How to read this page: each insight is a method — the trigger, the steps to a position, and the signal to watch when re-running it. The boxed line shows how it played out here. This page carries the methods
distinct from the
Jul 16 insights (which cover the trim-and-re-enter discipline, the ~2.5-year post-bubble clock, the bull's-own-numbers refutation, and the IPO patience rule — not repeated here). Timestamps deep-link into the video.
5:23 1. The 20%-max-downside screen — price the bear case first
The repeatable method
- Before assessing upside on a new buy, estimate its price in a normal index bear market (assume a garden-variety 20–25% drop, nothing catastrophic).
- Translate that index move into the name's own projected downside using its cyclicality/valuation — a high-beta, richly-valued group falls far more than the index.
- Reject anything whose projected downside exceeds ~20% from today's price, no matter how good the growth story — you only revisit after a big drop changes the math.
Here: the 12–15 highest-quality semiconductors (incl. MU) screen to ~40% downside in just a typical bear market — double the 20% ceiling — so the whole group is a pass; they even "trade down after stellar earnings" as investors debate normalized earnings 2–4 years out.
Watch for
- A candidate's modeled price in a routine 20–25% index decline; auto-reject if that implies >20% downside from here.
18:25 2. Base-effects inflation math — forecast the print, then the risk regime
The repeatable method
- Locate the one-off spike sitting in the year-ago comparison (here: the March oil spike). Inflation is measured year-over-year, so a high base a year ago mechanically lowers next year's reported rate — if the driver doesn't re-spike.
- Make the near-term call conditional on that driver: ceasefire holds → CPI can fall toward ~2% by next spring; oil stays high → stickier, and the Fed's path gets harder.
- Then invert it a year out: once you've cycled past this year's high-growth/high-inflation first half, next year's comparisons make growth look decelerating and inflation falling — a poor backdrop for risk assets even if nominal levels are fine.
Here: base effects off the March oil spike drive the ~2% "by March/April" call — and the same math, run into 2027, is the core of the "potential for a deeper decline next year."
Watch for
- One-off spikes in the trailing 12-month window; the conditional driver (oil) that decides whether the base effect actually shows up; the 12-months-later inversion as a risk-asset warning.
27:00 3. Map the momentum-rotation destination and get there first
The repeatable method
- Start from the structural fact that ~60% of trading is momentum players (retail + quants) who must stay fully invested — when they sell one trade, the cash has to land somewhere.
- Identify what they've been selling for months (the unloved sectors) — that's the pool of forgotten names cheapening while the crowd chases the hot trade.
- Accumulate those names incrementally now; when momentum flips, the rotation destination is "whatever they haven't been buying," and you're already positioned.
Here: semis-selling money is rotating into financials, healthcare, utilities and staples — the areas Oxbow has been adding to on weakness, so it's "already there" with the positions it wants (
24:52).
Watch for
- Which sectors have been persistently sold while one trade dominates; buy them before the momentum flip, not after the first up-week.
15:40 4. Separate the spike from the "raised floor"
The repeatable method
- When a geopolitical shock spikes a commodity overnight on thin liquidity, treat the spike itself as un-keepable — recalibrate/trim, don't chase it.
- But ask whether the shock has permanently raised the floor: recurring supply risk adds a durable premium to the commodity even after the spike fades.
- Re-underwrite the equities at the new normalized price, not the spike: if a "reasonable" price still funds strong cash flow, the stocks are as attractive as before the shock even though the commodity is higher.
Here: Brent gapped over $120 overnight (trim that) but geopolitics "raise the floor," so oil at $70–$80 leaves the E&Ps as attractive as at pre-war $60 — same enthusiasm, higher floor.
Watch for
- Overnight, low-liquidity commodity spikes (recalibrate); a structurally higher post-shock floor as the reason to keep owning the producers.
13:16 5. Build the asymmetry in advance — don't trade the "war default"
The repeatable method
- Don't keep a reflexive "if X happens, buy Y" playbook — markets move so fast on the news that you miss the initial move anyway.
- Instead, hold a balanced portfolio already tilted toward good risk/reward, so that whatever triggers the move, you benefit without having predicted the catalyst.
- Judge each position on upside-vs-downside asymmetry, not on a guess about the event — then let the surprise work in your favor.
Here: owning energy at a ~$60 oil "good risk/reward" before the war meant Oxbow participated when the Iran war took oil from $70 to over $110 — without having forecast the war.
Watch for
- Positions with asymmetric payoff you'd be happy to own regardless of the catalyst; resist the urge to chase the post-headline move.
23:30 6. The free-cash-flow disqualifier — use the bull report's own admission
The repeatable method
- For a story stock with an unbounded vision, put the vision in the "too-hard pile" — accept you have no edge valuing it — and instead find the one hard fact that settles the decision.
- Read the bullish sell-side report and pull its own cash-flow timeline. If the bull admits no free cash flow for many years, that alone disqualifies it for a cash-flow-focused mandate.
- Require a business that generates free cash flow and returns it (dividends/coupons); if the timeline to any FCF is a decade out, pass and let speculators have it.
Here: on SpaceX ("worth more than the entire Earth"), Chanos surfaced a bullish report whose analyst admitted no FCF until at least 2035 — an automatic pass for Oxbow.
Watch for
- The bull case's own FCF start date; a decade-plus timeline to any cash generation as the single disqualifying fact.
16:15 7. Express a commodity theme through cash-generating businesses
The repeatable method
- Decide, per commodity, whether to own the thing itself or the businesses exposed to it. Metals: own the metal + miners/royalties. Oil/gas: own the operating companies, not the barrel.
- Spread across the value chain to diversify the driver: E&Ps (price leverage), integrateds with refining (a margin tailwind), pipelines (toll-like income), and oil services (activity/repair cycle).
- Favor management teams that now return cash (dividends/buybacks) rather than reinvest into oversupply — the "smarter than decades past" capital-discipline test.
Here: oil is owned via a mix — E&Ps, integrateds with refining, pipelines, and a newer oil-services sleeve Ted built — betting on repair/more drilling ahead; contrast with gold/silver, owned as the metal plus miners/royalties.
Watch for
- Which vehicle fits each commodity; value-chain spread so no single sub-driver dominates; capital-return discipline as the quality filter.