Uber is experimenting with letting riders wait longer for a cheaper fare
qz.com
qz.com
I wonder if this is the first step towards offering a "limit" option. For example, where I live the average price from our house (Jersey City) to the mall is $7, but some days it jumps up to $15 or more. My wife and I usually choose to walk those days, but it would be nice to say "hey we want to go to the mall today, let's set a $7 limit and when that hits we'll go." For the flexible consumer it's easy to wait for a better price and not need to constantly look at the phone to see what the current market price is. The limit price could help create a queue which could provide a driver with a continuous supply of rides during non-peak hours.
What's more frustrating is when you are on the other side of a busy street, you wave to the driver or even call them to let them know you're on the other side of the street, and they just sit there...
What phone are you using?
In a decade of living in Manhattan, I've never had any issue like this. The location services on my phone (Pixel 2 XL) and every Android phone I've ever used have been spot on.
Location Services doesn't just use raw GPS, no, but that's what most drivers (and passengers) are using.
In any case, for several years, I only used the device GPS (instead of the full location services), and I never had any issues like what OP is describing.
I have no idea if that's true and have always been confused about whether the driver app even has maps built into it.
The Uber native maps app is almost invariably incorrect, at least here in Brooklyn. It does insane stuff like trying to put you on the BQE (interstate) to go like 800 yards or route you from Brooklyn to Brooklyn by transiting Manhattan and two bridges.
Until they figure out how to fix that they pretty much need to let drivers flip over to Waze or something.
That's unfortunate. As I said, every Android phone I had has been spot on everywhere in the city.
Can anyone recommend a carrier with better QOS here or is this just inherent to my new life in a giant city? I'll get over it and adapt, just keeping my options open.
Of course, over-subscription is a thing, so maybe things degraded?
If driver comes on the other side of street, maybe cross the street.
A tiny bit of effort goes a long way...
Whenever I called a rideshare, I very explicitly set the needle to exactly where I was. It's amazing how frequently I'd have a driver park across the street and a block away, and get angry when I didn't come to them.
I want absolutely cheap but non-shared point-to-point rides.
Beyond some very minimal, basic safety and competency as a driver, I stop caring and just want it to be as cheap as possible.
If I could indicate that I’ll happily be paired with a low rating driver if it is cheaper, I would.
And it sounds like adding this functionality could satisfy both of our use cases too!
Your comment rings sanctimonious to me.
While I agree that major airlines should do better in some areas, they are also just responding to consumer demand.
Should the people who cannot afford higher fares just exist in a lower caste where flying is not achievable? Those people for whom lower fares are the priority, you feel they just shouldn’t be able to attend a family member’s funeral or college graduation?
That is how things are in my extended family in the US. A spare couple hundred dollars for even a routine round-trip fair between major cities is still a very difficult stretch for them.
If grumpy people who can easily afford the flights are annoyed by the lack of frills and the cheapness, yet people who are less well off can possibly afford more flights, that seems like a positive overall for society to me.
Unfortunately in the airline industry we've tended toward an overall race to the bottom which explicitly budget airlines and different classes of service only help to a limited degree to separate out people with different preferences.
What it sounds like you are saying is that there is a group of middle class or upper middle class people who won’t choose to spend the amounts required for the luxury option (the equivalent of not taking the bus), and yet feel grumpy that the slightly higher prices they would prefer to pay end up pricing almost all flights out of the market for other middle class or lower middle class travelers.
“Cleanly separate the cheap vs comfortable use cases” seems simple: airline economics just implies that the prices for the comfortable (“not taking the bus”) option are staggeringly high, and that so many customers strongly prefer the cheap (“just treat the airplane like it’s the subway”) option, that airlines must structure the offerings around that fact.
That said, if you have Pre-check, airline status, airline club membership, maybe pay a bit more, etc. in practice you get a bit more legroom, early boarding, (usually) fairly fast security, etc. So, in that sense, there is a mild upgrade option relative to barebones cheapest tickets. (Depending on the airline.)
1. If the rider isn't sensitive about when they leave home, then they can happily wait until a driver who happens to have dropped someone off nearby is willing to take them for $7, because they don't need to drive more than 1 minute for the pickup.
