Cap Remap is a disaster

Let’s recap:

AURA, run by grifters and con-artists, bought in heavily to the grid redesign fairy dust theory. It didn’t hurt that their 2 chief transit gurus live on the northern part of the 5 route and hated the jog through Hyde Park and NUNA (that was responsible for a lot of riders, but not the 2 most important ones, obviously).

First few months look good. Single digit fixed ridership increases. Huge increases in rapid (but this is comparing to a year ago when frequency wasn’t good, especially weekend).

Now we’ve got our first comparison that matters: Sep 18 to Sep 17. UT is back in session; AISD kids are back in school and not able to joyride even though they’re still being offered free rides.

Guess what?

Fixed-route1 ridership is significantly down – even though one would expect the impact of Cap Remap in a flat ridership scenario to show more rides, because the redesign forces more people to take 2 rides instead of 1 to get to their destination.

september ridership table
image thanks to Caleb Pritchard


image of a portion of the tweetdeck complaints
I retweet a bunch every morning

And my “capmetroatx” column in tweetdeck suddenly became full of complaints about UT students being miserable for not having the RR and the 5 (both the 10 and the IF are now full to overcrowding, leaving many people at stops, and they’re pissed about it, in addition to the normal “where’s my supposedly frequent bus” complaint you’ll see represented to the right). This volume of complaints is huge compared to what I saw in spring of 18.


And a friend of the crackplog forwarded me this image showing how well one of the new frequent cross-town routes is doing:

Man, if only somebody had told Cap Metro ahead of time that this was a dumb idea.

Oh, wait! somebody did!

So we have at the very least some very convincing circumstantial evidence that the lower ridership probably isn’t due to weather2; it’s probably due to the fact that they took some buses away from places that were using them very well (the 5 through HP and NUNA; the RR); and put them on routes where they are not being used very well (i.e. the 335).
I’ll fill in this post with more details later. But suffice to say – the Pollyannas were wrong again; not that they’ll ever learn.


  1. normal buses, basically; remember that the Rapid changes weren’t part of the Remap – they preceded it 

  2. as is always Cap Metro’s bullshit excuse 

Let’s Go To The Movies: Also: Why Nobody Takes The Bus There!

In a happy coincidence, this twitter thread came to me as I was about to take the family over to our go-to theatre; the Alamo Drafthouse at Mueller, to watch a bougie family entertainment (I have kids, sue me; Christopher Robin was good). When I posted some details about our upcoming trip, friend-of-the-crackplog Hunter S Thompson Caleb Pritchard was appropriately dismayed.

More importantly, it’s worth exploring this trip from my house a few blocks on the interior of Hyde Park as an example of how even ‘better’ transit service has a long ways to go, but stay tuned after that; because there’s a little more to it.

Let’s talk time!

From a long time ago, you may remember cheezy minimalist graphics like the ones below, which made their first appearance in the best-est “Why Rapid Bus sucks” post that still holds for a little while longer until the two infill stops are fully integrated. In this case, you can see the components of the trip from my house to the Alamo Drafthouse in Mueller, as follows:

DRIVING TRIP (what we actually did): 7 minute drive, 1 minute to park, 1 minute to walk from the garage to the front door of the theatre.

TRANSIT TRIP (what we could have done): 9 minute (0.4 mile) walk down to 38th and Guadalupe; typical wait of 7.5 minutes for the 335 bus (15 minute frequencies on this route! Best we’ve got!); 14 minute bus trip to the closest stop in Mueller (Berkman and Simond); and an 8 minute (0.4 mile) walk to the theatre’s front door. 1

Now here they are combined on the same graph2.

See a problem yet? Hint: When just the walk to the bus stop from my house (the first part of four on the transit trip) is essentially as long (in time) as the entire driving trip, counting parking the car in the garage and walking from the garage to the theatre, you might not be competitive overall.

Well, you might be tempted to reply, surely the bus is cheaper!

Let’s talk money!

I have bad news for you. The Alamo Drafthouse at Mueller validates parking (which would have been $3.00 for us); but not bus fare. So here’s the way those trips cost out, and note that this is temporarily cheaper for the bus trip as my 2 kids that would normally have had to pay are actually riding free for the remainder of the summer.3

The driving trip costs around 60 cents, thanks to this handy but minimalist commute calculator, originally designed for bike commutes but usable for this purpose, that helpfully excludes bullshit like depreciation and insurance.4. The transit trip would have cost $5.00 (two day passes for the two adults).

So what?

Now for the interesting parts. What would we expect in a better world?

Let’s imagine for a moment that being “transit-oriented” in Austin actually meant what it means in the rest of the world. How would the equation above be different?

For one, the parking garage would have to be less convenient than the transit. This means that the transit needs to drop off in front; and the parking needs to be at least a short walk away (move the transit stop directly in front of the business; leave the parking garage where it is for now, I guess).

