方法论

研究框架、选股思路、思维方式 · 共 623 条 · 第 7 / 13 页

  1. 2026-03-24 方法论

    强调投资不是团队赛,别人要自己做研究和建立 conviction。

    @miraclemaster07 投资不是团队游戏!我只是把信息摆出来。 大家应该自己消化这些信息,再做自己的决定。 我也希望大家通过自己研究来建立自己的 conviction,而不是盲目追随我个人的 thesis 帖子。

    英文原文

    @miraclemaster07 Investing is not a team game! I’m just putting information out there. People should synthesize that and make their own decisions. And I’d hope people build their own conviction by doing research themselves rather than blindly following my own personal thesis posts

  2. 认为新瓶颈 / 卡点无法用线性收入模型预测,只能基于供应链位置做主观判断。

    像 $AXTI 或 $SIVE 这种新的瓶颈 / 卡点,前瞻收入根本没法建模。 你只能拿它跟龙头、它们的客户去做估值比较,再根据它在供应链里的位置做一个主观判断,比如 Jabil 的 1.6T transceiver、Ayar -> AI 芯片、$MRVL Celestial,或者更多还没公开的客户。 任何分析师如果说自己能准确建模收入放量路径,那基本都是在胡扯,哈哈。

    英文原文

    It's impossible to model forward revenue for new bottlenecks/chokepoints like $AXTI or $SIVE. It's valuation comparison to leaders, their customers, then just making a discretionary estimate on where it heads based on positioning (eg. Jabil 1.6T transceivers, Ayar -> Ai Chip, $MRVL celestial, and maybe more undisclosed customers). Any analyst who tries to model an accurate projection of revenue ramp is just giving BS lol.

  3. 2026-03-24 方法论

    说除非有风投愿意和自己合作,否则 Pandora’s box 会继续关着。

    @McLeary_77 除非有风投团队愿意跟我一起做个想法,不然潘多拉盒子就还是关着。

    英文原文

    @McLeary_77 Pandora’s box is staying closed unless a venture arm wants to work with me on an idea.

  4. 2026-03-23 方法论

    指出他们不是长线持有者,只是交易波动。

    @dadele2nd 这个判断很准确。 他们不是持有者,他们是在交易波动率。 这对股价不是好事,你更希望的是被动指数基金和长期持有人。

    英文原文

    @dadele2nd That’s an accurate assessment. They’re not holders, they trade volatility. Not a good thing for the stock, you want passive index funds and long term holders

  5. 2026-03-23 方法论

    说自己会用简单格式分享如何建立 conviction。

    @CifrBunny 很高兴听到这个! 我会尽量用一个简单的格式,分享我是怎么建立自己的 conviction 的。

    英文原文

    @CifrBunny Glad to hear that! I try to share my how I developed my conviction in a simple format

  6. 认为即使宏观不佳,光子供应链瓶颈在资金流入后仍会跑赢大盘。

    不会。实际上,像 $AXTI 这种票今年迄今回报 280%,对比 $SPY 还是 -3%,它们反而会大幅跑赢指数! 如果数千亿美元的超大规模云 CapEx 沿着供应链往上游流向这些公司,尤其是光子学这些瓶颈: 那么基本面的重估(尤其像 $SNDK 那样,靠提价带来经营利润的案例)会无视宏观环境。

    英文原文

    Nope. If anything they’ll all strongly outperform the index as seen with $AXTI 280% YTD return vs. $SPY -3% YTD! If hundreds of billions of hyperscaler capex flows upstream to these companies, especially in photonics sectors bottlenecks: The fundamental re-rating (esp. with price hikes -> operational income as seen with $SNDK) will ignore macro climates.

  7. 2026-03-23 方法论

    感谢别人多次提到公司,促使自己更深入研究。

    @Plaskpojken 谢谢你这么多次为这家公司发声!尤其是在私信里。 你其实是我后来更深入研究这家公司的很大原因之一 :)

    英文原文

    @Plaskpojken Thanks for shouting out the company so many times! Especially in DMs. You’re a large reason I ended up looking into the company deeper :)

  8. 2026-03-23 方法论 $AXTI

    说自己平时不会频繁 victory lap,但 AXTI 和光子选股最近实在太夸张。

    @IdeaLedger 抱歉,通常一个人每涨 40% 就 victory lap 一次,可能也得隔几个月。 但 $AXTI 和光子股最近的表现太夸张了,几乎每隔一天就来一次。 这也有助于我获得新关注。

    英文原文

    @IdeaLedger Sorry, normally when someone victory laps every 40% gain, maybe it’s every few months. $AXTI and optical picks have been an anomaly every other day. Also helps with getting new followers