2. Lease/finance, insurance, rent and food don't go up with the number of rides or distance. If a driver has the option to take someone right now, without driving far to pick them up, vs. waiting for a higher rate, then in some cases they'd maximise marginal profit by taking the 'low rate' ride.
they can happily wait until a driver who happens to have dropped someone off nearby is willing to take them for $7
(Of course if Uber were to introduce some 'limit price' feature, there would have to be some time-out/expiry (set by the system or by the rider). It doesn't make sense for a driver to pick up a rider days, weeks or months after the request was made.)
Cars that smell like cigarette (that's if the driver isn't outright smoking in your face), don't have AC, are old, dirty and with uncomfortable ripped-up seats.
Send your lower-quality drivers my way. I miss not wanting to vomit whenever I take a cab.
On my way back I used the Beat app. It's got a slightly wonky UI and the street names are in Greek, which I'm too ignorant to read, but for getting from the apartment to the airport it was great. Taxi was clean, driver was polite, fare was reasonable.
I don't doubt there are bad taxis, just want to point out there are good ones too.
4 years later, we have drivers that can't follow basic directions, that don't speak either English or Portuguese, that leave the turn-by-turn voice on, default to the terrible EDM stations (they used to default to neutral, low-key radio stations) and worst of all, drive very dangerously.
I've narrowly missed getting side-swiped more times than I can count, to the point where I'm seriously considering mounting a handlebar cam just to capture license plates.
So: if you're a Lyft or Uber driver, please please please please learn your obligations with respect to cyclists / pedestrians and associated infrastructure. (For instance: did you know that many jurisdictions have a mandatory minimum passing distance (usually 3ft / 1m), or that cyclists have "take the lane" rights in California? [1]) Otherwise, you shouldn't be on the road, and you definitely shouldn't be driving other people around for money.
[1] http://www.calbike.org/bicycling_in_california_sharing_the_r...
As a fellow cyclist, I don't think you've described any qualities that are specific to uber or lyft drivers, you've effectively described about 90% of the drivers on the road.
If I understand the math correctly, BART claims about 126 million rides (2015) and cost $906M (~$7 a ride, comparable to some uber pool fares)
This is because these are publicly funded organizations that are starved of funds for political reasons.
Public transit works well where it's funded well. It's a political problem, not a technology problem.
But that covers 100 percent of their expenses. If Uber had to shell out to fix the streets that they use, I bet it wouldn't be so cheap.
Streets are public infrastructure and are paid for by taxes, which Uber and all other companies that do business using roads presumably pay, just like everyone else. So, unless you have a specific reason to believe otherwise, it is fair to assume that yes, Uber is paying for the streets that they use.
If streets are not being adequately paid for, then that's a failure of the government that's responsible for them, not the businesses that are using them (streets are there to be used).
Uber doesn't pay a cent in gas taxes (they're not employees, they're partners, right?), and since they've never made money before, they don't pay a cent in taxes either.
It's a bit obvious you have a chip on your shoulder about Uber and ride share companies in general.
PS. I have an expensive piece of paper in economics and statistics, but I appreciate you telling me to be careful. Gotta have somebody looking out for me, i suppose.
There's a reason advanced cities in Europe and Asia invest heavily in mass transit without the expectation of a commercial return, it's simply more space efficient.
https://www.straitstimes.com/singapore/transport/government-...
...and of course, this being Singapore, the companies in question (SMRT Corp and SBS Transit) are themselves largely owned by various government entities (Temasek etc).
Comfortdelgro is not largely owned by the Singapore sovereign wealth fund, and I personally like its lines a bit better than the smrt lines.
It also manage the Copenhagen's subway
It makes profits
Scaling a mobile app is trivial compared to running a transit company.
BART owns/maintains infrastructure (including an underwater tunnel) and vehicles, provides benefits (health, retirement, etc) to their drivers, and operates a multi-jurisdictional police force.
Uber meanwhile hired a couple of tech bros (paid largely in stock instead of actual currency), pays their drivers near minimum wage with no benefits (as they're not actually employees), and disregards as many laws as possible (so there's little-to-no cost to enter a new market). It's not surprising they've been able to scale. What's surprising is how expensive their service still is.
What's your point with this?
This may be related
https://www.susanjfowler.com/blog/2017/2/19/reflecting-on-on...
Drivers aren't going to get more than a few percentage points increase without increasing the price to the end user.
But they should. Perhaps the pay scale is too top heavy. Uber's operational expenses should be trivial (hell, they straight up stole the most expensive bits -- the self-driving car IP). Instead they squeeze the drivers (passing along rate cuts) and riders (surge pricing during shootings, etc) pretty severely.
They can't easily scale these efforts back because the value of the service plummets for all parties if they lose their economy of scale.
