Curious about the gap between Sound Transit’s ridership models and reality, I had a discussion with ST planner Bob Harvey. He sent me this ST Transit Ridership Forecasting Interim Report (warning! 2 MB pdf) that explains the model ST uses for all its estimates across all modes, although the actual document is focused on North Link.
How it was computed: The Sound Transit model takes ridership data from existing Metro or Sound Transit routes in 2004. It takes PSRC data on population, demographics, employment, and highway congestion from 2004 and extrapolates them forward to 2010 or so based on PSRC assumptions. Then it applies the ridership improvements associated with Central Link’s level of service (speed, headway, etc.) to come up with the current ridership estimate.
What went wrong: Prior to Link opening, projected headways and travel time increased, cause ridership estimates to deteriorate somewhat. Once Link opened, the models overestimated regional employment. They also suffered from a lack of good ridership data for Rainier Valley-Seatac trips, a transit market that really wasn’t served before. And of course, no Metro route is completely replaced by Link.
Additional potential factors include the introduction of ORCA, transfer policies, and charging fares in the tunnel. Mr. Harvey, who didn’t work on the Central Link estimates, also notes that planners made “optimistic assumptions” about bus service realignment in Southeast Seattle, the segment that is “most definitely underperforming” compared to projections.
The estimates are actually worse than they first appear, given that they are geared to undercount ridership driven by sporting events and the airport.
The recent update replaces the insufficient bus data with Link data, but is still a victim to obsolete PSRC regional projections from 2004 data. Later this year, ST will release new projections based on a fully updated data set. It’s reasonable to assume that applying a standard growth rate to a baseline that’s well below trend will cause a general decrease in ridership estimates.
[ot]
Bruce, I won’t go as far as that, but it does point out the unrealistic and almost deceiving POV of ST at times.
First there was a tunnel, then no tunnel. Then there were going to be this much of a ridership base, now that’s off. Surprise, surprise….
I hate to say it, but if there’s one ST project here that needs drastic upswings in ridership numbers, its the Sounder Northline. I love these trains, but the amount of people riding them aren’t nearly as big as I hoped to see.
Sounder was largely a political project to provide a showcase for rail transit in the region without a huge up-front capital cost or long build times. It is not an optimal alignment in any way, and I suspect most transit planners are unsurprised with the north line’s poor performance; triply so, given the fact that CT is providing cheaper one-seat rides to far more of Snohomish County and has been for decades, and that the north line suffers horribly from mudslides for part of the year, which would be unaffordable expensive for ST to fix.
Oh I don’t know… Downtown to Kent in 12 minutes sounds pretty damn optimal to me…
All other things being equal, there’s more ridership going down 99, as Link will do. That’s what I’m referring to by “optimal”.
Downtown to Kent is 19 minutes.
I don’t know what Anthony meant by “First there was a tunnel, then no tunnel.”
OK, 11 mins to Tukwila, 19 mins to Kent. Still faster than driving. In rush hours, Sounder will still be faster than driving for the rest of the route past Kent.
With 3.5 million people in the region, accommodating more Density in the Kent area is a good thing. (IMO)
I don’t think North Sounder was ever expected to generate huge ridership numbers. As Bruce said it was largely political so Snohomish county got something for Sound Move. Though I can’t help but think the ridership would be better if the original plans for multi-modal terminals at Edmonds and Mukelteo had materialized.
The one bright spot of North Sounder other than Everett Station is it has paid for some track improvements the Cascades and Empire Builder get to use.
Still I think the Sounder North money might have been better spent on HOV improvements and more express bus service.
South Sounder always made a lot more sense than North Sounder.
[comment policy]
I think this is one of the main public enemies…wasn’t there a pretty comprehensive rail-bus integration plan that went awry?
And of course, you can never forget the recession. There’s good evidence that lower-income communities fared worse, which would disproportionately impact ridership.
I don’t know specifically about that, but I do know from talking to Metro planners that ST did not know the fare structure of Link at the time when Metro was considering potential network restructures in the RV. They didn’t know if a ride to downtown would cost $1 or $5. That makes a huge difference, especially in a largely low-income area the RV. Given this, Metro’s planners elected to first do no harm. Had the current fare structure been known at the time, there might have been more radical changes.
