Muni Will Never Make 85 Percent On-Time Goal

Exactly. Muni isn't running an airline here (though its service is far less reliable than even the worst airlines) and it's not as if anyone actually uses a bus schedule to plan their trips. Have you ever actually seen a Muni timetable? Has a Muni operator ever seen one? It doesn't matter one bit to most riders whether the bus comes at 8:22 or 8:32 as long as they know that buses come frequently and reliably. With a system like this, you don't have to worry about buses arriving at timepoints early, so you can actively manage schedules based on true vehicle positions instead of desperately clinging to a dream of a schedule.

For example, consider this scenario under the current system. You're waiting for, say, the 38L at Geary & Presidio. According to the timetable (hah!), a bus is supposed to come at 8:22am. If it happens to be four minutes late and arrives at 8:26, that's still considered on-time to Muni. The next bus is scheduled for 8:27, but let's assume it's right behind the first one as usual and also shows up at 8:26. You run for it, but it closes its doors just as you arrive. The driver is still stopped at a red light, but of course you're not getting on. The next bus after that is scheduled for 8:33, but it's ten minutes late and doesn't show up until 8:43. However, it's packed full of riders because it's the first bus anyone has seen in a while and you can't even get on. Now you've been to waiting 20+ minutes for the bus on a route that's scheduled for 5-6 minute frequency during rush hour (so your average wait should only be three minutes or so). The kicker is that only one of these buses is actually considered "late" to Muni and recorded in the statistics, despite this arrangement being completely useless to riders.

As a rider, I know that's what would be most useful to me. I can't realistically rely on Muni because I know all too well that the next bus might be five minutes away, but it could just as easily be 45 minutes or more. I'd happily accept slightly longer waits if I knew the service was reliable enough to depend on. It's darn hard to encourage people to take transit when there's a significant likelihood of being stranded in the rain for 45 minutes at the bottom of a big hill to get home at the end of a long day.

Of course, knowing how the SFMTA tends to operate, adopting a headway-based model for service performance goals would probably just lead to massive amounts of overtime, operators taking endless breaks in the name of "maintaining headways," management incapable of measuring the positions and speeds of vehicles, and a complete inability for anybody to make any changes that would actually improve performance. Oh wait; that's what we've got now. What do we have to lose?

I'm glad someone finally said it.

Vehicles coming along at consistent and reliable times almost sounds like a schedule to me. Although, I'd be all for it, just don't quite understand what the difference is.