L5 self-driving cars will be a game-changer for a while, assuming Uber gets exclusivity from Waymo (which now owns a stake in Uber).
I'm sure there's plenty of analysis of this out there, so I should read some articles.
Which is hilarious to me. Uber is spending a ton of money, and very little of it trickles down to the people that make Uber possible (the drivers).
(Not saying I agree or disagree.)
You seem to present that as a contradiction, but the obvious conclusion is that the GP believes Uber keeps an unfair share of the ride fees.
Based on my vague internet research, it appears Uber keeps 20-25% plus various fees, like how credit card settlement charges work, which can add up to 40%+ for some rides.
And the losses are so large that it likely means both raising prices to riders and reducing the net payout to drivers at the same time.
Any hot-take that centers around "It's just an app that a couple of guys wrote in an afternoon with some underpaid drivers doing all the work" is someone arguing in bad faith (and I suspect the person is well aware of it). There is a ton that goes into scaling a marketplace in general - let alone an on-demand one with thousands of corner cases, real time demand forecasting, real time fare calculations, real time capacity planning, customer support operations, etc.
https://en.wikipedia.org/wiki/Brian_McClendon
He's a great guy by the way, in addition to being super smart.
Your others points are not well thought through at all - I urge you to spend some time really thinking deeply about them if this is a topic you care about.
Every company drives around on roads they don't pay for. Roads are a benefit to society - part of the reason we value them is because they enable commerce. Do you complain that GrubHub doesn't use their own roads when they deliver your food? Do you complain that Pizza Hut doesn't own it's own cars when their delivery drivers bring you pizza? Does United help pay for the airports that enable them to exist?
Look, the fact of the matter is that consumers LOVE ride sharing services. They're fast, they're cheap, they're clean, they're safe, and the customer support is incredible. Ride sharing today is simply 10x better than any alternative in most cities. If you're going to argue for cabs being superior, then you either do not live in America or you've never taken a cab before. Not only that, but many drivers love ride-sharing too. Whole classes of people that couldn't find work before can now easily find work with Uber or Lyft (including the deaf, mute, retired, students etc). They can work when they want to, as much as they want to (or as little) and set their own schedule. This is simply not possible with any other low-wage work.
Having said all of this, I don't mean to imply that we should absolve Uber of responsibility for their business practices or their culture problems. We should absolutely push them, and the industry to be better. But this isn't just about Uber. This is about ride-sharing in general, which completely changed what transportation in cities means. It is absolutely innovative in every sense of the word.
Yeah, but when I get into a car it's not Uber Maps on the phone, it's Google Maps. So I'm not very concerned about a guy who worked at Uber for less than two years.
>He's a great guy by the way, in addition to being super smart.
I'm sure he is, which is why he got the hell out of there.
>Do you complain that GrubHub doesn't use their own roads when they deliver your food?
Yes, I would like for them to pay their fair share in road taxes.
>Do you complain that Pizza Hut doesn't own it's own cars when their delivery drivers bring you pizza?
...yes, but now we're being redundant.
>Does United help pay for the airports that enable them to exist?
Now this is just ignorant. Yes, large airports charge millions in landing fees. Train companies pay absolutely egregious property tax rates. It's just cars and trucks that get away with paying 18.4 cents per gallon. Oh, and me who has to pay taxes to fix the roads here while congestion and air pollution here in Chicago have absolutely skyrocketed.
>which completely changed what transportation in cities means.
Uber's success is basically because they were able to push all of the legacy costs of public transit onto taxpayers and claiming that they're more efficient. It's VC-funded bus routes but using a phone. DISRUPTION.
Navigation might be Google Maps or Waze, but everything else is Uber. My point is that Uber is 100% serious about mapping and they are investing a ton of resources into it. They're likely one of the few players who are doing this at scale, internationally.
> I'm sure he is, which is why he got the hell out of there.
Now you're projecting. He left Uber so that he can run for office in his hometown of Kansas City. I think he has a lot to contribute outside of technology. I got to know him because I went to a talk he gave once. And no I don't work for any of these companies.
> Taxes & Roads
Ultimately, everyone pays for roads. Not just drivers. And we pay for them because they're for the benefit of everyone. They enable services like GrubHub/Uber/UPS/Dominoes as well as commerce more generally - which in turn makes our lives better. It's a complex system, and I suspect that the tax revenue the government gets for roads is higher after the introduction of Uber/Lyft. With that being said, if you feel a tax is necessary then I respect that opinion. Even with a tax like you're suggesting, it won't change the demand curve meaningfully or affect Uber's profitability. It's likely serve to further entrench them if anything at this point.