  9. 2026-03-23 方法论 $AXTI

    回顾自己对 AXTI 的 150 美元目标价,认为是靠抢先识别瓶颈。

    大家都觉得我疯了,说我把 $AXTI 的目标价从 12-15 美元上调到 150 美元。 然后怀疑者一下子都没声音了? 我今年年初至今回报超过 600%+,原因就是我总是比别人先识别超大规模云供应链里的最大瓶颈。 然后再做多。 https://t.co/lgmu8p7F5a

    英文原文

    Everyone thought I was crazy when I gave $AXTI a $150 PT from $12-15. All the doubters suddenly disappeared? Reason my YTD is over 600%+ is because I identify the biggest chokepoints in hyperscaler supply chains before anyone else. Then go long. https://t.co/lgmu8p7F5a

  10. 2026-03-23 方法论

    说这只是随手分享信息,通常只会挑最喜欢的赢家深入写 thesis。

    @devandraan 这只是很随意的东西,就是一路把信息倒出来。 通常我只会挑一个赢家,或者我最喜欢的那个,然后再专门写 thesis。

    英文原文

    @devandraan This is just informal stuff, just dumping info along the way. Normally I'd just the winner or one I like the best then do a thesis on it.

  11. 2026-03-22 方法论

    说会去找市场还没发现的更小众垄断。

    @YungSmaaash 我会尽量去找那些市场还还没意识到的小众垄断。

    英文原文

    @YungSmaaash I would try and find more niche monopolies markets don’t know about yet

  12. 2026-03-22 方法论 $AXTI

    认为世界正在围绕石油、稀土和半导体打架,很多供应链已被武器化。

    世界现在在争夺的就是: 石油、稀土和半导体。 原因在于……每一条供应链都已经被武器化了。 而且碰巧,美国最大的地缘政治对手们掌握着大多数瓶颈点。 我最早把 $AXTI 的瓶颈和光子学 + AI 超大规模云供应链联系起来。 但还有很多我不敢公开说的,单纯因为它们涉及国家安全漏洞。 不过这些主题上的投资逻辑依然成立: 大量资金会被投向这三个板块,不是为了去打它们的战争,就是为了在国内把它们保住,或者在 2026 年绕开它们。 从主题上说,如果你的组合没有重仓在全世界都在争夺的这三个方向上…… 也许你哪里弄错了?

    英文原文

    The World is fighting over: Oil, Rare Earths, and Semiconductors. The reason being... Every supply chain is being weaponized right now. And it so happens, US biggest geopolitical enemies own most of the chokepoints. I was one of the first to connect $AXTI's bottleneck to photonics + AI hyperscaler supply chains. But there are many more that I'm scared to broadcast, just due to national security vulnerabilities. But the thematic investments hold true: Massive amounts of funding will be poured into these three sectors to either go to war over them, secure them domestically, or engineer around them for 2026. As a theme, if your portfolio is not overweight into those three right sectors the entire world is fighting over... Maybe you're might be something wrong?

  13. 2026-03-22 方法论 $AXTI

    说自己 12 月几乎天天发 AXTI,现在只是等市场晚一点消化信息。

    @maxmay2024 是啊,我去年 12 月几乎隔天就发一次 $AXTI。 现在只是等市场在几个月后慢慢把这条信息消化掉。

    英文原文

    @maxmay2024 Yes, I posted almost every other day about $AXTI back in December. Now I'm just letting markets price in that information few months later.

  14. 2026-03-21 方法论

    认为股票长期是正和游戏,公开分享研究有助于信息更快扩散。

    谢谢!我个人觉得,股票从长期来看是正和游戏。所以价值创造和新信息的传播,最终大家都会受益,没必要设付费墙。 以前的标准模式是:一份 2 万美元的分析师报告 -> 对冲基金先拿到 -> 股价上涨 -> 散户在市值已经到 200 亿美元以上后才知道。 现在有了 X,散户可以和机构抢先一步,第一时间拿到同样的信息发现。 (我并不反对分析师或想把成果变现的人;我只是更愿意打破这个老模式,把内容公开发出来。)

    英文原文

    Thanks! I personally think stocks are a positive sum game long term. So everyone benefits from value creation and novel information distribution and there’s no need to paywall. Before the standard model went from $20,000 analyst report -> hedge funds accumulate -> price goes up -> retail finds out after something is $20B+ Now with X, retail are able to frontrun institutions, by getting the same information discovery instantly. (No problem with the analysts or people who want to monetize their work, I prefer just to break the trend and post it publicly)

  15. 2026-03-21 方法论

    解释自己额外页面是用来放研究材料和零散想法的。

    别觉得有压力去看,我不会把我的 thesis 帖子设付费墙。 那个页面只是多出来一块空间,用来放一些研究材料或者沿途漏掉的随机想法。我不想把主时间线刷得太满。 但如果其中某个点子或思路,能启发你自己的想法,那也挺好。

    英文原文

    Don’t feel pressured to, I don’t paywall any of my thesis posts. That page was extra space for research material that gets lost along the way or random thoughts. Didn’t want to flood my main timeline. But might be helpful if one idea or thought processes helps spawn some of your own