As to the cock-up with the location and design of the Mount Baker TC, I’ve never been able to get any straight answer, on or off the record, from anyone as to how the hell that happened.
If this hasn’t already been covered, it sounds like a good future STB topic.
More importantly (IMO) is how to fix it ASAP. And I’m not just talking about just from the technical side, but just as importantly the political.
It’s not just politics (although that will be a part of it when the time comes) it’s the fact that Metro doesn’t have the resources to push through major restructures in multiple parts of the county at once. These things are huge undertakings that require tens of thousands of hours of staff time, public outreach etc. No major restructures will be on the table for the south end probably for a couple of years at least. Most staff time now is focused on the north of the city and RapidRide restructures.
There would have been better integration if the county council hadn’t “saved” the 38 and 42. That took hours from the 39, which covers two of the underserved east-west corridors.
In fairness to the council, they only “saved” the 42. The 38 was a new creation. The 38 had its chance, and now it is going bye-bye.
I’m sure we’ll hear it here if the old “Save the 42” sign goes back up on the ACRS building. It’s not just up to the council to say “No”. We need to out-comment ACRS, and out-letter them with constituent letters to Councilmember Gossett.
I already put in a bad word for the 134, which is one of my routes.
I think the 38 has been around for years. Metro did discontinue the portion between Beacon Hill and SODO when Link began service.
Yeah, the 38 wasn’t new in the last few years. It might be about 10 years old now. But the route was truncated when Link came along.
And don’t forget that even though the 38 is a short route with low ridership, it goes up a VERY steep hill. I’m in pretty good shape and it was a struggle for me to walk up Beacon Hill (as I recall there aren’t good sidwalks there either). Elderly or disabled people could not do it. It really illustrates the trade-off between productivity in terms of ridership and providing a basic level of mobility for vulnerable people. I’m not saying we should keep it if the $20 CRC goes through…I just think people need to be mindful that every route has a reason and not narrowly view maximizing ridership as the only goal.
“you can never forget the recession. ”
Really? I thought despite the recession the population here is on an unstoppable upward curve? And wouldn’t people be opting to take the low cost mass transit route when unemployed rather than private expensive car ownership? When gas prices went up, and people were still commuting downtown from the suburbs, there were reports that many switched to express buses. Why wasn’t there a similar intracity trend?
[ot]
OK, I can save ST some time and money because I’ve already worked out the ridership for University Link. Here is what I pulled out of my Analytical Statistical Survey.
Columbia City and Othello?
Where’s Capital Hill?
Oops, Columbia 2,100 and Othello 2,300. Those are in the 59,700 total. Capitol Hill. Thanks for the corrections, we can now advance this from Preliminary to Final Draft :=
No Problem. Why more for Othello than Columbia City?
Simply because current boardings are slightly higher at Othello.
Bernie,
what methodology are you using to generate these estimates?
I think your numbers for both Capitol Hill and Westlake are a bit off. I expect Westlake will stay by far the busiest station in the system even at full ST2 Link build out.
Given the high transit mode share on Capitol Hill I expect the numbers for that station will be much better than you predict as well.
Metro should shoulder some blame in Link’s underperformance.
Metro chooses to run a whole fleet of buses right by a good connection to Rainier Beach Station.
Metro chooses to incentivize cash payment, which creates a financial disincentive to jumping on the train.
Metro is choosing to try to keep all the SR 520 riders going downtown from using UW Station as a transfer point.
Metro is keeping quite a few neighborhoods from having access to the south end of Link, and in most cases putting a lot of unnecessary service hours into doing so. I’m not just talking about RBS. My neighborhood (South Park) has also been begging for a connection to TIBs, to no avail.
ST can only control so many variables. Metro’s erratic and illogical behavior is one of those variables it can’t control.
100 percent on target assessment.
“Metro chooses to run a whole fleet of buses right by a good connection to Rainier Beach Station.”
If you’d bothered to read my comment above, you might understand why.
“Metro is choosing to try to keep all the SR 520 riders going downtown from using UW Station as a transfer point.”
