杆。 且看能小资金3年能不能走出来。
1.7KFollowing
1.3Kfollowers
Feed
Feed
Pinned
This month, I really have to thank $SNDK and $SKHYNIX
and also Elon Musk's big rocket $SPCX for the opportunities, which allowed the account to have a relatively good increase.
Victory and odds were controlled, and I also appreciate the company of brothers along the way. Let's see how far we can go this year, proving that a small capital of 1500u can also compound and grow.
I hope the market continues to provide opportunities next month, aiming to reach 12,000u next month #Elon Musk responds to Morgan Stanley, $3.5 trillion revenue may come seven years early
-2.46%
Snapshot at 29 Aug 2026, 16:13
Korea is not opening the market today, the Japanese index has a decent rally, and the Taiwan stock market has even hit a historic high. Tonight when the US market opens, this semiconductor sector is expected to perform well.
There haven't been any major news changes these days, so just keep holding the trend stocks and wait for the market to choose a direction. AI hardware funds have to pick a direction to go. For AI application fields, it depends on whether the funds will flow in or not.
Long positions: The odds for current $SNDK and $MU long positions are getting smaller and smaller, and arbitrage is not very meaningful anymore. Better to miss out than to make a wrong move.
It's better to hold more $ETH and Bitcoin.
Short positions: Short Hynix and oil, the odds are very high once this trend forms. Just don't go heavy all at once; short on rallies at high points.
#霍尔木兹仍未开放，OPEC+维持11月产量不变
-8.07%
Snapshot at 05 Oct 2026, 14:50
Recently, I haven't been doing arbitrage because the volatility is low, and it's time to choose a direction.
The rebound in AI tech stocks is in its final stage; except for giants with sufficient cash flow, for those that have fallen a lot, buying is basically fine.
For those traditional cyclical stocks using AI narratives up to now, if they haven't broken previous highs and are consolidating, don't try to guess the bottom. Going long now only has a 3% stop-loss space, and the odds aren't high. It's really better to wait for a clearer direction.
There are still quite a few surprises to look forward to. Previously, there were many positives; going forward, there will gradually be random negatives.
Those starting to short for trend should begin testing positions. If the previous highs aren't broken, and the rebound is within a downtrend space, short on rallies, give the market some time. The odds here are much better than going long.
So for upcoming trend trades, unless unexpected events occur and the logic is falsified, just hold on. Use time to exchange for space with high odds.
Finish the longs on this wave of Tem that don't break the lower band, then go all in on shorting Hynix.
#非农降温难压美债收益率，长期利率压力仍在
-6.62%
Snapshot at 03 Oct 2026, 15:09
Yesterday's nonfarm payrolls were positive, and the semiconductor rebound was quite good, with the Nasdaq pulling back after breaking new highs.
However, storage has been affected by news; previously, despite price increases, no new factories were being built. But now, some are being constructed. The next question is whether Seagate and Western Digital will follow suit, and it's highly likely they will expand production. Data center growth is still ongoing, but locking orders without production or delivery turns into financial leverage. HDD leads the way; will others be far behind? So the reason SanDisk and Micron also pulled back yesterday mostly relates to supply-side bearishness. Currently, there's no way to boost gross or net profit again; price hikes won't last long due to supply expansion, similar to the logic behind Micron's recent decline—physical limits.
To capture data center market share, they can't keep locking orders indefinitely. They can afford to wait because the difference lies in the grid connection time, which is longer than the storage cycle.
Therefore, the market is a bit chaotic now, lacking the resonance and various optimistic expectations seen before July. It's better to be cautious and trade less.
With the rebound relatively high, continue shorting Hynix.
$SKHYNIX $SNDK
#美国9月非农仅增2.9万，失业率升至4.2%
-6.35%
Snapshot at 03 Oct 2026, 14:26
Reviewed last month's trades, and indeed they were a complete mess, not following the pattern. Started with a rebound on very low volume, doing range arbitrage, and although the three trades based on news didn't lose money and even made some profit, it was clear that I didn't control the little voice in my heart, lacking patience.
Used naked K-lines of different timeframes all at once. Frequently switched targets, trying to apply four different timeframes to different targets, which definitely didn't work.
Got greedy, causing issues with the entry points of each trade. Not patient enough to execute. Attention was not focused enough.
Next, after adjusting my state, I will separate by entry points and return to trading based on the trend of the timeframe. What to do and what not to do at the end of a rebound cycle. Will strictly adhere to this.
After all, missed trades cost nothing, but reckless trades can lose everything.