With respect to your point about Airports, they're not even close to being paid for by the airlines in any meaningfully fair way. First of all, the government subsidizes it up front with billions of dollars, expensive zoning, planning resources etc. That's a huge capital risk that society takes on for the long-term benefit of everyone (you can't opt-out if you don't fly on planes). Yes, airlines pay usage taxes and some other fees that help pay for the airport but it stands to reason that the government won't recoup that kind of money in even a single decade. The government props up the airlines industry in large part because we demand it and it makes our lives better.
> Uber's success is basically because they were able to push all of the legacy costs of public transit onto taxpayers and claiming that they're more efficient. It's VC-funded bus routes but using a phone. DISRUPTION.
Absolutely wrong. I use Uber/Lyft every day. I don't have a car. Uber/Lyft are nothing like a bus route - not even close. To even suggest that they're similar to a traditional bus service route is to reveal that you haven't used the services in any meaningful way.
They are absolutely 100% disruptive (even more so if you're a woman and safety is important to you). I'm telling you, ride sharing has changed how I get around in cities, period. There's a reason drivers drive for these services. Ride sharing is creating a whole class of jobs that have a very important role to play in society. Stay-at-home moms that use Uber/Lyft to supplement their income while being able to have flexible work. Students that are just getting started in their career and are using Uber/Lyft to sustain themselves in the interim. Retired veterans who don't have any other way to earn a steady income. Deaf people who can't find jobs otherwise. Immigrants who are trying to build up a support base for themselves. The list goes on. Talk to the drivers, this kind of work is a big boon to a lot of people.
Now, should these be the only career options for people? Obviously not. I think that's where most people's concerns are valid. We can't create a society where this is the only type of work available. But can this type of work exist alongside other meaningful options? Of course!
So you're comparing scaling out a mapping back end to building and running massive transit infrastructure?
> With respect to your point about Airports, they're not even close to being paid for by the airlines in any meaningfully fair way.
Take a look at JFK. Airlines absolutely do pay significant amounts into airports.
> Absolutely wrong. I use Uber/Lyft every day. I don't have a car. Uber/Lyft are nothing like a bus route - not even close. To even suggest that they're similar to a traditional bus service route is to reveal that you haven't used the services in any meaningful way.
FWIW, BART doesn't run buses.
> Ride sharing is creating a whole class of jobs that have a very important role to play in society.
Minimum wage jobs with no benefits (thus externalizing the cost of providing the jobs)? That's not much of a win in my book.
> Now, should these be the only career options for people? Obviously not. I think that's where most people's concerns are valid. We can't create a society where this is the only type of work available. But can this type of work exist alongside other meaningful options? Of course!
And absolutely none of that means Uber is doing something more challenging than running a transit agency like BART or NYMTA. Uber's externalized most of their costs, disregarded most laws, and is scaling a very specific part of their software stack. It's not surprising to see growth under those circumstances.
Ah, I thought things like "God mode", Boobr, and doxxing critics felt creepy and unsafe. But that's just me, and I'm not a woman.
> There's a reason drivers drive for these services.
Sure. We like to call those predatory financing and exaggerated claims about compensation. Your buzzword laden defense of Uber misses the point though: scaling Uber is significantly easier than scaling a transit agency like an airline or BART.
Have you talked to actual women? Ask them if they feel safer in an Uber, in a Taxi or in BART. Seriously go try this experiment. Assuming you're in the Bay Area, ask 10 of your closest female friends.
Also remember that Uber is just one ride sharing company.
> Sure. We like to call those predatory financing and exaggerated claims about compensation. Your buzzword laden defense of Uber misses the point though: scaling Uber is significantly easier than scaling a transit agency like an airline or BART.
People are not stupid. If it was a raw deal for them then they'll jump ship (there are huge communities of drivers that constantly crunch numbers, share spreadsheets, etc). You can drop in and drop out at any time. That's one of the benefits of doing Uber. Work at McDonald's? Now you have to fight to get enough work hours for pay, you can be overworked, you can be shortchanged, you can have grueling schedules, you can't take vacations easily, the actual work is stressful....have you sincerely ever worked a job like that?
> Your buzzword laden defense of Uber misses the point though: scaling Uber is significantly easier than scaling a transit agency like an airline or BART.
