以下是内容的中文翻译： 在他们“最大输家”一期节目的后续中，主持人肖恩·奥马利和丹尼尔·芒克讨论了他们的“最大赢家”，回顾了这些投资成功的原因，并探讨了它们今天是否仍值得买入。他们还提到了那些他们曾分析但最终没有买入的公司，并强调了从中吸取的教训。主持人强调了他们“利益攸关”的投资方式，对于所分析的每一只股票，他们都在个人投资组合中持有重要头寸。 **Alphabet (谷歌)** 谷歌是他们最大的持仓，自最初买入以来已实现近100%的收益。肖恩最初的估值是每股约180-200美元，基于分部估值法。围绕谷歌的叙事发生了巨大转变：最初人们担心它会被OpenAI颠覆并面临司法部的审查，但其Gemini大语言模型后来居上，谷歌也将AI整合到搜索中。这表明市场极易受叙事影响，导致数万亿美元的价值波动。尽管早期担忧集中在B2C大语言模型变现（计算成本高昂）上，但主持人认为，真正的利润在于B2B人工智能产品，而Alphabet的产品套件（Gmail、Google Drive）在这方面具有显著优势。 当前的担忧围绕着人工智能和数据中心的大规模资本支出（CapEx），今年预计为2000亿美元，明年可能达到3000亿美元。这种支出导致谷歌第二季度自由现金流自首次公开募股以来首次转为负值，导致债务增加，并可能发行股票而非回购。然而，谷歌云业务增长惊人（同比增长80%以上），拥有高达0.5万亿美元的巨额积压订单，表明需求强劲。尽管面临这些挑战，主持人仍对谷歌的长期内在价值复合增长充满信心，不过丹尼尔指出，他不会在其当前“乐观”的估值（70倍自由现金流）下增持谷歌，而是会让他的赢家股继续增长，遵循“伟大企业总是带来意想不到的惊喜”的建议。 **亚马逊 (Amazon)** 亚马逊是近期才持有的股票（2月份买入，5个月内上涨35%），也面临巨额资本支出（超过2000亿美元），导致第二季度自由现金流为负。然而，亚马逊在AI竞争中拥有独特优势：它建立了价值200亿美元的内部芯片业务，并通过AWS、内部效率以及AI代理可能带来的电商顺风，获得了巨大的收益。与谷歌不同，亚马逊的估值被认为更具吸引力（经调整后为营运现金流的17-18倍），丹尼尔表示，他目前会将新资金投入亚马逊而非谷歌。该公司垂直整合的模式和多元化的业务，从云服务到支付和广告，使其无论AI的最终影响如何，都能良好发展。 **Reddit** Reddit一直是一个“过山车式”的赢家，部分获利了结后实现了140%的收益，后续再次买入的头寸还有60%的未实现收益。当估值变得极端时，主持人选择削减该头寸，理由是Reddit的多元化程度低于Alphabet。尽管Reddit第二季度营收增长60%，但登录用户同比增长仅为7%，这是一个管理层决定停止报告的指标——此举引发了对透明度和可持续性的担忧。 历史上被认为难以变现的Reddit，其营业利润率达到30%的速度远超预期。然而，它面临对谷歌搜索流量的依赖，而且AI概览正在降低用户点击外部网站（如Reddit）的比例。尽管Reddit的数据对大语言模型至关重要（谷歌每年支付6000万美元，这被认为是一个较低的数字），但考虑到AI驱动的搜索行为变化，主持人对其公司的长期用户增长战略表示质疑。尽管有这些担忧，丹尼尔仍看好Reddit在AI主导的世界中作为人类互动平台的潜力，尽管考虑到波动性和用户增长问题，他不太倾向于积极增持。 **从分析过但未买入的公司中吸取的教训** 主持人还讨论了几家他们曾分析但未纳入其内在价值投资组合的高表现公司，并从中吸取了关键教训： * **Remitly**：一家汇款公司，丹尼尔个人投资组合中大量买入（上涨80-90%）。肖恩对支付公司的怀疑（价格竞争激烈，加密货币颠覆）阻止了它被纳入主要投资组合。Remitly展现出强大的“盈利能力”，利润率迅速提高。这里的教训是，内在价值投资组合专注于长期、具有护城河的企业，这使得将需要更积极管理或投资逻辑快速变化的公司纳入其中变得困难。 * **Crocs**：一家时尚零售公司，在丹尼尔分析后价值翻倍，但他们卖出过早。这凸显了跟踪较小、波动性更大的头寸的挑战，以及更广阔的结果范围，导致错失上涨机会。 * **台积电 (TSMC)**：一家他们同意买入但“忘记”了的公司，因为肖恩对地缘政治风险（中国-台湾）和快速变化的半导体行业有重大担忧。台积电凭借其“令人惊叹的护城河”和在芯片制造中的关键作用，价值翻倍。教训是不要对风险过于两极化；即使存在有效的长期担忧，一个小额对冲头寸也能抓住显著的上涨机会。 * **戴尔 (Dell)**：受益于AI基础设施需求，其基础设施业务同比增长70%。尽管AI服务器合同利润率较低，但巨大的销量推动了可观的利润增长，使其市盈率（PE）倍数急剧扩张（从历史平均8.5倍增至今天的22倍）。主持人认为，随着资本支出最终放缓，这种倍数是不可持续的，这突显了市场“前瞻性”但可能过于乐观的本质。 * **Comfort Systems**：一家为数据中心提供HVAC（暖通空调）服务的公司，他们最初认为这是“无聊”的业务，但其价值增长了5倍。其在复杂数据中心方面的专业化使其成为AI建设意想不到的受益者。教训是，AI等宏观顺风对即使是成熟、不那么引人注目的企业也能产生强大而常被低估的影响，同时警告说，如果AI支出周期减速，此类估值倍数可能会收缩。 最后，主持人引用彼得·林奇的名言——“一生中成功的投资只需少数几个大赢家，它们的收益将远远超过那些表现不佳股票的损失”——强调了持续的学习过程和寻找长期复合增长股的目标。

In a follow-up to their "biggest losers" episode, hosts Sean O'Malley and Daniel Munker discussed their "biggest winners," reflecting on why these investments succeeded and whether they remain buys today. They also touched upon companies they covered but didn't buy, highlighting lessons learned. The hosts emphasized their "skin in the game" approach, holding significant positions in their personal portfolios for every stock covered. **Alphabet (Google)** Google, their largest holding, has seen a nearly 100% gain since their initial purchase. Sean's original pitch valued it around $180-$200 per share, based on a sum-of-the-parts valuation. The narrative around Google shifted dramatically: initially feared to be disrupted by OpenAI and facing DOJ scrutiny, its Gemini LLM later became a leader, and Google integrated AI into search. This demonstrates the market's susceptibility to narratives, causing trillions in value swings. While early concerns focused on B2C LLM monetization (expensive compute), the hosts believe the real money is in B2B AI products, where Alphabet's product suite (Gmail, Google Drive) offers a significant advantage. Current concerns revolve around massive capital expenditures (CapEx) for AI and data centers, estimated at $200 billion this year, potentially $300 billion next. This spending has made Google's Q2 free cash flow negative for the first time since its IPO, leading to increased debt and potential share issuance instead of buybacks. Google Cloud, however, is growing phenomenally (80%+ YoY), with a massive $0.5 trillion backlog, indicating strong demand. Despite these challenges, the hosts remain confident in Google's long-term intrinsic value compounding, though Daniel noted he wouldn't add to Google at its current "optimistic" valuation (70x free cash flow) but would let his winners run, following the advice that "great businesses surprise you to the upside." **Amazon** Amazon, a more recent holding (February, up 35% in 5 months), also faces substantial CapEx ($200 billion+), resulting