  16. 2026-03-21 方法论 $MU

    回顾自己用便签式测算提前推演 MU 毛利率,结果与实际财报非常接近。

    回到一月份: 我当时用便签式的粗算方法,拿美光的财报毛利率预测和官方指引做了对比。 我的原话是: “如果 $MU 指引 2026 年 Q2 的毛利率为 68.0%,而 NAND 价格从估算的 33-38% 直接上涨 100%, 那毛利率预测就可能被推到 73-75% 以上。” 华尔街当时的共识大概就在 67% 到 69% 左右。 而当 $MU 真正公布财报时? 74.9% 的毛利率,正好落在我当时预判的 73-75%+ 区间。 这就是为什么信息会在财报前就被提前计价。 所以这根本不是一个“意外结果”,因为连我这种普通人一月份都已经能估出来 / 提前计入定价了。

    英文原文

    Back in January: I napkin-math modeled Micron's earnings margin projections vs. their official projections: My quote: "If $MU guided for 68.0% gross margins Q2 2026, and we see a 100% NAND hike from 33-38% est. That could bring gross margin projections over 73-75%." Wall Street consensus was sitting right around 67% to 69%. When $MU reported earnings? Their 74.9% margin landed in that predicted 73-75%+. This is how people price in information early, before the actual earnings. So it's not a "Surprise Result" when even people like myself could estimate it/price it in, back in January.

  17. 2026-03-21 方法论 $MU

    说明存储与光子股在宏观下跌中仍然坚挺,因为利润与增速过高,重定价通常发生在财报前。

    同意。记忆芯片 / 光子相关标的在更广泛的宏观抛售里还能撑住,原因就是它们的净利润(存储)和收入增长(光学)都高得离谱。 如果没有伊朗战争,$MU 可能会高得多,但核心观点仍然成立:绝大多数重估发生在财报前,而不是财报当天,除非真的出现重大意外。

    英文原文

    Yep agreed. Reason why memory/photonics is holding up against broader macro selloffs is that their net income (for memory) and revenue growth (for optical) are just way too high. If there were no Iran war $MU would likely be tons higher, but the core point still stands that most repricing happens before earnings, not on the actual day, unless there's a genuine surprise.

  18. 2026-03-21 方法论 $MU

    总结 MU 的长期看多逻辑,并强调期权要在信息被市场完全定价前买。

    是的,我确实认为 $MU 长期还会继续走高。 如果要给一个 TLDR: 1. 财报周不要去买被做空的期权,因为大概率财报信息已经被计价了。 2. 如果真要买期权,要在市场开始计价信息之前就下手(比如 DRAM / NAND 价格上涨的传闻阶段)。 3. 我不认为资本开支是这次抛售的主因,更多是期权资金流相关,然后市场需要一个故事去解释它。 4. 宏观逆风也很多,但总体来说,重估发生在财报前,靠着不同信息碎片慢慢拼出基本面变化。 财报只是对这一过程的确认而已。

    英文原文

    Yeah I do think $MU is going higher long term. If I had to give a TLDR: 1. Don't buy shorted options on earnings week, since earnings is more likely than not priced in. 2. If you're going to buy options, do it before markets price in information (eg. around rumors from DRAM/NAND price hikes) 3. I don't think capex was the reason for selloff, lot of it is option flow related and then there needs to be a story for it. 4. Lot of macro headwinds too, but in general, repricing happens leading up to earnings, piecing together different information that changes fundamentals. Earnings is just confirmation of that.

  19. 说自己的 UPWK 逻辑在 Reddit 上花了很久才兑现,并认为短期价格波动常被人拿来挑错。

    谢谢!是啊,我的 $UPWK 逻辑大概是在 Reddit 上花最久时间才兑现的,哈哈。不过我觉得这些帖子从方向上通常还是对的,只是时间上不一定精确(我记得我 Reddit 的 PT 帖大概 11 个里对了 8 个左右吧,不是每次都准)。 很多人喜欢抓短期价格来挑错(比如今天 $SIVE 的下跌,或者 $NBIS 在可转债消息后的下跌)。但我觉得,我大多数 thesis 帖从长期来看还是正向兑现的。

    英文原文

    Thanks! Yeah my $UPWK thesis probably took the longest time to play out on Reddit haha, but I think they're usually right directionally over time (I remember being ~8 for 11 or something from my Reddit PT posts, don't always get them right). People like to cherry pick short term prices (eg. $SIVE drop today or $NBIS drop post convertible note) to say I'm wrong. But, think most of my thesis posts play out positively over the long run.

