I stopped using the BART recently because it's dangerous. Several people I know personally have been held up at gunpoint on the BART, or in the station. That's not counting the various news reports, like one [1]. I'll go back when it's safe, but not before then.

[1]: http://www.sfgate.com/crime/article/BART-takeover-robbery-50...

> I'll go back when it's safe, but not before then.

BART is already exceedingly safe. Even with robberies rising 45% this year the overall robbery rate is projected to be 1.7 per million trips[1].

To put it another way, if every time you rode the BART you also bought a CA lottery scratch-card you'd be about as likely to get robbed as you would to win the jackpot (normalized odds for a $2 scratch card: 0.00000167[2], normalized odds for robbery 0.0000017).

Do I think BART is a great system? Absolutely not. Do I prefer taking a Lyft to the airport rather than the BART? Yes. But I've worked in cities with unsafe public transit systems, and the BART simply doesn't compare.

[1]: http://www.sfgate.com/crime/article/Crime-stats-show-where-s... [2]: http://www.calottery.com/LotteryHome/play/scratchers-games/$...

That's interesting, because it is quite improbable that someone would "know personally" several lottery winners at ~590000 to 1 odds, even if his acquaintances routinely bought two tickets a day. GP's post is just a single datum, of course, but it's enough in itself to raise doubt about the accuracy of that 1.7/megatrip statistic.

Numerically, suppose GP knows 30 people who make 600 trips on BART each year. In one year, the chance of at least one robbery in that group, given the quoted odds, is three percent. The chance of multiple robberies, i.e. at least two, is under 0.05 percent – over 2000 to 1 odds against. Even over a ten year period, the odds are 25 to 1 against. If GP's personal experience is not unusual, then something is screwy with the stat.

> GP's post is just a single datum, of course, but it's enough in itself to raise doubt about the accuracy of that 1.7/megatrip statistic.

In itself it's not that great. You have two obvious options:

1. Robberies on BART are generally not reported. You'd need a very high rate of nonreporting to mess with the conclusion that overall BART trips are quite safe.

2. Certain BART trips, by location, time of day, or some other metric, are much more dangerous than the average, and the BART-using public preferentially avoids taking those trips.

If you believe in case 2, it's not at all surprising that someone might know several people who have been robbed on BART. You'd conclude that their social circle has atypical BART habits. How many robberies-at-gunpoint do you think occur in the packed rush hour trains?

Unfortunately many of these crimes will go unsolved because BART refuses to release videos. Apparently, releasing the videos “could lead to stereotypes”

https://www.google.com/amp/sanfrancisco.cbslocal.com/2017/07...

BART doesn't release the videos to the public. They do release them to the police. There are legitimate reasons not to release them to the public, including the fact that the media will sensationalize them to make BART look far more dangerous than the statistics actually reflect.

Replace [BART] with Police and [police] with [congress]:

"The Police doesn't release the videos to the public. They do release them to Congress. There are legitimate reasons not to release them to the public, including the fact that the media will sensationalize them to make the Police look far more dangerous than the statistics actually reflect."

Would you still want videos withheld from the public?

After all, BART is a public agency. We are not babies who need to be protected by The Master Director because we are "the lowly citizen." That's not how it works in America.

Videos of what?

If the police have videos of X person robbing Y person, with no police around, then I don't mind if they keep those videos private.

It's not about whether you mind or not.

It's about whether Y person will get justice.

Let me put it differently:

Police actions need to be transparent. There are strong reasons to release recordings with police actions in them.

But if the recordings are unrelated to any police officers or police actions, then all of that goes away. It's completely fine for police to not release footage of two random people interacting. And that scenario is what is analogous to BART surveillance videos.

  two random people interacting

That's an odd way to paraphrase the pairing of a felony perpetrator and a felony victim.

It's in the context of the previous comments, "X person robbing Y person". I didn't want to be redundant. No matter what crimes are happening, there are no police involved, so there is no police transparency motive to release the video.

The motive is to provide the public with information. Some of this information may be useful to identify the (gang of) attackers.

That's a perfectly fine motive, but it's not enough to force anyone that has such a video to release it completely publicly. And BART in this case is basically an "anyone". They have no particular involvement.

Without AMP: https://sanfrancisco.cbslocal.com/2017/07/09/bart-withholdin...

It would be nice if a community fixed its problems, whether systemic or otherwise, rather than hiding facts in the name of equality. Citizens should demand as much.

What third-world country do you live in that has such unsafe public transport?

San Francisco

Or if we're a little more honest public transportation is an Inferior Good [1] that people stop using as soon as they have the means to.