Hogwash. Metro is trying nothing of the sort. No formal plans are in the works for restructures resulting from University Link, but some planners do have interesting ideas for it, which they don’t want discussed widely yet — they depend on too many other external factors, like how tolling affects bus reliability on 520.
Really, if you can’t be bothered to step away from your keyboard and talk to people and find out what’s actually going on, you should shut up.
Bruce,
I do talk with transit officials and politicians, and hopefully in a more civil tone than you do.
If you’re talking about the lack of planning staff time, Bruce, then isn’t it obvious Metro is making a poor choice to allocate resources toward keeping service hours on the road over hiring an extra planner or two to save several FTEs of operators?
Metro has been focusing all their planning resources on finding ways to cheapen costs while maintaining existing service, rather than restructuring the system for better service.
However, I don’t think you can call this erratic and illogical. It’s just a symptom of an under-resourced planning department. With the big drive for cost-cutting there’s no money in the budget to even look at incremental improvements, including the zero-cost ones. Even this major eastside service restructure required by RR-B seems to have mostly been phoned in and smells a bit of design-by-committee (still an improvement, though).
I’m with Metro on this one. The bus stop is in the right place according to the ridership numbers, it’s the Link station that was poorly sited. UW medical is a higher ridership draw at that location than all the bus/rail transfers put together.
I’m not happy with anyone over the loss of the Flyer stops, though. Metro did a poor job of advocating for the only all-day two-way cross-lake transfer for the highest ridership route in the system.
Metro still can’t explain why UW Med is such an important destination for SR 520 commuters, but not for downtown commuters. ;)
I’m taking bets on how soon SR 520 service from UW Med gets cut below 30-minute headway after 8 pm once UW Station opens.
WSDOT, SDOT, Metro, Sound Transit, and the UW should all be flogged for how poorly integrated and coordinated the billions in transportation and other investment targeted for Montlake and the triangle are.
The biggest epic fail of course is the loss of the Montlake flyer station.
Off peak is when the flyer stop connection is most important so this could end up being an improvement. I think the current top spot for epic fail is the second bascule bridge across the Montlake Cut.
Bernie,
Thanks, I wasn’t aware that some of the problems with eliminating the Montlake Flyer stop had been addressed.
I agree with you on the second Montlake bridge. Personally I’d be discouraging traffic through the area by making the 520 ramps HOV only during peak hours.
Keep beating the drum! It’s starting to be heard.
Just toll the Montlake ramps.
That could work since electronic tolling will be in place for the bridge anyway. Use a combination of Transit Only HOV/HOT lanes and control the flow so that everything isn’t just cued up waiting for the Pacific triangle. Maximize throughput and bring in a little revenue to offset the cost of the lid.
This isn’t just with Link. There are many spots where Sound Transit bus service runs in parallel and near perfect competition with Metro service. A prime example is the 212 and 214 vs. the 554. Yes the 212 and 214 are faster and more direct, but is that worth losing a broader network of service? Somebody seems to have decided “Yes”.
I frequently ride the 554 to Issaquah and shake my head every time I see the 214. Even though the 214 is an “express” and makes 2 fewer stops than the 554 it only has about a 2 minute advantage over the 554. That route should be axed immediately. The increased ridership on the 554 would increase its productivity and save ST some money.
The 212 I don’t have as big a problem with just because of the huge volume of rush-hour commuters between downtown and Eastgate.
“Metro chooses to run a whole fleet of buses right by a good connection to Rainier Beach Station.”
As has been pointed out before, there’s a significant tradeoff to truncating the 101, 106, and 150. It would add 5-10 minutes of travel time PLUS the wait for the train. At that point you’d have to ask whether Renton and Kent are getting adequate service. It should not take most of an hour to get from downtown to a close-in suburb. Especially since we’re talking about regional centers, not isolated residential neighborhoods.
Maybe Metro will have to truncate the routes off-peak to make budget, but it should be presented as acknowledging the downside (longer travel times) while emphasizing the upside (greater frequency, if the bus frequency is doubled). But it’s unjust to say Metro has made a completely idiotic decision (to keep the 101 and 106 going to downtown) as if it’s a 100% no-brainer with no tradeoffs involved, because there are significant tradeoffs.