  20. 2026-03-20 方法论

    说自己故意把内容写得很简单,方便散户理解半导体术语。

    @stfbutnou @SemiAnalysis_ @FundaAI 嗯,没错,SemiAnalysis 的研究确实更强,但谢谢你突然这样怼我 lol。 不过顺便说明一下:我故意把内容写得很简单,就是为了让散户更容易看懂半导体术语。

    英文原文

    @stfbutnou @SemiAnalysis_ @FundaAI Well no sht semianalysis does better research but thanks for randomly putting me down lol. But just an FYI, I publish stuff extremely simple on purpose so retail investors can understand semi jargon better.

  21. 2026-03-20 方法论

    说这些内容只是出于好玩分享,觉得把不同主题结合起来会很强。

    @Ayouubb_ 我发这些纯粹是为了好玩,免得大家觉得我这些想法 / 回报没意思。 之前还出现过 AI Displacement ETF,后来又有了 AI bottleneck ETF! 感觉把这三个东西组合起来会非常强。

    英文原文

    @Ayouubb_ I just drop these for fun in case people find any ideas/returns interesting. There was also the AI Displacement ETF not too long ago and then the AI bottleneck ETF! Feels like combining the three together would be pretty OP.

  22. 2026-03-20 方法论

    表示自己只是为了好玩做加权组合,让大家可以直接按板块投资,省去部分 ETF 费用。

    我只是为了好玩才这么做,万一大家只是想直接用这些权重去投资整个板块呢! 这样也能避开现在不少 ETF 的管理费(而且理论上应该会跑赢)。 很多基金其实没覆盖这些名字里很大一部分,因为不少资管经理会忽略上游那些细微但关键的瓶颈!不过我也没打算正式推出一个官方产品 lol

    英文原文

    Je fais juste ça pour le plaisir au cas où les gens voudraient simplement utiliser ces pondérations et investir dans le secteur dans son ensemble ! Ça permet d'éviter les frais de gestion de pas mal d'ETF actuels (et ça devrait surperformer). Beaucoup d'entre eux n'incluent pas une grande partie de ces titres, car beaucoup de gestionnaires d'actifs passent à côté des goulots d'étranglement subtils en amont ! Mais je ne prévois pas d'en lancer des officiels lol

  23. 建议把仓位投向即将到来的硅光 / CPO 方向,并在高信念和分散之间做平衡。

    @moderndayvenom 我会把资金投向硅光 / CPO,因为我们知道这会从 OFC 和 $NVDA GTC 继续推进。 像 $TSEM、$SIVE、$SOI 这种名字,是我最喜欢、信念最高的三个。 然后我会稍微分散一点,今天价格下的 $RDDT 也可以作为多头,另外当前收发器周期里的 $AAOI 也不错。

    英文原文

    @moderndayvenom I’d send it into silicon photonics/CPO since we know it’s coming from OFC and $NVDA GTC. Names like $TSEM, $SIVE, $SOI were my favorite three higher conviction ones. Then I’d diversify a bit, maybe $RDDT as a long as of today’s prices. And $AAOI for current transceiver cycles.

  24. 强调股票长期是正和游戏,价值创造最终让所有人受益。

    @SaiseiInvesting 股票是正和游戏!所以从长期看,价值创造会让所有人受益。 如果看 $LITE 或 $AXTI,过去几个月基本就是一路直线上涨。

    英文原文

    @SaiseiInvesting Stocks are positive sum! So everyone benefits long term from value creation. If we look at $LITE or $AXTI it’s just been a straight line up past few months

  25. 2026-03-19 方法论

    提醒别忘了大盘也曾回落,等市场恢复时自己持有的名字会更高。

    @BasedCapitalLP 我觉得你忘了更广泛的大盘也曾经回落。 等它恢复之后,只会意味着我持有的那些名字会比现在更高! https://t.co/DPDl1yhgK3

    英文原文

    @BasedCapitalLP I think you’re forgetting the broader market did falter. When it recovers, that just means the names I own go higher than they are now! https://t.co/DPDl1yhgK3

  26. 解释自己很早知道 SIVE,但因为觉得硅光周期还早而没买,直到 GTC 出现催化。

    我之前就知道这家公司一段时间了,因为 Reddit 上一直有人狂吹 $POET,而那也是他们最喜欢的股票。 Sivers 也时不时被提到。 只是我一直觉得硅光 / CPO 还有很长时间才会真正落地,所以从没真正做多。 但 $NVDA GTC 成了它需要的催化剂,尤其是在……

    英文原文

    Was aware of it for awhile from all the $POET shilling back on Reddit as it's their #1 favorite stock over there. And Sivers got name dropped from time to time. Just felt like silicon photonics/CPO was quite a ways away so never went long. But $NVDA GTC was the catalyst it needed, especially with the new Jabil announcement. So ended up going long.