[1] https://en.wikipedia.org/wiki/Inferior_good

Not when it works well. I live about a mile from work. My kid goes to school perhaps 2mi from home. Mass transit has made it very easy for my family of three to downsize to one car. (We are not means-limited. It's fantastic to not have the extra crap to deal with from a second car.) When there's a thunderstorm or excessive ice, or I need to get home faster for whatever reason, the bus is great. But: I happen to live right off of a major bus corridor, so there's a bus every five minutes or so.

We use Lyft to fill in the gaps if something comes up that requires both adults to drive somewhere.

Mass transit plus a little bit of easily-available ride sharing _can_ be significantly better than car ownership. But...

It is an inferior good in that usage decreases as income rises. But in some locations, mass transit (especially subway) can be a pretty good alternative to congested surface streets--especially outside of rush hour, not carrying luggage/groceries/etc.

This is one industry where "build it and they will come" will most probably hold true. Especially in dense areas.

Not to mention safe and clean. In bigger cities, public transportation can put you in physical danger and have you walking out smelling worse than Theon Greyjoy

More people die in automobile crashes than from train crashes and crimes in public transit systems.

This doesn't mean anything. Although many more die from automobile crashes doesn't make using public transportation in some major cities inherently dangerous or unpleasant - even unhealthy.

Being on a busy road surrounded by cars all burning petrol is not exactly a healthy place to be either.

http://www.bbc.co.uk/news/health-35504768

At least I rarely have to fear for my life while driving through a bad part of town, which is definitely not the case while using public transportation or walking. Which is my main point, health was a side note.

>I rarely have to fear for my life while driving through a >bad part of town >which is definitely not the case while using public >transportation Source? Are you sure that this perceived difference in safety in 'the bad part of town' is not completely in your mind?

thousands of people take public transit through 'the bad parts of town' in places like Chicago and NYC every day and they (relatively speaking) don't have to fear for their lives either.

I understand this is purely anecdotal, but every time I'm driven through Compton, I've never had a problem. Almost every time I have to walk down Central, even to grab a drink from the gas station, someone see's me, a smaller white guy, and it always ends with me waving my Kershaw.

Also the reason why I moved down to Orange County, yet I still somehow landed in one of the worst neighborhoods in Anaheim.

And let me tell you this, we see the highest amount of NO2 pollution in traffic in the inside of cars. A good filter may help you, but many many cars don't have them.

The statistics don’t really tell that story.

Getting mugged in a station doesn’t typically kill you, and you’re not travelling, so those stats won’t be counted.

If there was an “incidents per mile travelled”, metric for my trip during daylight hours that would be one thing. But there isn’t any such statistic.

It’s almost certainly cheaper, safer, more time efficient and dramatically more convenient to take a car. In my personal, typical case, I live and work in the same city. Drive/bike/bus/walk time is 10/25/45/90 minutes, respectively. Cost if daily commute assuming IRS costs is about 20% more for driving, but with shorter more predictable and more flexible trips.

> The statistics don’t really tell that story.

> In my personal, typical case

They may not tell that story in this particular case, but unless something is wrong with the way with the statistics are gathered, it does present a better top-level picture than personal anecdotes. (While personal anecdotes can also be useful in understanding how to design around edge cases, but are not useful when figuring out the best way to design a system).

Highway safety statistics have significant limitations with respect to fine grain detail and data collection capability. Any datasource that uses police data must account for jurisdictional differences in data collection as well.

Nothing is wrong with the statistics, it's how you use them. It's just difficult to meaningfully say that it's safer to take the bus from point A to point B using statistics that are generalized over time and geography.

For example, 30% of motor vehicle fatalities are alcohol related, and the vast majority of those fatalities take place after most people are done commuting for the day. Uber probably has a significant impact on preserving human life, after dark.

The purpose of the anecdote is to illustrate how those stats relate to decision making and risk/reward management by the travelling public. Am I at more risk driving three exits down the highway vs. crossing a busy avenue, sitting at a bus stop, riding a bus, crossing three other busy streets? I don't know -- and those statistics that you are citing don't answer that question. My guess is that I'm safer in my late-model car in a typical automobile accident than in any traffic or crime incident in the bus case.

They still have an important point about what statistics to use. Total death count is useless. You want numbers for per-mile and per-hour, and you want to look at more than just deaths.

Yes, but that doesn't actually prove anything about the safety aspects of switching from public transit to a taxi-like service. The comparison would have to be made like-for-like trips.

However, the risks for both are quite small, and that's why the quite big gap in convenience dominates the equation, anyway.

> efficient

Specifically, it should be efficient and preferably fast for me, not just in aggregate

This can be represented by placing emphasis in 95th percentile stats, similar to how sites should analyse 95th percentile request handling

It can be fast for everyone else, because then you'll have all those empty roads for yourself.