in negative Q2 free cash flow. However, Amazon has a unique advantage in the AI race: it has built a $20 billion internal chip business and benefits immensely through AWS, internal efficiency, and potential e-commerce tailwinds from AI agents. Unlike Google, Amazon's valuation is seen as more attractive (17-18x operating cash flow adjusted), and Daniel indicated he would deploy new capital into Amazon over Google currently. The company's vertically integrated model and diversified businesses, from cloud to payments and advertising, position it well to thrive regardless of AI's ultimate impact. **Reddit** Reddit has been a "rollercoaster" winner, with a 140% realized gain after taking profits and a 60% unrealized gain on subsequent re-buys. The hosts chose to trim this position when valuations became extreme, citing Reddit's lower diversification compared to Alphabet. While Reddit's Q2 revenue grew 60%, logged-in user growth was a modest 7% YoY, a metric management decided to stop reporting—a move that raised concerns about transparency and sustainability. Historically seen as difficult to monetize, Reddit achieved 30% operating margins much faster than anticipated. However, it faces dependency on Google Search for traffic, and AI overviews are reducing click-through rates to external sites like Reddit. Although Reddit's data is crucial for LLMs (Google pays $60M annually, a figure seen as low), the hosts question the company's long-term user growth strategy given the AI-driven changes in search behavior. Despite these concerns, Daniel remains bullish on Reddit's potential as a platform for human interaction in an AI-dominated world, though less inclined to add aggressively given the volatility and user growth issues. **Lessons from Companies Covered But Not Bought** The hosts also discussed several high-performing companies they covered but didn't add to their intrinsic value portfolio, drawing key lessons: * **Remitly**: A remittance company Daniel bought heavily for his personal portfolio (up 80-90%). Sean's skepticism about payment companies (race to the bottom, crypto disruption) prevented its inclusion in the main portfolio. Remitly demonstrated strong "earnings power," rapidly increasing margins. The lesson here was the intrinsic value portfolio's focus on long-term, moat-driven businesses, making it difficult to include companies requiring more active management or with rapidly changing theses. * **Crocs**: A fashion retail company that doubled in value after Daniel's coverage, but they sold too early. This highlighted the challenge of tracking smaller, more volatile positions and the potential for wider outcome ranges, leading to missed upside. * **TSMC (Taiwan Semiconductor Manufacturing Company)**: A company they agreed to buy but "forgot" due to Sean's significant concerns about geopolitical risks (China-Taiwan) and the fast-changing semiconductor industry. TSMC, with its "lollapalooza moat" and critical role in chip manufacturing, doubled in value. The lesson was not to be overly binary on risks; a small, hedged position could have captured significant upside even with valid long-term concerns. * **Dell**: Benefited from AI infrastructure demand, with its infrastructure business growing 70% YoY. Despite low margins on AI server contracts, the sheer volume drove substantial profit growth, causing its PE multiple to expand dramatically (from 8.5x historical average to 22x today). The hosts believe this multiple is unsustainable as CapEx eventually slows, highlighting the market's "forward-looking" but potentially over-optimistic nature. * **Comfort Systems**: An HVAC service provider for data centers, which they initially viewed as a "boring" business, saw a 5x increase in value. Its specialization in complex data centers made it an unexpected beneficiary of the AI build-out. The lesson was the powerful, often underestimated, impact of macro tailwinds like AI on even mature, less glamorous businesses, while cautioning that such multiples might contract if the AI spending cycle decelerates. Concluding with a Peter Lynch quote—"All you need for a lifetime of successful investing is a few big winners and the pluses from those will overwhelm the minuses from the stocks that don't work out"—the hosts emphasized the ongoing learning process and the goal of finding long-term compounders.