Video summary

Bad News, Bearish Sentiment… and a Market Breaking Out

Main summary

Key takeaways

Finance

Finance-focused summary (markets/investing/trading)

Macro / rates / oil backdrop

  • The discussion centers on a powerful stock rally despite negative headlines.
  • Inflation / rate anxiety
    • Repeated references to yields being over 5%, framed as a risk to equities.
    • A “doom” narrative emerges if yields move toward ~5.5%.
  • Oil
    • Oil is referenced as over 100.
    • They debate a scenario where oil retreats (hypothetically to ~70), which they argue would support equities.

Fed “data” dynamic

  • After “the Fed finally spoke,” markets initially reacted negatively.
  • The next day, they claim the reaction reversed, described as typical behavior and evidence equities were not breaking down despite “incoming data.”

Bonds as the key tell (portfolio factor)

  • Primary bond instrument: TLT (iShares 20+ Year Treasury Bond ETF).
  • They argue bonds broke a prior low but then failed to extend downside (“couldn’t pick up any downside”), interpreted as improving conditions / support buying.
  • The broader framing: bond direction matters for the “AI trade,” because higher yields pressure valuation-heavy growth/AI stocks.

Positioning / crowding vs confirmation (“gates”)

  • They argue crowdedness alone isn’t enough; markets can remain crowded.
  • A key fear: participants may be contrarian in bonds the wrong way (e.g., “buying the turn” too early/too confidently).
  • Their trade decisions require multiple confirmations, not positioning by itself:
    1. Positioning / sentiment indicators
      • Mentions COT/COT-like positioning.
      • Mentions sentiment discussions on X and Discord.
    2. Market confirmation
      • They want price action to validate the thesis.
    3. “Gates” concept
      • Sentiment must imply bad outcomes are being dismissed and price action must stop confirming the bearish narrative.

Equities: “bad news failure” / AI-linked leadership

  • Despite negative AI headlines, they emphasize AI/high-growth exposure led the rally.
  • Examples mentioned:
    • SOXX: down 5.5% on Monday, then recovered; end of week it was up.
    • SOX performance: cited as +7% and +12% over the week (as attributed by speakers).
  • They also note S&P 500 (S&Ps) and Nasdaq (Q’s) were near/at all-time highs during the rally period.
  • Core idea: “Who is not going down when the market is going down?”
    • Instead of buying broad factor exposure, they prefer identifying the strongest component within the theme.

Trade examples / stock ideas (leading themes)

Strong / leading AI hardware

  • SMTC: cited as a strong “component,” breaking August highs (later described as pulling back but still showing strength).
  • AMD: referenced as hardware leadership.

Cybersecurity / software strength (new highs)

  • CRWD (CrowdStrike): new highs.
  • NET (Cloudflare): new highs.
  • OKTA (Okta): new highs.

Other software / communications

  • PLTR (Palantir): holding an “earnings gap” and pushing higher (not necessarily new highs).
  • TWLO (Twilio): “software coming back to life,” rising since June; linked to communications/agent/bot use cases.

Healthcare “AI improves diagnosis” theme

  • HT (HeartFlow): applying AI to testing/heart diagnosis.
  • Tempest AI: described as “beat up” but leading off a market bottom.
  • CI (Cardinal Insights): described as strong (“health field”).
  • NTRA (Natera): genetic testing; described as one of the biggest positions.
  • GH (Guardant Health): cancer/oncology genetic testing.
  • PFE (Pfizer): mentioned but described as not as strong.

Infra / data / other hardware-software (near all-time highs)

  • NTAP (NetApp): mentioned.
  • SNOW (Snowflake): mentioned.
  • COR / CoreWeave: described via mis-transcription as “a core weave”; the context contrasts prior failure/breakdown with “great stuff out there.”
  • CoreWeave: characterized as an AI infrastructure story that “never quite gets going” (stuck in the mud).

Sector / ETF risk framing

  • They repeatedly contrast:
    • Buying TLT (bonds)
    • vs buying stocks/AI leaders that are already showing relative strength.
  • Stated logic: if equities are working, you may not need oil/bond moves to make the stock trade work (though those moves could still act as tailwinds).

Valuation / capital cost discussion

  • The question posed: if rates rise from ~5% to 6%, does it meaningfully impact companies growing ~70%?
  • Response emphasizes financing:
    • Many AI-related companies need capital for capex; higher funding costs worsen their economics/capex math.
    • They argue the government/industrial policy environment and competition suggest capital availability should persist.