If I can't adjust this month, then it's time for more self-discipline. Stay out of the market and understand myself before coming back.
If I want to make another 100x in 3 years, some ups and downs are actually good. The early stage was too smooth, which felt a bit unusual.
Wishing everyone a new month with your accounts soaring all the way.
#财报观察员：美光上调指引，存储需求继续走强
-7.35%
Snapshot at 02 Oct 2026, 17:10
A big pullback on the first day of October, wiping out the $TEM profits 😂. Holding on to wait a bit longer, it’s going to retest the launch platform pivot, which is quite exhausting, but the direction is confirmed: if no stop loss is hit, just ignore it; if stop loss is hit, close the position and look for other opportunities. Wait for another retest of the lower track to stabilize before acting.
The market is too chaotic right now. Range arbitrage has been going on for 2 months. Now we have to wait for the rebound to end before entering and adding to short positions. First wait for the rebound and retest to confirm the end, then expand the short position. Try to trade as little as possible.
Stop arbitrage for now, switch everything to trend trading: $TEM long trend, $SKHYNIX short trend.
See how the non-farm payrolls go tonight; good news could ease some rate hike expectations, which is also good. Because even if they hike, it won’t be much before hitting another peak.
Still watching when US Treasury yields can recover a bit; without ending the stop loss on tech longs, I’m not confident setting it too high.
#9月非农今晚公布，加息预期成焦点
-13.96%
Snapshot at 02 Oct 2026, 14:33
Just reviewed Micron's earnings report and organized some key points. $MU
Revenue: The consumer segment's share is almost gone. Now it's mainly data centers.
The earnings report's long-term orders lock in about a quarter of 30 years' revenue,
with a margin of 32 billion and receivables of 150 billion. About 17% is deposits.
The current production expansion bottleneck is the lack of clean rooms, but the clean room construction cycle is about 2-3 years, corresponding to storage starting from 2024. It's roughly okay now. The equipment for clean rooms is not very scarce; the shortage is in manpower and time. Physically, the limit is about this time.
Also, the data center cycle is 2-3 years, and the current limitation for data centers is the queue for grid connection. This is a physical limit that can't be solved by stories, so although computing power is still lacking, the essence is the inability to get data center grid connection approvals.
If the factory building is ready but can't be used, buying chips first doesn't benefit oneself, so locking orders plus queuing for data center grid connection is a win-win.
Therefore, the key is to watch the speed of data center grid connection. This is the hardest and requires a 3-5 year queue, during which storage prices can drop before buying.
So, in terms of the cycle, it's time to start shorting, although there can still be short-term rebounds, the space is insufficient.
But Micron is at least US-based, so still short Hynix.
#财报观察员：美光上调指引，存储需求继续走强
-1.57%
Snapshot at 01 Oct 2026, 16:07
After the $MU earnings report came out, it neither rose nor fell. Although it is not far from the previous high, yesterday the Nasdaq plunged at the close, yet funds flowed into the semiconductor sector. The more important thing is how it performs after today's opening.
PCE met expectations, reducing the probability of inflation-driven rate hikes, combined with falling oil prices, but the Nasdaq did not show much reaction, and US Treasury yields remained unchanged.
I still feel that the rebound cycle is nearing its final stage. The rebound has been oscillating around this level, with little arbitrage space, so only short-term trades are possible, which I am a bit reluctant to do.
Waiting for a higher probability opportunity to open short positions. If the Nasdaq does not break the previous high and Micron remains stagnant, then it's time to increase short positions.
Open a small short position on Hynix first to see the direction of this wave. Set a 10-point stop loss if it follows a trend. Waiting for specific nodes, as the current movement is indeed too chaotic.
As long as TEM's trend is not broken, I will hold on a bit longer.
Going forward, I will no longer do short-term arbitrage, only trend trading.
The direction will emerge.
#加息预期推迟，9月非农成下一关键
-1.81%
Snapshot at 01 Oct 2026, 12:48
Yushu closed his position, too bad he didn't send a holiday red envelope 😂. Next, I'll focus more on short positions. For the $TEM long positions I hold that aren't weakening, I'll continue to keep them.
I'll also look for suitable entry points to open short positions on $SKHYNIX and crude oil.
Tonight probably won't be calm. Let's watch the show; if there are upper and lower wicks, it’s still okay to open a short position.
Starting from 8:30, there will be wicks, and at midnight, another earnings report will cause some volatility. Let's see if it will consolidate or prepare to retest the previous high.
#财报观察员：美光财报临近，AI存储需求成焦点
-1.98%
Snapshot at 30 Sept 2026, 14:45