  27. 总结自己的研究流程:先发 thesis,再做跟进研究,最后在 thesis 兑现后庆祝并持有。

    谢谢! 我的风格通常是: -> 先发 thesis 帖 -> 再做后续研究(比如 $AXTI 的 7N 铟出口管制、瓶颈) -> thesis 兑现后再庆祝 如果之后也没什么新东西,就继续拿着。 市场里大部分钱,都是靠找出别人不知道的东西赚到的,所以我现在就是在继续做这类功课,尤其是 $SIVE。

    英文原文

    Thanks! My style is usually: -> Thesis post -> Follow up research (eg. $AXTI 7n indium export controls, bottleneck) -> Then celebration when thesis plays out And then just sitting on it (if there's nothing too new). Most of the money in markets is to be made finding out stuff others don't, so just doing some of that work now with $SIVE.

  28. 2026-03-18 方法论 $XLU

    认为 XLU 是一笔 boring 但很长周期的交易,自己会继续持有。

    @Darkhawk720 @_stockResearch $XLU 没什么变化,它就是一笔无聊但多年的交易。所以我个人只是继续拿着仓位。

    英文原文

    @Darkhawk720 @_stockResearch Nothing's changed with $XLU, it's boring multi-year trade. So I'm personally just sitting on my positions.

  29. 认为 Soitec 的上涨验证了自己“免费公开研究会被迅速定价”的观点,散户在 X 上甚至领先机构。

    Soitec <$SOI / $SLOIF> 自从我一周前发帖后,已经涨了很多。 因为它们是西方在硅光和 CPO 基板上的事实垄断者。 我免费发想法,它们会立刻被计价进去。 这和以前的模式不同: -> 银行分析师卖 1 万到 4 万美元以上的研究 -> 对冲基金低价先吸筹下一个瓶颈 -> 散户等到涨到顶部后才知道。 而第一次,X 上的散户在架构范式转移这件事上,比机构更早。

    英文原文

    Soitec < $SOI / $SLOIF > is now up, a lot since my post a week ago. As they’re the Western monopoly over silicon photonics and CPO substrates. I post my ideas for free, and they get priced in immediately. Instead of former models where: -> Bank analysts sell research for $10,000-$40,000+ -> Hedge funds can accumulate the next chokepoints at low prices -> Retail investors find out at the top after they’re already priced in. And for the first time: Retail on X are earlier than institutions to the architectural paradigm shifts in AI.

  30. 说明自己在容量等于收入的假设下估算 ASP,并引用 AAOI 与 LITE 业绩电话会作支撑。

    @FlashBozZ 是的,我确实写过那篇分析。ASP 是一个估算值,而我假设产能最终会转化成营收。 从 $AAOI 的财报电话会来看,他们说超大规模云厂商几乎什么都买,只要他们能造出来;再加上 $LITE CEO 也明确说他们已经完全卖光了。 https://t.co/UFwfdo0xzg

    英文原文

    @FlashBozZ Yes I did a writeup on it. The ASP is an estimate, and I’m making the assumption capacity translates to revenue. Given per $AAOI ER they stated hyperscalers were buying anything they could build. As well as $LITE CEO statements about being completely sold out https://t.co/UFwfdo0xzg

  31. 2026-03-18 方法论 $AXTI

    说自己在 $12 时就感觉 AXT 能到 140 美元,并强调自己的判断有尽调支持。

    @FoundersX76 我的意思是,我在 $AXTI 还在 12 美元的时候就“感觉”它能到 140 美元……现在 2-3 个月就到 50 美元左右了? 当然,我不是只靠感觉,我会用尽调来支撑这种感觉,比如超大规模云供应链映射、营收估算或者估值比较。

    英文原文

    @FoundersX76 I mean I "felt" $AXTI could hit $140 when I was back at $12... And now it's ~$50 in 2-3 months? Of course I back my feeling with DD like hyperscaler supply chain mapping, revenue estimates, or valuation comparisons.

  32. 2026-03-18 方法论

    请求优化 ticker 识别逻辑,避免把市值误判为公司名。

    @nikitabier 很高兴听到。最近 AI 垃圾信息确实不少。 顺便问一下,能不能推一个 ticker 识别的改动? 这样它就不会把市值误识别成公司名了,比如这种情况。 https://t.co/AiVhGjgs4n

    英文原文

    @nikitabier Glad to hear. There’s been a decent amount of AI spam recently. On a side note, it possible to push a change for tickers? So it doesn’t classify market cap as company names eg. https://t.co/AiVhGjgs4n

  33. 强调自己会先用前瞻基本面和供应链终端需求建模,而不是先看图形。

    我会先把前瞻性的基本面(营收、利润)建模进去,然后再让市场去给 $AAOI 这样的公司定价。 或者像 $SIVE 那样,去看供应链终端客户,比如 $AMZN 或 $MSFT 对光源的需求。 市场里大部分 alpha,本来就是这么找到的,不是靠画一条歪歪扭扭的线。

    英文原文

    I try to model in forward fundamentals (revenue, profit) before markets price the company in with stuff like $AAOI. Or with $SIVE, looking at supply chain end users like $AMZN or $MSFT for light sources. That's how most of the money is to be made in markets finding alpha, not drawing a swiggly line on a chart.