Sentiment “bubble” argument

  • They reject “bubble” framing.
  • Investor sentiment survey:
    • 53% bears (and also cited that the one-year highest bearish reading is 60%).
  • They argue that being only single digits off all-time highs while having 50%+ bearish is inconsistent with classic frenzy bubbles (e.g., South Sea bubble dynamics being far less).
  • Conclusion: even if stocks are rising, it may not be a bubble; psychology is viewed as more like “not frenzy,” partly because there’s social pushback against AI/data centers rather than widespread mania.

Indicators / monitoring framework (explicit tools)

Guardrails and confirmation signals

  • Up/Down Volume Ratio (20 days)
    • Was below a black line during a correction (described as net selling).
    • Recently moved back over 1 and is ramping higher.
  • JNK (junk bond ETF) as a bond-market proxy
    • Said to be below a declining 21-day average (“below a declining 21” / “21 average”).
    • They want JNK to reclaim above the 21-day average to confirm bond stress is easing.
  • Emphasis on watching the tape (price action), not just narratives.

Risk management / execution cautions (explicit)

  • Strong caution against overly aggressive leverage:
    • References “full margin” and short-term call options.
    • If wrong, options losses of roughly 30%–40% (or similar) can severely damage an account.
  • Execution principles:
    • Don’t fight the tape—wait for confirmation.
    • Use sizing: start with partial exposure (they cite 20–30%) and add as the trend confirms.
    • Avoid all-in / all-out decisions driven solely by headlines; allow for back-and-forth.

Timeline / market regime commentary

  • They describe a frustrating regime over the past year:
    • September to March: “nothing to do.”
    • After mid-May: more favorable, but much of the remainder is sideways/choppy.
  • They contrast this with earlier eras like 1996–1998, when there were months of consistent gains.

Explicit tickers / instruments mentioned

  • TLT (Treasury bonds ETF)
  • USO (oil ETF)
  • SOXX; also SOX (Philadelphia Semiconductor Index mentioned indirectly via SOX)
  • S&P 500 (“S&Ps”)
  • NASDAQ (“Q’s”)
  • JNK (junk bond ETF)

Individual equities mentioned

  • SMTC, AMD
  • CRWD, NET, OKTA
  • PLTR, TWLO
  • HT (HeartFlow), Tempest AI (company name as mentioned), CI (Cardinal Insights)
  • NTRA, GH (Guardant Health), PFE
  • NTAP (NetApp), SNOW (Snowflake)
  • COR / CoreWeave (infra/data-center-related name as transcribed)
  • OpenAI / Anthropic (discussed as AI labs; no tickers provided in the subtitles)

Methodology / step-by-step framework (as described)

1) Trade thesis construction (“gates”)

  • Check whether positioning/sentiment is lopsided
    • Uses COT/positioning and social sentiment (e.g., Discord/X).
  • Require confirmation from market price action
    • “Let the market confirm it.”
  • Add an “event logic” check
    • “Let there be some horrible news” scenario that would normally move bonds/stocks—yet price fails to react bearish.

2) Selection approach

  • When theme risk exists (e.g., bonds/oil), buy the strongest component
    • Rather than defaulting to the factor hedge/ETF (e.g., prioritize strongest AI-linked leaders over just TLT/SOXX).
  • Core question: “Who is not going down when the market is going down?”

3) Trend monitoring / confirmation

  • Use indicators:
    • Up/Down Volume Ratio (20 days)
    • JNK vs the 21-day average (want JNK back above the ~21 average)
  • Follow the tape
    • Don’t switch bearish purely because sentiment is bearish—confirmation must come first.

4) Risk management

  • Avoid all-in / full-margin approaches.
  • Scale exposure (starting around 20–30%) and adjust as trends confirm.
  • Treat losses carefully, especially with short-dated options.

Key numbers called out

  • Yields: over 5%; “doom” scenario at ~5.5%
  • Oil: over 100; hypothetical retreat to ~70; worst-case hypothetical to 300
  • Sentiment: 53% bears; one-year highest bearish reading 60%
  • SOXX/SOX move
    • SOXX down 5.5% on Monday, then recovered by end of week
    • SOX cited up +7% and +12% (as stated)
  • JNK indicator
    • Want recovery relative to the declining 21-day average (“21” / “21 average”)

Disclosures / disclaimers

  • No explicit “not financial advice” disclaimer appears in the provided subtitles.

Presenters / sources mentioned

  • Matt Caruso (host/co-host)
  • Jason Shapiro (co-host)
  • Historical/trading references mentioned:
    • George Soros
    • Felix Dennis
    • Trading-book references (e.g., “Market Wizards” authors; additional names referenced but not fully clarified in captions)

Original video