Actionable insights — SoH Crisis Takeaways

1. The surprise / non-surprise scorecard — separate outcomes from priors

1. After a shock, list what actually happened, then bucket each item as a surprise (violated your prior) or a non-surprise (confirmed a known tendency). Force the distinction — it stops you from re-narrating the event to fit the position you already had.
2. For every "surprise," name the specific mechanism you missed and ask whether it's sustainable or temporary — a durable regime change vs a one-off drawdown.
3. For every "non-surprise," don't act on it as news; it was already in expectations, so its only value is confirming the base case.

• Any post-crisis narrative that only confirms your prior — that's the tell you haven't actually run the scorecard.

2. Find the hidden marginal actor — not just the obvious swing producers

1. Decompose the headline supply/demand shock into its pieces and attribute each to a specific actor (redirected exports, SPR draws, demand destruction).
2. The pieces everyone already models (OPEC/Saudi policy, US shale response) carry no information. Hunt for the actor whose behavior is not in consensus — here the demand-side buffer, not a producer.
3. Once identified, test durability: is it drawing a finite stockpile (temporary) or a structural change? A finite buffer means the moderating force reverses later.

• China's import run-rate returning toward 12 mb/d (the buffer exhausting); any new demand-side actor quietly absorbing or dumping barrels.

3. "Do the opposite" — read the term structure as sentiment, not signal

1. Anchor on the long end: 60-month-forward crude has held a stable $65–$70 band. Treat that as the structural value; the front end is where emotion lives.
2. Measure front-to-back spread. Steep backwardation (month 1 well above month 60, here $25–$30 at the peak) is a stress reading — a reason to fade bullishness, not chase it. Slight contango is the opposite — don't turn maximally bearish into it.
3. Apply George Costanza: at the extremes, lean against the crowd. This is psychological discipline, not a trade signal — size and timing are separate.

• The 1-to-60-month spread hitting an extreme (either steep backwardation or unusual contango) while commentators extrapolate that extreme forward — the contrarian entry to the psychology, not necessarily the price.

4. The corporate returns standard — judge the business, not the price forecast

1. Stop trying to guess where oil settles. Instead set a through-cycle bar for the company: it should earn a decent return at a normal trough and never lose money at a deep trough.
2. Sum the full cycle of ups and downs and require a mid-teens cash-on-cash return or better, plus competitive per-share growth.
3. A name that clears that bar is ownable regardless of the near-term price path; one that needs a high oil price to work fails the test.

• Balance-sheet break-evens and trough free-cash-flow; per-share (not headline) production/reserve growth as the durability check.

5. Re-map the risk premium — find which link in the chain actually gets hit

1. Don't assume "geopolitical turmoil = bullish crude." Trace the specific disruption to the specific segment: upstream fields, shipping/choke-points, refining, or LNG.
2. Ask where a sustained (not one-off) disruption can land, and cross it against each segment's supply-growth outlook. A segment that is both attackable and starved of new capacity carries the real, persistent premium.
3. Distinguish concentrated single-point risk (one hub) from a growth market that is diversifying its supply base — the latter self-heals, the former doesn't.

• Drone/standoff-weapon strikes shifting from fields to downstream infrastructure; segments with attackable choke-points and stalled capacity additions (refining) vs those with heavy diversified CAPEX (LNG).