By the time this is published, the deadline at Los Angeles and Long Beach has passed, one way or the other. That is the point. On October 27, gCaptain reported that about 250 port watchmen represented by ILWU Local 26, about 2 percent of the union's workforce at the 2 ports, had issued a strike notice covering 8 terminal locations, with the current contract extension expiring on November 1, and that the terminals involved handle about 85 percent of Long Beach's cargo and 33 percent of Los Angeles's. On October 29, SupplyChainBrain reported that a strike could begin at 5 p.m. Pacific time on November 1, and that 35,000 containers were already waiting for rail loading across the 2 ports as of October 28.
Every operator with cargo on the water toward San Pedro Bay faced the same structure: a binary event, a known date, an unknown probability, and a set of hedges whose lead times were longer than the time left. Whatever happened on Sunday evening, the decisions that mattered were made before anyone knew the answer.
The breakeven probability is the only forecast that matters
Nobody can forecast a negotiation. Nobody needs to. The decision needs 1 number: the probability at which hedging becomes cheaper than waiting.
Take an illustrative vessel 5 days out with 4,000 import containers for the affected terminals. Diverting to another gateway costs, say, $400,000 in extra steaming, terminal fees, and inland repositioning. Arriving into a stopped terminal costs demurrage, detention, missed rail connections, and customer penalties: call it $125 per container per day for 5 days, or $2.5 million. The hedge pays if the probability of a strike exceeds 400,000 divided by 2,500,000, which is 16 percent.
That is the whole analysis. At 16 percent you divert, even if the most likely outcome is that the strike is averted. The hedge was not a bet that the strike would happen. It was a premium paid against a large tail.
Options expire before information arrives
The second structural feature is timing. The vessel 5 days out must decide on day 0, when the information is worst. The container already on dock has a different menu: pull it to an off-dock yard now, or leave it and accept the rail queue.
Each unit of cargo has a last moment at which each hedge is still available, and that moment almost always arrives before the deadline. A plan built around the deadline is a plan built around the 1 date on which nothing can be done. The useful schedule is the list of decision deadlines, unit by unit, each with its own breakeven.
Criticality is positional, not proportional
The notice came from 2 percent of a workforce and reached more than half of a complex's throughput. That is not unusual. In the networks we have modeled, a small number of nodes, roles, or handoffs gate a disproportionate share of flow, and they are rarely the ones the organization chart or the payroll would suggest.
That map should exist in advance: for each small element, what fraction of flow stops if it stops, and what the alternatives cost. The arithmetic above cannot be run on a Friday if the dependency is discovered on the Friday.
Score the decision, not the outcome
Whichever way it went on November 1, the operators who diverted at a 16 percent breakeven and a 40 percent likelihood were right, and the operators who waited and were spared were lucky. Over a year of such events, the first group pays premiums and avoids the catastrophes. The second group saves premiums until the year it does not.
The queue of 35,000 containers makes the arithmetic sharper, not different. A disruption that lands on a loaded system is recovered from slowly, because the backlog compounds behind it. The strike notice was the headline. The queue was the exposure.