Understand that forecasts are based pn past performance. You get the first few years of a good PDO and you'll underestimate as conditions are better than historic. Then, was conditions decline you overestimate. Couple this with occasional blob...

Further, the impact of the AK pink situation adds an even-odd component (which actually may have affected survival somewhat back when there were on wild pinks). Also note the steep decline in hatchery fish which places more pressure on the wild fish in either consumptive or release fisheries.

I harp back to a model developed in the OPI that very accurately accounted for ocean survival and estimated return to the coast. Vet accurate. It used 4 separate ocean metrics. There were two problems. The first was that the last metric was available just before the fish hit the coast; too late for the ocean mixed-stock fishery. The second, but which we should all remember, is that any one of the metrics could drive overall returns. If #4 was good it would compensate for poor 1-3 with the opposite just as possible.

Fish terminally on known (ie updated) numbers or develop and implement oceanic in-season updates and modify accordingly.