Yet we know that even slower than car conventional intercity rail is not restricted to just “those with no alternative”, since increases in service frequency and service reliability result in increase in ridership. So clearly there is a spectrum.

1. Average load factors of 2.2 are not achieved by loads all clustering between 2 and 2.4 per car, but are achieved by one and two and three and four and five occupant cars.

2a. For those who do not enjoy driving, and in particular those who prefer visual or audiovisual entertainment to audio-only entertainment, several hours driving is several hours that could be used to relax.

2b. For business travelers, two or three hours spent driving is two or three hours that could be used more productively if the employee was free to concentrate on work.

3. The traveler may want to use the car at the end of the trip, but the traveler may also find the car at the end of the trip to be nothing but another chore in terms of finding a safe and secure parking place in an unfamiliar area.

4. Some car trips can be made at the spur of the moment, but others require advance preparation, in some cases when when traveling with children, in other cases in order to book a company car for the trip.

5. It takes some time to find a place to park the car on arrival at the final destination. The parking place hunting time can cancel the access/egress time advantage of cars over HSR, especially when traveling to sites with convenient transit connections.

6. The idea that its not possible to get food or to walk around to stretch your legs on a HSR trip without stopping the train is hilarious. At least the first five points were relevant to transport choices for some people – the fact that the car has to be stopped to get food or stretch legs (and indeed ought to be stopped to stretch your legs every two hours to avoid an increased risk of traffic fatality) is a clear benefit for HSR across the board.

Rather than the fantasy of there being a “speed threshold”, the reality, starting at speeds below time-competitive with cars, is that there is a continuum, with quicker trips on trains being an advantage if a train is brought from a time-disadvantage to cars to being time-competitive with cars, to having a time-advantage over cars, to the point of being time-competitive with flying, to the point of having a net-time-advantage over flying (when the time overheads connected with flying are included).

The thresholds as opposed to the continuous responses occur as specific destination/origins enter new market segments. For example, over three hours, the ability to compete for same-day trips starts to fade. That, of course, interacts with frequency, since an actual arrival before the start time and actual departure after the end time is required to compete for a specific same-day trip.

While intercity transport demand has higher and lower levels at different times of day, commuter demand has a stronger peak, which means that you get to buy more trains for the same number of riders, because you have to have enough trains to provide seats for the people riding during the commuter peak.

Also, commuter demand tends to be for shorter trips than intercity transport demand. So focusing on commuter demand implies more seats have to be empty for a longer part of the in-bound commute in order to be available for the closer commuters, and more seats will empty early and remain empty on the out-bound commute.

Normal commuter rail services often address this by having standing room, so that the seated capacity can be added to substantially by standing room. But since the HSR will be designed for intercity transport, they will be all-seated trains, and can’t use that.

That is why all around the world, HSR service providers often choose to price commuter-distance trips at a premium price, in order to ensure that the shorter trip tickets that are sold pay a sufficient amount for the smaller number of passenger-miles that they provide in order to compensate for the reduction in average load factor.

And, why this is wrong:

In any sane rational world view, you want to maximize paying asses filling the seats of your trains. The more of them there are, the more successful (by any measure) your service is and the better the economics are.

… in a sane, rational world, intercity rail services are concerned about the passenger miles of transport that they are providing, and not just about the number of people buying a ticket.

As stated, short distance HSR piggybacks on top of longer-distance service. Frankfurt-Cologne is a great example – that line was so expensive nobody would have built it just to service that city pair. But it is heavily used by short-distance travellers just the same. I used to take the train over that segment weekly (from Stuttgart to Duisburg) – and the vast majority of those on the train were travelling between those two cities: the train would fill up in Frankfurt and empty out again in Cologne. [sorry can't work up any numbers from the net, other than the 11 mio annual on that segment]

For short-distance HSR to be attractive, access time is crucial. Since most German cities are very dense and with good transit to rail stations and difficult auto accessibility to the center city, short distance HSR is a hit there (I used to regularly get from my apartment in Stuttgart to my friend’s in Frankfurt [about 130 miles] in under two hours – just about half an hour faster than driving). Bringing this back to the CA context, HSR from Sacramento to downtown SF would be a big winner because of the local transit and density of SF, whereas HSR from Sacramento to Silicon Valley would be more variable and depend on future development patterns.