For two, the parking can’t be free while the transit costs money. But do you think a business in Mueller as it exists today would be willing to make this trade? Of course not; each and every business in Mueller would die without the influx of cars from neighbors like me. They are nowhere near dense enough (by orders of magnitude) to get the clientele they need from people within walking or biking distance. And don’t forget to remember this when people credulous of the grid redesign cargo cults think they will be great for ridership – the places you’re reorienting service to serve better (your ‘secondary centers’) all have free parking, meaning the competitive transit-versus-driving value proposition is horrible compared to the traditional downtown.

For three, also, the realignment of service for Mueller focused on running things down Berkman, which might be better for the residents, but definitely not for anybody trying to get to the Town Center [sic] from outside the area. If Mueller’s reality matched its promises, transit would go straight to/from the Town Center, and residents would be within a short enough walk that they would want to go there to board buses. The actual reality isn’t that great; Mueller is too spread out, as discussed ad nauseum, so the transit has to run on Berkman.

So the real answer gets to an even more fundamental flaw in Mueller: It tries to be a “center” a la “centers and corridors” from Imagine Austin, but the density it has is nowhere near large enough to justify free transit and expensive parking, so it ends up in the uncanny valley of density. Difficult to drive in, unpleasant and expensive to take transit to, and with a tiny fraction of the people within walking and biking distance that would be required to keep their businesses in business without those drivers and their necessarily subsidized parking.

What’s the solution? Centers have to be orders of magnitude denser than this, so that parking doesn’t have to be free to keep businesses alive, or, you know, stop trying to pound a square peg into a round hole and just resume densification of our existing center where parking already doesn’t have to be free. Either way.

Also, though, please note that for a single person, this particular trip would still have been a dumb financial decision, but not quite as dumbererer as for the whole family. Consider that you have to pay a bus fare for each person (including the kids when the free summer ends), but a car full of 4 dinguses costs the same as for a single dingus driver.

Also also, though, please check out Caleb’s new gig – we expect great things!

(I’ll try to fix this up and flesh it out over the course of the week to make it a little less bare-bones but had to get this out there in case I run out of time. Remember this isn’t my day job and I don’t even have time or the cost-benefit ratio for it to be a frequent hobby anymore. Fuck AURA.)


  1. I’ve used the common practice here of assigning half the ‘headway’ to ‘wait time’ as in – the average time we’d have to wait for the bus if we just show up is 7.5 minutes. The minimum ‘wait’ time for the bus if you plan on a scheduled departure is typically 5 minutes – you’re supposed to give that long in case it’s early. The ‘wait’ time for your car is ALWAYS zero, of course. 

  2. updated on 8/14/2018 to make the phases the same across the bars, but the color scheme is still gonna suck because there’s only so much of my work time I’ll spend on making this pretty for y’all 

  3. details: two day passes at $2.50 each, kids are currently free. 

  4. Note here that even though my company car gives me gas for free, I costed out this trip as if I had to pay for it 

Know how you can tell they’re not honest?

Capital Metro edition

Yes, it’s been a while1.

In a recent twitter thread, Karl-Thomas Musselman posted the tweet below. I am making this blog post to capture it so that this well-made point is not lost in the twitter memory hole.

The graphic comes from Capital Metro’s 2016 approved budget on page 48. The full graphic is after this paragraph. What do you think this kind of choice in axis scaling suggests about Capital Metro’s honesty on rail subsidies?

Page 48, Capital Metro 2016 Approved Budget
Page 48, Capital Metro 2016 Approved Budget

  1. Note: I have not blogged much this year because the actions of Julio Gonzalez-Altamirano and others, especially linked with AURA, have made my investment in public affairs significantly less effective. This lack of content is likely to continue as long as the urbanist community decides his approach and style are preferable.  

Central Austin boxes: 2014 versus 2000

I’ve started this spreadsheet (read-only link; you can save and edit as you like) of mostly strong pro-rail boxes in 2000 (basically central Austin, until I hit the part of town where it started to lose – so a few non-yes precincts are included for geographic completeness). I didn’t go as far south as some people would; I consider central Austin to stop at Oltorf and go no further north than approximately Koenig. I did include some of lower East Austin.

Screenshot sample here:

20141115railspreadsheetpicture1

2014 results: http://traviselectionresults.com/enr/contest/display.do?criteria.electionId=20141104&contestId=71

2000 results: http://www.centralaustincdc.org/images/Rail_2000.pdf

So far, it looks like in extreme central Austin alone, a margin similar to that seen in the 2000 election would have yielded around a 5600 swing in votes (2800 nos changing to yeses, essentially). This is not yet sufficient to change the balance of the 2014 election (margin was about 27,000), but it is clearly a major portion of the swing.