  34. 2026-03-17 方法论

    认为很多 agentic 场景需要本地硬件,尤其是跟自动化相关的 T&S。

    @wanna_B_trader 很多 agentic 用例都需要本地硬件,因为和自动化相关的 T&S 需求很多。

    英文原文

    @wanna_B_trader Local hardware is needed for a lot of agentic use cases because of automation related T&amp;S.

  35. 2026-03-17 方法论 $RPI

    指出本地模型不直接跑在树莓派上,而是把 picoclaw / openclaw 等当作编排器。

    @MingleSach 你不会直接在 Raspberry Pi 上跑本地模型。你会把 picoclaw / openclaw 之类的东西当作编排器去移植。

    英文原文

    @MingleSach You don’t run local models on raspberry pi. You port picoclaw/openclaw/etc. as an orchestrator.

  36. 2026-03-17 方法论

    认为便宜的本地硬件适合运行 openclaw 这类 agent 编排工具。

    @pablocuellar_3 便宜又独立的硬件,很适合跑 openclaw 这类 AI agent 编排工具。

    英文原文

    @pablocuellar_3 Dirt cheap isolated hardware for running openclaw (ai agent orchestration)

  37. 拿 ARKK 和语言教育类 AI 公司举例,认为投资瓶颈型光子供应链可能更好。

    @zzenrocker 不如你去买 Cathie Wood 的 $ARKK ETF,里面有像 $DUOL 这种“下一代 AI 公司”,也就是被 AI 驱动的语言教育受益股。 也许这比去押注真正支撑超大规模云 AI 基础设施的关键 CPO 光子瓶颈更好。

    英文原文

    @zzenrocker Idk, you can buy Cathie Wood’s $ARKK ETF, for “next generation AI companies” like $DUOL, the "super stock" set to benefit from AI-driven language education. Might be better than critical CPO photonics bottlenecks powering hyperscaler AI infrastructure.

  38. 认为光子 TAM 太大,完全可以做一个相关 ETF,但自己更喜欢直接持有个股。

    很高兴听到!是啊,光子对 AI 的 TAM 实在太大了,所以会出现很多不同名字。 你当然可以把 $FORM、$KEYS、$VIAV、$AEHR 以及另外 20 多家公司做成一个 ETF,然后坐等。 不过……我个人更喜欢自己挑赢家。

    英文原文

    Glad to hear! Yeah, there's so many different names since photonics TAM for AI is massive. You can definitely make an ETF of stuff like $FORM, $KEYS, $VIAV, $AEHR, and the 20+ other companies. Then probably sit back and chill. But... I personally like to try and pick a few of the highest upside winners personally.

  39. 2026-03-17 方法论

    承认 neocloud 初期资本开支极重,也不喜欢稀释。

    @Blue_1Trades 对,neocloud 行业不幸的地方就在于,最初扩张阶段的资本开支极重…… 我也确实不喜欢稀释。

    英文原文

    @Blue_1Trades Yeah, that's the unfortunate part about neocloud industry is that it's extremely heavy capex during the initial buildout... I'm really not a fan of dilution either.

  40. 对比不同赛道的毛利率和资本开支,认为 photonics 相比重资本 neocloud 更轻。

    它们属于不同的行业、不同的毛利率,也有不同的资本开支需求。 $NBIS 和 $IREN 通常需要很重的稀释或融资来买 GPU,但从长期看通常仍然是增值的。 光子业务的资本开支通常轻得多,毛利结构也不一样。

    英文原文

    They are different sectors, gross margins, and different capex requirements. $NBIS and $IREN typically require heavy dilution or financing to buy GPUs, but typically are accretive in the long run. Photonics operate might lighter in capex, different gross margin profile. Different growth rate. So can’t compare 1:1

  41. 认为 CPO TAM 呈指数级增长,自己看多 SIVE、AXTI 等受益于 architectural shift 的标的。

    看看这张图就知道了。未来几年 CPO TAM 是直接一条直线上去的。 市场通常会害怕那种“一次性”周期,但这看起来是指数型的。 我说过 $SIVE 可能是下一个 $LITE,所以我会长期持有。基板供应商比如 $AXTI 和 $SOI 也是一样。

    英文原文

    Just look at this graph. Straight line up for CPO TAM next few years. Normally markets are scared of one-and-done cycles, but this looks exponential. I said $SIVE could be the next $LITE so I’m long for awhile. Same with substrate providers like $AXTI and $SOI. As for current transceiver bottlenecks, would re-evaluate next year.

