Thanks for the level-headed view, Bob.

Shaking from local quakes is more likely to cause RR damage in SoCal than something in Cascadia, as you note. In 1971, freeway bridges on I-5 did fall on the now-Metrolink Palmdale line, blocking it for several days (also some other track damage near the ground rupture, which was fairly easily and quickly fixed). The tunnel between the Valley and Santa Clarita was apparently not affected by either the 1971 or the 1994 quakes despite the freeway interchange on top of it falling over.

Another potential source of locally-generated tsunami in SoCal is the collection of faults in the offshore area south of Palos Verdes - several spots where compression or extension exists that could cause vertical movements and tsunami even in a moderate quake per some recent studies. In any case, as you note, locally-generated tsunami are likely to cause major but not spectacular damage. I wouldn't want to be on a Surfliner near San Clemente when it happens, though.

For more information it might be worth looking at SoCal effects of the 1964 Alaska tsunami. That's probably documented somewhere at USGS (they have tons of reports on that quake, which was the largest ever recorded in N. America and 2nd or 3rd largest recorded, period). I vaguely recall that, in addition to the major damage and deaths at Crescent City, there was some damage in harbors around the Bay Area).