Some key notes:

I have excluded Mueller because nobody lived there in 2000, so we can’t assign a reasonable value for their margin. The “mostly Mueller” precinct went for by 64% (1580/2470) – if I was forced to guess how they would have voted on the 2000 plan, I’d say 70%+. I also excluded one precinct on Auditorium Shores where 1 guy voted.

The 2000 report only has ranges for precinct margins. You can change the assigned value for each range in your copy of the spreadsheet if you want. I chose 75% for the “over 70%” boxes, 65% for the “60-70%” boxes, etc.

The precincts do not line up exactly. A few shifted boundaries, some were combined, numbers changed, etc. I have noted which precincts in 2000 I considered the most relevant for 2014. Again, you can save the spreadsheet yourself and change this if you wish. If more than one 2000 precinct was used for comparison, I averaged their assigned values first.

Will pull some interesting nuggets out of here later today as I get to them. I’m sick as a dog and at my kid’s chess tournament all day today, and my battery in my phone, being used for access, is dying fast.

 

Rapid Bus has degraded bus service overall

This VMU on Lamar at North Loop (google maps link; as of 9/5/2014 the streetview picture is from construction) is open now. I like it. It has a bus stop right in front of it! Streetscape is good. There’s actually a new Taco Cabana across North Loop from it, unfortunately with a drive-thru, where the pretty image to the right has a grassy field1. The property to the south of the Taco Cabana appears ripe for redevelopment soon as another VMU; I’d be surprised not to see it go that way within a couple of years.

052512_wheatsville_1479950a

Let’s imagine the resident of one of these new apartments wants to take the bus to Wheatsville Co-Op, an urban grocer located at about 31st and Guadalupe. Lots of people used to ride the bus to Wheatsville last I checked.

For background, the VMU ordinance was enacted as a quid-pro-quo for the McMansion ordinance – the logic was that we would build tall apartments (for Austin, anyways) over walkable retail on corridors where transit frequencies and usefulness was high. Lamar/Guadalupe definitely fit that bill, at least originally.

Before the implementation of “Rapid Bus”, the #1 ran about every 13 minutes during peak periods on this route. Google maps says that the bus portion of this trip takes 8 minutes on the #1. Note that Google doesn’t even consider the 801 a viable option for this trip, unlike Capital Metro themselves. We’ll get to that in a minute.

We can use the same “show up and go” calculations from this post to come up with this graph. Short summary: If transit service is to be truly useful as a replacement for the car, it needs to be frequent enough that you don’t bother to check a schedule; you just show up at the stop and a bus comes pretty soon (and by the way this was one of the big marketing points for the #801; so this isn’t just a condition I’m placing on them to be mean). Note that the walking time on either end for the #1 trip is essentially zero – there are bus stops for the #1 (but not the #801) directly in front of the VMU building and the grocer.

Originally, when frequency was every 13 minutes, a trip to the grocery store would involve a 0 minute walk, an average 6.5 minute wait (half of frequency), and a 8 minute trip on the bus, for an expected trip time of 14.5 minutes. Not bad.

However, in the world we live in now, Capital Metro has cut half of the #1s and imposed instead the #801 in place of the #101, stealing the local frequency for the express. How does that service work for our apartment resident?

Same calculations as above – we end up with an expected wait of 13 minutes (it runs every 26 minutes during peak)2. Total trip time is now 21 minutes, if you can get a seat on this bus, which has been a problem ever since the 801 change happened.

But surely the 801 made up for this drop in service, right?

Again, Google won’t even give this as a trip; but Capital Metro’s trip planner does.

20140905capmetrotripplanner1

Huh. Cap Metro expects the user of this ‘service’ to walk about a half mile north to the “Brentwood Station”, wait (12 minute frequencies during peak), ride the bus to the “Hyde Park Station” (7 minutes), then walk about a half mile south to Wheatsville. Hey Google, how long will those walks take? Google says 8 minutes each, roughly.

So let’s graph those new trips, shall we?

20140905stackchart

The results show that, and all of this is compared to the conditions before the #801 started (“old #1” in the graph), a resident of this apartment building can now either pay the same amount of money for a much less frequent service (#1) that will now take about 50% longer to get where they want to go, or they can pay double the price for a reasonably frequent service (#801) that will take more than twice as long to go where they want to go. People boarding a bus at this stop and travelling to Wheatsville have seen a significant degradation in quality of bus service.

What’s the conclusion? Well, even if you are foolish enough to think a 26 minute frequency local service still qualifies as “show up and go”, the residents of this VMU and many others in the area are unquestionably much worse off after the implementation of MetroRapid. And what’s worse – the developments resulting from the VMU ordinance were sold to surrounding neighborhoods as less of an impact on their daily lives because we all assumed many of its residents would ride the bus.

