Because of the traffic and the hills, SF has always been hard on vehicles and vehicle systems. Think those brakes will last 100K miles on your car used mostly in town? Think again. But ever since LRVs replaced streetcars it seems to have gotten worse on the rail side. Political decisions (such as forced purchase of the Boeing cars rather than Siemens/DuWag, and probable shenanigans in the Breda deal) haven't helped matters.

Sacramento looks like it comes out smelling like a rose, but it really doesn't if you look at the passenger miles. And Sac has the advantage of a fleet of old Siemens cars that are still the most reliable things they have at almost 30 years old (except for the a/c that was added on and can't handle summer temps), a fleet of modern CNG buses that are regularly replaced and serviced, and probably lower unit staff costs than SF. And Sac has NO HILLS and little stop/go traffic affecting bus & rail lines. The other cities in the study that look much better than SF are also much lighter-duty environments. Houston? Give me a break - one short LR line and a few social service bus lines - everybody drives.

You also have to note the definitions (yes, I did read TFA and also the audit report linked from there). The 617 miles is ALL FAILURES including those that cause a vehicle to be taken out of service by policy, not because it stops running. The policy part can get interesting - an area might get a good total failures number by reducing policy requirements for taking a car out of service due to some minor failures. That said, SF also has the highest major failure rate at about 2500 miles per - not a good number in a place that has the most vehicles running and the most passenger miles carried, resulting in the probability that there will be one or more failures taking a train or car out of service every day. The SF ratio of total failures (including policy-related) to major actually seems to be a bit less than many of the others, though.

Who else has the Breda cars? Somebody in the article mentioned LA Gold Line - confirmation please? What's their failure rate like? How many of SF's failures are with the stair mechanism, which is unique to SF, and how many are for other reasons that are common to the comparison group? The audit report didn't discuss that. It's clear in Sac, however, that the newer (CAF) cars not initially as reliable as the old-tech Siemens, and even after 10 years of operation and debugging are still more fragile especially in terms of door issues and high-temp electrical failures (in an area where summer temps often run between 100-110F in the PM peak; SF doesn't have to worry about that until global warming gets much worse). I suspect that's the case in most places with newer cars and limited maintenance budgets. Though I have to admit that the only train I was on in Sac that ever caught fire was an old Siemens - I guess that would count as a major failure. How often do Bredas catch fire?

Finally - why are the interminable Muni problems an argument for a VLF? More money - probably, yes, if it can be guaranteed to go to reliability improvements and not just higher driver salaries. But transit is primarily a social service operation - even, technically, in SF where people actually use it (sometimes) in preference to owning a car even though they could easily afford one - so it makes more sense to me to use a broader-based tax such as a sales tax bump or something attached to the property tax (allowed to be reflected in rents).