I think you forget that the OP that prompted my response said "I genuinely don't know what their expenses are other than giving away shit for free", which is what I was responding to.
If you genuinely agree with the OP and you've made up your mind, then I don't think there's anything I can say to convince either of you that scaling Uber isn't a matter of giving away shit for free and building an app in a weekend.
You're welcome to believe what you want.
I am in the Bay Area, and the biggest proponent of taxis over Uber that I know is female. The only woman I know that uses Uber does so because her boyfriend is a code monkey at Uber and so they'd get a few rides for free each month.
Of my friends that use "ride sharing" apps, they almost all use Lyft exclusively because they find Uber so reprehensible. I also see plenty of vehicles signed for Lyft alone or Lyft + Uber, but almost never see Uber only vehicles.
As a woman how do you feel about Uber fighting background checks tooth and nail?
Or how about this?
http://time.com/5023287/uber-threatened-journalist-sarah-lac...
Or this?
https://www.nytimes.com/2017/06/15/technology/uber-india-rap...
Or this?
http://www.newsweek.com/uber-and-lyft-drivers-accused-more-1...
And, really, Kalanick using Uber for sex (a.k.a. Boober) doesn't bother you in the slightest? Good for you then.
> People are not stupid. If it was a raw deal for them then they'll jump ship
https://www.recode.net/2017/2/28/14766964/video-uber-travis-...
So this guy is an idiot then? How exactly do you get out of an auto loan that you're now underwater on because Uber decided to turn the screws on you?
> I think you forget that the OP that prompted my response
I am the author of the comment you replied to. I'm happy to be convinced but marketing doublespeak like "engineering excellence" isn't convincing in any manner. Uber develops an app and spends a ton on marketing. Beyond that Uber externalizes almost all of its costs. That's trivial to scale compared to actually scaling a transit company that has to invest in infrastructure.
I think you are overstating it somewhat. Ride Sharing is just a doublespeak name for owner-driver minicabs, which have been around for ages.
But when all these services end up doubling in price, as they're going to inevitably have to do (and the autonomous vehicle cavalry isn't going to charge in to save the day), it will be interesting to see what happens. My own experiences suggest there are a lot of people who don't want to get sweaty walking 3 blocks or to mingle with the masses on mass transit.
Nope! In the grand scheme of things, building an app, and some backend software while maintaining bunch of servers doesn't compare to building:
[1]
- 46 BART stations: 17 surface, 14 elevated and 15 subway stations.
- 669 revenue vehicles
- 112 miles of railway track
- a police force with 206 legally sworn-in law enforcement officers and 90 other staff members
- a wireless network that supplies cell phone service to riders underground
BART isn't even that good - I live in Sydney and we are always complaining about our double-decker dead-silent very clean commuter and city rail. Buzzfeed [2] even thinks our train system is from the future!
I love software - I want to make a career out of it, but we can't forget "real" engineering fields and the effort that goes into them.
[1]: https://edition.cnn.com/2013/07/04/us/san-francisco-bay-area...
[2]: https://www.reddit.com/r/sydney/comments/8rulvn/mfw_american...
The comparison between the two can't be made without adding in more factors.
I've wondered about that for some time.
How about the new cars that Uber is encouraging their drivers to buy with tantalizing financing?
Creating a market for new cars only ensures that new cars will continue to be produced in high volume.
I think that Uber from an investment perspective is looking good but from a social perspective it's looking bad.
operates a joke of a police force.
So these costs are passed along in the cost of an uber ride.
You did point at something I missed though. Technically every ride is being subsidized by investors (because uber operates at a loss afaik). To whatever extent they're losing money on every ride is a sort of disadvantage to uber ...
And of course, anecdotally we've all driven behind an Uber car that suddenly pulls over to the side of a traffic lane to let its fare out.
Let's see what the long term outcome is for Uber before we give them too many high fives...
I think a better comparison than total cost is how much money they are losing. The difference between what I'm paying and what I feel like I'm getting with uber feels like... a lot, and that lines up with everything I've read of their financial situation.
If you are willing to lose substantial money on a per-ride basis, sure, you are going to have a pretty nice transit product.