  42. 2026-03-17 方法论

    解释自己白天发的多是研究材料和思考过程,正式观点会放到主时间线。

    @blue__mount 这只是我在白天分析供应链或者交易想法时,随手记下来的思考 / 研究材料和过程。 等我整理完之后,才会正式发到主时间线。 我这样做是为了不让主时间线塞满那些别人不一定想看的东西……

    英文原文

    @blue__mount It’s just my thoughts/research material + process throughout the day when I’m analyzing supply chains or trade ideas. I post formally on my main timeline after I’m done. I made it so I don’t fill up my main timeline with stuff people don’t want to see. https://t.co/TLC4Y1ybyJ

  43. 2026-03-17 方法论

    博主解释关注价值在于行业趋势洞察而非股票推荐

    很高兴听到这些。这再次印证了我的想法:很多人,特别是行业领袖关注我,不是为了股票推荐。而是因为我通常能提供有关前沿领域发展方向、时间框架的洞察,以及该领域的关键公司有哪些。我通常会把这些映射成投资论点给交易者参考,但人们也可以忽略这部分,只想了解酷科技或隐藏威胁——比如'关键材料→机器人领域'。

    英文原文

    Glad to hear that. And it reaffirms my thought that a ton of people, especially industry leaders follow me, not for stock picks. But because I usually provide good insight into where frontier sectors are heading, the timeframe, and what the critical companies in the space are. I typically map that into investment thesis majority of the time for the traders, but people can ignore that if they just want to learn about cool tech or hidden threats “eg. Critical materials -> robotics sector.”

  44. 博主说明发帖时机与原则:不鼓励跟单,愿分享想法让粉丝独立判断

    当我的投资论点可能发生变化时我会发帖,比如 $SNAP 的股权激励(SBC)。或者当有重大新闻时发帖,比如 $VPG 可能进行大规模生产,当时有特斯拉签约数百家中国供应商的消息。然而,我不想让人跟单交易或在我卖出时同步卖出。我更愿意分享我的想法和风险提示。然后让人们得出自己的结论。但如果人们在随机帖子中直接问我,我通常会说是否已减仓或退出了仓位。只是X平台不会把所有帖子都展示给粉丝,所以我确信很多人错过了很多更新。

    英文原文

    I post when my thesis might have changed such as $SNAP SBC. Or with possibly $VPG with mass production, when there was news on Tesla contracting hundreds of Chinese suppliers. However, I don’t want people to copy trade or sell at the same time I do. I’d rather post my thoughts, and risk disclosures. Then let people come to their own conclusions. But if people ask directly in random posts, I often say if I de-risked or exited my position. It’s just that X doesn’t show all my posts to followers, so I’m sure people missed a ton of updates.

  45. 说自己去年看多 ALAB 和 CRDO,今年则转向纯光子方向,并会在仓位变化时公开说明。

    @StephanC86 我去年发 thesis 的时候,是做多 $ALAB 和 $CRDO 的。 但今年我已经转向纯光子方向了。 如果我改变仓位,我通常也会发帖说明。不过我一天会发很多帖子,而 X 的算法又不会把大部分内容推给关注者……

    英文原文

    @StephanC86 I was long on $ALAB and $CRDO last year when I posted my thesis. But I pivoted to pure play photonics this year. I do make posts when I change my positions, but I make a ton of posts every day and X algo doesn’t show most of them to followers. So not everyone sees my updates.

  46. 强调独立研究的重要性,指出恐惧驱动短期价格,但经营收入驱动长期价格。

    提醒一下:当整个社区都在恐慌发帖讨论内存股名字的时候。从$EWY(SK Hynix、Samsung)、$MU和$SNDK。在伊朗冲突初期。说“Kospi和内存股会因氦气或LNG下跌30-40%”。我是全程保持看多的少数分析师之一。因为我实际上做了研究来支持我关于LNG、原油和氦气几乎没有实质性影响的观点。伊朗冲突仍在进行中,关于潜在油价飙升仍有很多未知因素:但恐惧和煽动性标题驱动短期价格。经营收入驱动长期价格。只要知道从方向上看,内存对AI建设至关重要。美国不会让他们的AI建设因中东冲突而停滞,并在AI——这个历史上最变革性的技术——上失去对中国的优势。

    英文原文

    Just a reminder: While the entire community was doomposting about memory names. From $EWY (SK Hynix, Samsung), $MU and $SNDK. during the initial Iran conflict. Sayin “Kospi and memory names would drop 30-40% because of Helium or LNG.” I was one of the only few analysts that remained bullish the entire time. As I actually did the research to back up my claims about little material disruption of LNG, Crude, and Helium. The Iran conflict is still ongoing, and there’s a lot of unknowns still regarding potential oil spikes: But fear and sensational headlines drive short term prices. Operating income drives long term prices. Just know that directionally, memory is critical to the AI buildout. United Stares wont let their buildout stall, and lose China, in AI - the most transformative technology in history. Over a Middle East conflict.