LINK ridership numbers (and Sounder ridership rates, for that matter) actually matter very little. The voters have spoken – the agency has its mandate and will carry out the directives. Whether or not people now are choosing to use this alternate mode of transportation does not matter. We’re building an 80-year system, and that’s the required time horizon!
Exactly right. Doesn’t matter if it makes any sense, or not — it will be built.
This is pretty much the definition of “boondoggle.”
It’s not simply a matter of choice. If the bus doesn’t connect, I don’t ride the train. Metro made that choice for me. :(
You expect people to be riding on 40 mph electric trains in the year 2090?
They have been for over 130 years, why not in the future? Are you thinking more Jetsons-style hover cars? Higher priced oil is not the only problem with SOVs. No matter their energy source, SOVs are not the answer to all of our transportation needs Mr. Bailo.
I optimistically expect the Rainier Valley segment will be put underground by then, as the NYC subway was. (It was elevated before it was underground.) Of course, the Duwamish/Georgetown bypass will probably have been built by then too.
An awfully long report to explain why the numbers are so wrong. The forecasts really don’t matter much, except for two reasons.
1. Given the same methodology is used across all modes, then it’s a comparative tool to evaluate how well one mode is in relation to another.
2. It’s a financial forecasting tool for decision makers when looking at the costs of various projects and the ‘forecast’ benefit of the expenditure. If you’re hell bent on building something then it can be used to say almost anything you want, within reason. Once a project breaks ground, the models are history. When was the last time government quit building something because of errors in the model? Do you believe the DBT model is pure of heart?
Now that Link is up and running, it would be foolish to not adjust your assumptions for budgets using real data. Ridership determines level of service, costs, OR/OE, staffing, and a ton of other real-deal decision making inputs.
The price to pay for bad forecasting is in you’re credibility. Sound Move was fool me once. ST2 was fool me twice. ST3 needs some wonderful news when ULink opens.
Also, this has to be high on the FTA’s radar scope, reading the letter to them in the report appendix. Good luck getting much in the way of FFGA for N. Link, as most of the other applicants that have made good forecasts and will highlight this little problem ST now has.
I think Bernie’s numbers are still high, but that’s just my own opinion (worth $.02).
I agree with most of what you say here, except that 326 words isn’t very long. I’m also not inclined to panic about federal funds, at least as a result of this.
Sorry Martin, I was referring to the ST report and appendix.
Stellar job reporting on this, as it does matter in some circles, mostly in DC, and depending on who’s holding the gavels.
My own read on future funding is mostly based on my belief that our economy is headed for a ‘double dip’, and the bottom of financial, housing, and most everything else is headed south about the 1st Qtr of next year. Cash for new starts will be significantly lower than in the past, and what funds are available will be extremely competitive, and may get down right nasty.
The problem is not a double dip. During the Great Depression there was gobs of mass transit funding.
The problem is a brainless austerity-mad hard-money set of people in the federal government, who don’t want to understand that Keynes was proven essentially right, again. See Krugman for more…. we need an FDR-style New Deal, and we ain’t getting it any time soon.
…so you may be right. We may have stupid enough people running the federal government that there is a shortage of transit funding.
Given that this will be due to *stupid people*, however, don’t expect what funding there is to be dispersed rationally. Expect it to be dispersed completely *irrationally*.
“ST3 needs some wonderful news when ULink opens.”
The plan WAS to put ST2 on the 2016 ballot, which would have been before U-Link realistically would have opened. Now that the lawyers are gone though, there won’t be an ST3.
derp – “The plan WAS to put ST3 on the 2016 ballot . . .”
The most important thing is to get ST2 finished. That covers the core central part of the region, and it does even more than earlier hoped (by terminating in Lynnwood rather than Northgate). ST3 would certainly help (with Link going to Everett, Redmond, Tacoma, and a Burien-Renton line), but it’s still less of an impact than ST2.
Of course, there’s also the additional lines in Seattle, which I would like to see. But again the same principal holds. For somebody in Ballard, being able to transfer to Link at Brooklyn is better than having no Link at all. It would certainly help Ballard-Capitol Hill, Ballard-Redmond, Ballard-Lynnwood, and Ballard-airport trips, even if Ballard-downtown is a wash.