(Personally, I'd be pretty okay with the government losing some of my tax money on public transit projects... though I think if the government is gonna do it, I want them to lose the money up front on expensive, durable, high-capacity infrastructure, then come closer to breaking even on a marginal ride basis)
hence why we have taxi services, why all cities across the world have taxi and similar services. transit cannot go everywhere nor can it quickly adapt to changes in public needs. buses are the best bet for adapting but services need both rolling stock, drivers, and usage cases, to move whereas taxi and ride sharing services react on need.
plus in the current meta where "soon" there will be self driving cars and buses all which are EV it will completely transform public transit. I just hope to be there will it does.
Yeah, that is part of why I support systems that lean towards higher up front and lower ongoing costs.
>plus in the current meta where "soon" there will be self driving cars and buses all which are EV it will completely transform public transit. I just hope to be there will it does.
Self-driving won't help with congestion problems. unshared on-demand vehicles (and it doesn't really matter here if the on-demand vehicle is driven by a human gig economy worker or a computer) help with parking, but if anything, they make congestion worse, as they've gotta drive to pick me up in addition to driving to my destination. (of course, shared on-demand vehicles can help with congestion, at least when they are shared.)
The only "game changer" here is that potentially a self-driving uber could be cheaper (either by directly lowering operating costs, or if we get the hang of things, by lowering the number of injuries caused by mistakes.)
The problem is that if you charge less, I'm going to use the non-shared on-demand service more often; I only use the shared service to save a few bucks.
My point here is that we can probably predict what self driving cars will get us by looking at what heavily subsidized ride sharing services get us. self-driving cars might even get us less; if I could get uber-like point to point without having to park service out of a personally-owned car at a price point that isn't much more than a regular personally-owned car? I'd probably share rides a whole lot less than I do now.
Uber also has "express pool" option (again in bay area), wherein you have to wait for 2 mins to get matched.
As a rider, these options are great. However, not sure if it's profitable for these companies. I guess with higher rider/driver density, they can make it work?
F.D. I work for Uber, but not in the rides business.
https://therideshareguy.com/uber-is-ripping-off-frequent-rid...
hiking up return fares
[1]: https://therideshareguy.com/uber-is-ripping-off-frequent-rid...
[2]: https://www.theverge.com/2016/5/20/11721890/uber-surge-prici... - while it was only a study in 2016, in 2018 this is definitely true. Your upfront fares are hiked if your battery say 5%.
Do you have a source for this?
In a nutshell, this Uber feature is already there for me. I will admit - Uber's prices are mostly lower than Lyft.
I commute via Uber/Lyft every day and this happens to me often and I've found that relying on the actual address instead of using location sharing gets me the best results. In most open places Live Location Sharing can be great especially in crowds or when dealing with intersections but it is too sensitive in areas with congested Wi-Fi signals or tall buildings.
Without breaking NDA, it was something along the lines of "Why would Uber let customers wait longer?"
It's an interesting product problem but I felt that it would be more relevant for an economist to analyze, given that it's more of a supply/demand equilibrium challenge.
Why not offer better service for higher prices (instead of the reverse)? Getting service from a "super" tier of drivers, ones that have:
* a history of ON TIME pick-ups
* fluent spoken English
* straightforward following of GPS directions
* GUARANTEED air-conditioning
* non-smoking
* quiet not talking loudly on the phone
* not a POS car with destroyed vehicle suspension/shock absorbers and trashed seating
So, essentially, it's a tiny bus that makes stops along a route. But with less capacity than a bus and less consistent scheduling and routing.
(To be clear - I am pretty skeptical of the idea of demand-responsive "microtransit" such as Uber Express Pool).
Even when my ride is paid by someone else (business travel), I will often use Pool and now Express Pool as I find it efficient.
Your city clearly has different bus drivers to every city I've ever lived in. The best moment I've ever had on a bus was when the bus driver stopped on a level crossing while a train was coming. I've also had bus drivers clip signs and parked cars, crash into other cars and busses, brake and accelerate like they were trying to dodge a heat seeking missile, and just generally bad drivers. And in my current city, the bus drivers (and tram drivers) won't do anything about bad passengers, they sit in their little perspex box and ignore the passengers.
I was on a tram the other day while an ex con was screaming obscenities at old Asian lady. The lady complained to the tram driver, who did literally nothing. She had to get off the tram because he was getting so abusive I swear he was about to hit her.
Trains seem to be mostly OK, maybe because they're bigger I can just avoid them. But I've felt seriously unsafe on trams and buses before at night, and I don't scare easily. There's nothing quite so unsettling as a junkie having a psychotic episode in the back of the bus. You don't want to look at them in case they engage you, but you want to keep your eyes on them in case they feel like jabbing you.