The seductive story, and why it is the dangerous one
The received wisdom on geopolitical shocks is tidy: markets sell on the headline, panic overshoots, and the disciplined buyer picks up quality at a discount while everyone else is frightened. It is a good story. It is also the single most survivorship-biased idea in trading.
We remember the sell-offs that recovered because those are the ones that became anecdotes. The events where the discount was not a discount — where the market had correctly priced a structural change and kept going — do not get retold, because "I bought the dip and it kept dipping" is not a story anyone repeats.
Why the pattern is not a rule
The usual framing says: immediate sell-off, dip-buying within a day or two, real repricing inside a week. That does describe many events. The problem is that it describes them afterwards, and the sample is tiny.
Genuine geopolitical shocks are rare, each is structurally different, and there is no meaningful statistical base to generalise from — a handful of events across decades, in different rate environments, with different starting valuations. A pattern derived from a dozen non-comparable observations is a narrative, not an edge. Anyone quoting a win rate for trading geopolitical events is quoting a number they cannot have.
The distinction that actually matters
Not "is this an overreaction" but "does this change cash flows".
A conflict that disrupts a shipping lane changes input costs for identifiable companies, durably. An election result that shifts regulatory expectations changes the discount rate applied to a sector. Those are repricings, and they should not revert.
A headline that is frightening but leaves earnings untouched is a sentiment event, and those do tend to fade. The work is telling which one you are looking at, and that work is specific to the companies you hold rather than to the event.
Where sentiment analysis is at its worst
Worth saying plainly, because it is our own tool: this is the scenario where a sentiment reading is least useful.
Sentiment measures the volume and tone of discussion. Geopolitical events generate enormous discussion almost entirely independent of their market impact — a distant conflict with no earnings consequence can dominate every feed for a week. The signal spikes, the fundamentals do not move, and a tool that reads text cannot tell those apart.
During these events, treat a sentiment shift as a measure of how loud the news is, not of how much it matters. The reasoning attached to a reading names its event count and sources, which is the part to read — forty events from one wire story is a different thing from forty from forty.
What to actually do
- Decide before the next one happens what you would do. Policies made under stress are not policies.
- Ask the cash-flow question for each position you hold, not for the index.
- Size any response smaller than feels right. The defining feature of these events is that the information is incomplete and arriving fast.
- Accept that doing nothing is a position, and across a small sample of unrepeatable events it is frequently the best one.
The honest summary: geopolitical trading is where confidence most outruns evidence. The traders who come through these events well are usually the ones who had already decided their exposure, not the ones who reacted cleverly.
Nothing here is financial advice. Geopolitical events are not statistically tractable and past market reactions do not indicate future ones.