Are you talking about the memo in appendix C, “Updated Treatment of Bus Speeds in the Sound Transit Model”? That’s from August 2002. It sure looks like a lot more than 326 words (five pages) but I’m not seeing any other letter to the FTA referenced in the appendix.
Oops, this was supposed to be a reply to comment-173463 Mike Skehan says:
June 22, 2011 at 1:24 pm
Yes, as far as the FTA radar scope reference.
The FFGA reference was from the Intro on page 2.
“The next ST model update will be 2012-based using new surveys and counts data with the general
incremental modeling framework. The updated ST model is intended to be used to produce
ridership forecasts in support of the New Starts application of the North Corridor Transit Project to
FTA for entry into Preliminary Engineering. The following presents a brief history of ST transit
ridership forecasting.”
In other words, ST needs a healthy dose of FTA funds to complete N.Link, or scale back the opening date.
Seems like this is a cautionary tale for the Tunnel, 520 replacement, etc, etc, etc
If they are so incredibly wrong about a big project like this, then why should anyone believe them about anything else.
Right now the taxpayers are getting soaked for a system which is terrible and costly in an era where budgets are being cut for education and basic services.
I would call inexpensive rapid transit a basic service.
North Link between UW and Northgate doesn’t need New Starts money.
Besides ST has a good defense of the model they used for Central Link. The PSRC population and employment numbers are the same numbers all other transportation planning is based on.
Now it is true Sound Transit will need New Starts money for the North Corridor (Northgate to Lynnwood). However I don’t know of a case where Federal grant money was denied to a project because the agency proposing the project missed a prediction for a prior project. Where grants have been denied it has almost always been on the merits of the project itself.
To that end both North Link and the North Corridor are likely to look very good under New Starts criteria. Furthermore some of the factors in why the Central Link numbers were off aren’t present in the U-Link/North Link/Lynnwood corridor. For one it is pretty easy to guess transit demand based on current bus ridership as existing bus service more closely parallels the rail line. For another this corridor has some of the highest transit demand in the state. Finally ST, CT, and Metro are much less likely to completely drop the ball in restructuring bus service to feed the corridor rather than largely leaving current bus service as is.
Over the last 20 years the SOV mode of travel to work has jumped 10%. Transit mode has stayed the same (negligible outside of King County). The change to drive alone has come entirely from the carpool share. That I would have never guessed. Is it because of the increased employment in the suburbs, a wealthier population, a shift from large employers to more small business?
Certainly a part of the growth in SOV has to have been the continuing artificially low price of petrol in the US and the very modest tax we pay on it.
True that from 1980 to 2000 the inflation adjusted cost of gas was dropping. Since then it’s almost doubled (right back to where it was in 1980) so it would be interesting to see the stats for the last decade. But the average mpg has also improved by about 50%.
I seem to have dropped this:
>
I’ve never been a big fan of CAFE standards but like seatbelts and 5 mph bumpers sometimes it’s the only tool that works. For years car makers have lobbied to keep light trucks exempt which has in part fueled the SUV craze. In fact with the Cash for Clunkers program there was a $1,000 incentive to buy something classified as a SUV/light truck (like our Subaru Outback, go figure). General Motors CEO Dan Akerson recently said, “You know what I’d rather have them do — this will make my Republican friends puke — as gas is going to go down here now, we ought to just slap a 50-cent or a dollar tax on a gallon of gas.” And as the famous Engine Charlie quote opines, “For years I thought that what was good for our country was good for General Motors, and vice versa.”
@Mike Orr; Mike, I was referring to the original tunnel that was supposed to be built through the Rainier Valley. Yes, I’m still peeved at ST for their lie on that one, and understandably so…
Bruce, I see your point about CT serving a large portion of Sno. County, but the fare of bus vs. train is nominal (in my case), only 50 cents more for rail, iirc.
Mike Skehan said it best, credibility is at stake here. If ST wants to keep the transit ball rolling, and with supporters to bat for them, then they need to provide accurate, non-biased info. Heck, I want them to succeed, but I’m having a hard time trying to figure out why to give my unequivocal support to an organization that hasn’t been straight up in the first place.