  47. 2026-03-16 方法论 $SIVE

    说自己先把 SIVE thesis 给订阅者看,但不喜欢长期付费墙,所以会公开发出来。

    @BonyBallf2 我是先把 $SIVE 的 thesis 分享给订阅者的。不过我不喜欢把信息整合永久锁在付费墙后面,所以我总会在之后把想法公开发出来。 https://t.co/MYHYzPOaWD

    英文原文

    @BonyBallf2 I shared my $SIVE thesis with my subscribers first. However, I don't like keeping my information synthesis permanently paywalled, so I always release my thoughts publicly afterward https://t.co/MYHYzPOaWD

  48. 表示无法控制市场如何给免费公开研究定价,并认为自己只是把 Sivers 用可读方式梳理出来。

    我没法控制市场如何给我公开发布的免费信息整合去定价。 WSB 那帮人已经关注 $POET 一阵子了,也提过 $SIVE,但没人把 Sivers Photonics 以一种容易读懂的方式梳理出来。我也觉得他们错过了时机。 随着资金轮动到硅光 / CPO,我个人觉得现在是合适的时候。 我持有多头仓位,但我并不是建议别人也去做这笔交易。我只是想分享我的研究和想法。

    英文原文

    I can't control how the market prices in free public information synthesis posted publicly. The WSB crowd has been on $POET (and talked about $SIVE) for a little bit, but nobody was synthesizing Sivers Photonics in a readable way. I also think they missed the timing. With capital rotating into silicon photonics/CPO, I felt it was the right time personally. I hold a long position, but I am not recommending others take this trade. I just wanted to share my research and ideas

  49. 认为 neocloud 的数据中心如果有更酷的命名,估值可能会更高。

    这是我的一个热观点: 像 $NBIS 或 $IREN 这样的 neocloud,如果把数据中心起个更酷的名字,估值可能会高 1.3 倍。 酷名字示例: - $POET 的 “Starlight” - $NVDA 的 “Blackwell” 但数据中心呢? - $IREN 的 “Prince George” - $CIFR 的 “Barber Lake” 为什么要把 AI 数据中心命名成你的理发师名字,或者英国小孩的昵称? 我发誓,除了现有投资者之外,没人会对 Iren 的 “Prince George” 数据中心上线感到兴奋。 但如果大家都看到 $NBIS 的 “Singularity” AI 数据中心上线,那听起来就酷多了。 Nebius 现在本该值 400 亿美元了。

    英文原文

    This is my hot take: Neoclouds like $NBIS or $IREN would command 1.3X higher valuation... If they renamed their datacenters to something cool. Cool Name Examples: - $POET "Starlight" - $NVDA "Blackwell" But datacenters? - $IREN "Prince George" - $CIFR "Barber Lake" Why are you naming an AI datacenter after your hair cut stylist or a British Kid nickname? I swear, nobody outside existing investors are going to hyped over Iren's "Prince George" datacenters coming online. But if everyone see's $NBIS "Singularity" AI DC coming online, that just sounds badass. Nebius would be worth $40B by now.

  50. 介绍自己的订阅区只是“洗澡时的灵感区”,不会把主线内容付费锁住。

    哇,谢谢大家这个月的 4300 位订阅者! 我最近刚把它设好,但还没正式发过帖。 这是我的第一篇! 不过,这段时间我一直收到大量私信,所以顺手做个 PSA: 我做的任何 DD(比如 $SOI 到 $AAOI)最终都会出现在我的主时间线。 我不会把任何 thesis 帖永久锁在付费墙后面,但我可能会在订阅区稍微提前发一点。 尤其是因为 HFT 算法也会爬我的主帖,而我更愿意给零售投资者一个先手。 不过本质上,我把订阅区当成一个“洗澡时的灵感”空间。 所以那些零散的供应链、笔记、交易想法,比如: - 套利 / 期权 - DCA 加仓位 - 还没变成完整 thesis 前的早期想法 - 或者随机的 AMA 问答 我不想让主时间线塞满这些不正式的思考过程或评论。 所以我觉得,顺着这个过程分享我学到的东西,同时先给真正关注的人一个提示,会更有意思! 如果你想看我白天到底在想什么,那这个频道就是干这个的!

    英文原文

    Wow, thank you all for 4300 subscribers this month! I recently set it up, and never made a formal post about it. This is my first one! But, I've been getting floods of DMs about it. So just to clarify things with a quick PSA: Any DD I do ( $SOI to $AAOI) will always end up on my main timeline. I won't gatekeep any thesis posts I make, but I might post a tiny bit earlier in the sub spaces. Especially since HFT algorithms scrape my main posts too and I like to give retail audiences a heads up before the algos. But I basically use the subscription channel as a "Shower Thoughts" space. So any random supply chains, notes, trade ideas like - arbitrages/options - DCA levels - early ideas before they become full thesis or actionable ideas - or random ama questions I just didn't want to flood the main timeline with informal thought process or comments. So thought it would be more fun along the way to share what I learn and give real folks a heads up first. If you're interested in what I'm thinking about throughout the day, that's what the channel is for!