Still true? Doubt it.

More to come.


  1. Chris Bradford bait 

  2. most people would not consider this “frequent” and thus probably wouldn’t even consider the ‘show up and go’ approach, but let’s keep going 

A Really Crappy Chart For Your Monday

Lots of people, vitamin more about including some of my favorite people at Capital Metro, page are claiming that the Red Line is now “meeting projections”. Hmm. Let’s analyze this claim by looking at the archives, shall we?

1. The Red Line was opened up in 2009 – projecting 1700-2000 boardings/day, from Day One, with the following schedule:
This is the last monthly data we get before Big Changes make for a big discontinuity in the graphs. December is, anaemia as Capital Metro wants to make sure you know, drugs a low ridership month. As usual, click for larger versions. Analysis follows the pictures.

Continue reading “A Really Crappy Chart For Your Monday”

3000 words on rail ridership

I posted this link on twitter with the caption: “Austin Urban Rail Goes To Hell”. Note entry number for giggles.
I really don’t have time for this, more about salve with the 60 hour workweeks, shop young family including baby that still doesn’t sleep nights, and impending back surgery, but I have to say something, so I’ll be brief.
I offered a year or more ago to become involved with Leffingwell’s team on the urban rail project. I was ignored. (Note: I offered quite nicely.)
Recently, the plans have crystallized – and it’s bad. Shared running almost everywhere – except for one (admittedly long, but not really relevant) stretch from I-35 to the airport, the trains will be stuck behind cars – or at best, buses (including local buses). No, a ‘possible future transit lane’ on Guadalupe/Lavaca doesn’t mitigate; unless it’s reserved for ONLY Rapid Bus and the train – and I don’t see that happening; it’s going to be stuffed with locals too, and that’s if it even happens.

Unlike Brewster McCracken, who talked up reserved guideway everywhere except the leg out Manor to Mueller, Leffingwell’s team has relented and the plan now calls for the trains to be stuck in traffic almost everywhere important. McCracken talked about “time certainty” being a big deal on a trip to/from the airport (or to/from work, of course). You don’t get that without your own lane – period. No amount of Rapid [sic] Bus technology is going to get you there.
This rail plan, in its current state, is not worth fighting for. In fact, it’s probably worth fighting against, as was the 2004 plan that so many of the “why don’t you just stay civil” folks failed to affect in any way, shape, or form.

Be ready for a lot of the same people who claimed from 2004-2010 that car drivers would switch in droves to a train that required them to ride shuttlebuses to claim that the fact that these trains are stuck in traffic won’t keep people from switching to them.
Remember who was right before, and who’s been wrong the entire time. Or just be lazy and maintain access to the gladhanders to stay “civil” – and hold hands as we all ride the train off the cliff together – your choice.
And Not a done deal, you say? The engineering docs look pretty much done-deal level to me; as do the interactions with the media (note: the ONLY media outlet to cover the issue of guideway AT ALL was “Impact Central Texas”; their story here – good job guys; and shame on everybody else).

The urban rail system route is expected to follow Guadalupe and Lavaca streets, San Jacinto Boulevard and Congress Avenue. It will travel with traffic and may potentially receive signal priority at traffic lights, similar to Capital Metro’s buses.
An urban rail system in Austin is expected to cost $200 million in its first phase of development. The track will be 33.8 miles in length and extend from Mueller to downtown to the Austin Bergstrom International Airport. Photo by Bobby Longoria/Community Impact Newspaper. Click for a larger image.
“Big difference between this and a bus is that it can fit 170 people, mostly standing, where a bus caps out at 60 or 80,” Spillar said.

Hey guess what another big difference between this and a bus is, Rob? The bus that’s stuck behind somebody double-parked can change lanes. A train sharing a lane with cars is the worst transit possibly imaginable in a city where most people drive – it has the worst aspects of buses and the worst aspects of trains with almost none of the good parts of either.
More background on Why Streetcars Suck courtesy of Jarrett Walker here: streetcars: an inconvenient truth
So I guess I need to update my “IT’S NOT LIGHT RAIL” chart:
If your train runs on freight tracks, can’t run in the street, and requires shuttle buses – IT’S NOT LIGHT RAIL. Know what else? If your train doesn’t have its own lane – and relies on the same crap Rapid [sic] Bus uses to get a leg up, IT’S ALSO NOT LIGHT RAIL.
Summary: If you want to live in a city with good urban rail, your best bet now is probably to move away. Seriously.

spreadsheet behind this image coming sometime down the road, maybe.

CM flacks like JMVC and board members like Mike Martinez are making statements that rail ridership has improved. Here’s 3000 words to the contrary, pulmonologist fresh off Capital Metro’s presses. First two pictures directly from them; third one directly from me.

Continue reading “3000 words on rail ridership”