Anthony, please provide your evidence that Sound Transit ever planned to build a tunnel through the Rainier Valley.
I was active then in rail transit planning, and I have no such recollection. In fact, the citizens who were involved in planning events back then were opposed to an underground line. They were happy that Sound Transit (then called the RTA…) was not proposing a tunnel.
As Transit Voter said, I don’t think ST ever considered a Rainier Valley tunnel. They may have been open to it before the initial cost comparison was made, but not after. It was neighborhood activists (and others like myself) who wanted a tunnel. But ST never promised them a tunnel, just like it has never promised Surrey Downs a tunnel.
Mr. Harvey (of Sound Transit) states:
( planners made “optimistic assumptions” about bus service realignment in Southeast Seattle, the segment that is “most definitely underperforming” compared to projections. )
Facts are often uncomfortable. Here’s the fact’s citizens;
Link pathway from downtown Seattle to the airport had 2 logical routes. And 1 really bad route (Raineer valley).
Good Route 1) A straight line to the airport, passing Boeing field, with at the most two low cost stops. No deep Beacon Hill tunnel and no ultra deep Beacon Hill station.
Good Route 2) Or, West Seattle to the airport. Less costly by far, than the MLK (Rainer Valley) route. No deep Beacon Hill tunnel and no ultra deep Beacon Hill station.
Both good routes would have completely avoided a $300 million dollar Beacon Hill tunnel and the $250 million dollar deep vertical bore station. The Beacon Hill light rail station services a lightly populated residential area, with high Rainer valley unemployment (the residents have a low demographic to travel to jobs) and low business occupancy rates (very few residents of other places travel to Rainer valley for jobs). This demographic is true not just for Beacon Hill, but all Rainer valley.
The Boeing field route would have cost so much less, that Federal Way would have had its light rail. But Federal Way is not powerful in the political game.
What of West Seattle route? It would have cost about halfway between the Beacon Hill-Rainer valley route and the very low cost Boeing Field route. But ridership would be nearly double, if the bedroom community of West Seattle had been served. West Seattle residents have above average demographics for employment in Seattle, University district, Bellevue and Redmond. But alas, again West Seattle is not politically powerful. In fact, it’s actually best described as Seattle’s retarded step child whom just gets to pay tax.
Why was Rainer valley so politically connected? The racist and profiteer Ron Sims. Who by the way is on his way back to Seattle (shhhh Ron Sims), now that this president has finally figured out Ron is a nasty liability? Please Ron, find another place to victimize.
We (all of America) are losing the econmonic war with China. Yet we (the suckers) still play the racsit game, until we all go out of business. Ask yourselfs, what route would the Chinese business model have taken? Social engineer is almost allways a failure. Ron Sims got rich. So be happy for that.
A line going down the Duwamish Valley would have performed extremely poorly. Probably 1/3 or less of what the current line gets for ridership.
A line to West Seattle would very likely have run in the same cost range as the Rainier Valley route if not more. First there is the need to cross the Duwamish which would require either a tunnel or a new bridge. Then there isn’t a good ROW for surface running between the Junction and the Airport. Elevated would very likely have been a non-starter at least between the Junction and Morgan. That leaves a tunnel which wouldn’t have been cheap even as cut/cover. I’m not convinced the ridership numbers would have been any better than they are for Central Link either.
As for the Beacon Hill Tunnel and Station, they gave ST experience with building underground in the region. First the experience convinced Sound Transit they didn’t want any deep mined stations on U-Link/North Link due to construction risk. Second the experience means there is much less risk in digging the tunnels necessary for U-Link and North Link.
Maybe a 1/3 less than current ridership but not a 1/3 of current ridership (facts not in evidence for either claim; it’s a blog, not an EIS). Thing is, the huge savings in cost would allow ST to build farther south and recoup the losses in the RV. The time penalty of the RV detour will forever limit the effectiveness of that extension. Maybe that’s not all bad. It will in the long term force dense development closer to the CBD. Ironic that what was sold as Social Justice will really be Gentrification. And Federal Way really should be a suburb of Tacoma, not Seattle!