Seattle light rail in the first nine months of 2013 hit a weekday average ridership of 28,717, which despite strong summer ridership is 12% below the pre-construction forecast of 32,500 supposedly to be reached by 2011.
This page is a living document that is revised and updated periodically, most recently November 24, 2013.
The next page update will come when final 2013 ridership numbers are available.
Here is Sound Transit’s best possible spin on long-run failure to achieve its forecast: Average weekday ridership of Central Link is up 11 percent over the 25,870 boardings per day in the same period of 2012.
This updated page covers Seattle's light rail passenger counts from the first day of revenue service July 20, 2009 through September 30, 2013. Sound Transit has provided PITF with the data presented here, all of which are estimates subject to change. Archive of data sheets from Sound Transit is at the bottom of this page.
Sound Transit's short-run forecast for Central Link Light Rail ridership in 2013 is 27,900 weekday boardings and 9.2 million annual boardings. Through the first three quarter of 2013, the weekday average was 28,717 which exceeds the all-year forecast, an unsurprising consequence of a normal seasonal pattern in Seattle, with the typically weak winter quarter now likely to drag down the all-year average closer to the forecast. Light rail ridership since opening in 2009 has risen month by month in the March through August period, and then begins a downward slide that lasts well into the following winter. Year over year, however, the ridership has risen; ridership in 2013 through the first three quarters is up 10.5% over 2012.
Sound Transit's Central Link Light Rail between downtown Seattle and Sea-Tac Airport achieved its highest ridership ever on the occasion of Paul McCartney’s concert at Safeco field on July 19, 2013, attracting 41,938 boardings. Ball games and other events in the Century Link and Safeco stadiums near Stadium Station are proving to be an important source of ridership on light rail. PITF calculates that from May through September 2012, sporting events brought in an additional 71,000 riders per month.
Higher ridership means more standing on the light rail instead of sitting, as shown on the front cover of Sound Transit's 2011 annual report at right. Meeting the ridership forecast would require that even more passengers must stand instead of sit for part of their trip. This condition has been totally expected by agency planning, but perhaps not so much by Seattle transit riders. Crowded trains may discourage ridership by those who have options. While light rail cars can be packed safely by design with up to 200 riders, PITF has observed the reluctance of passengers on a platform to board after about 150 folks have stepped into a rail car. Sound Transit light rail trains are currently limited to two cars by the track capacity where trains reverse direction at the north end of the Downtown Transit Tunnel.
Short-term ridership forecasting at Sound Transit is now improving year by year. In 2012 the target was exceeded by one percent. In 2011 the average daily weekday boardings came in at 23,617, 5.5% below a dramatically-lowered budget target for 2011 of 25,000, set after the year began. In 2010 the actual ridership of 7 million all year was more than one million below the forecast of 8.1 million. The average weekday ridership for all of 2010 was 21,026, which was 21% below the short-run forecast. The 2010 forecast was made in fall 2009, a year after the beginning of the recession in 2008 that was incorrectly blamed by Sound Transit for the ridership being below the short-run forecast in 2010.
More significant has been the been the failure in the agency's long-run forecasting, PITF has recently learned that the Federal Transit Administration (FTA) was told by Sound Transit in 2003 -- when the $500 million Full Funding Grant Agreement (FFGA) was awarded -- that the predicted fall 2011 average weekday light rail ridership would be 32,500, a number not yet reached as of the end of November 2013, and higher than the most recent forecast for 2013 of 27,900. Quoting Sound Transit's July 2012 draft Before and After Study, "For all of 2011, Central Link carried 7.8 million passengers compared to the FFGA prediction of 10.7 million, or about 27% lower than predicted." Only in the summer of 2013 has the 15-day moving average of ridership reached above 32,500.
Authorities plan for the majority of light rail riders to be people commuting to work, and thus there is a focus by Sound Transit and Federal Transit Administration on weekday ridership on non-holidays. The blue line on the next chart below shows non-holiday weekday ridership day by day on Central Link light rail since July 2009. Friday is often the most crowded day of the week on the trains. The red line on the chart shows a 15 day moving average -- amounting to three weeks -- that smoothes out the daily fluctuations and allows detection of up or down fluctuations in the weekday ridership. The trend is up, with seasonal variation, high in summer, lower in winter. At the end of December, 2012, weekday ridership was trending strongly downward, the same trend seen in December of 2010 and 2011. As of March 2013, the ridership trend headed upward for the spring and summer tourist season, but by September had begun the expected fall downward slide that will likely not bottom out until late winter 2014.
45,000 boardings per day indicated by the straight orange line high on the chart was the 2020 expectation for the Airport to Downtown train service presented frequently to the public while the line was under construction prior to 2009. Long-run forecast ridership has been shown on other light rail lines to be reached about 18 months after the line completely opens, which would have been July 2011 for Seattle. In response to not yet even coming close to 45,000 per day, Sound Transit claims its light rail is maturing more slowly than the average USA light rail service, which experience so far indeed indicates to be true.
Sound Transit has given up on the forecast of 45,000 weekday average by 2020 for the current number of stations. Last year, the actual ridership results from 2010 and 2011 plus Sound Transit's own forecasts for 2012 through 2015 formed a clear trend line seen in the chart below as the lower line that ended up in 2020 at around 36,000, not the 45,000 that was bandied about prior to 2009. Now this year a new, more optimistic forecast has the light rail ridership headed toward 42,000 per day, closer to what was promised to the Federal Government back in 2003 in exchange for a $500 million grant to build light rail from Tukwila to Westlake Center, called "Initial Segment." Significantly, the original forecast did not include riders from Stadium Station, a boarding location critical to the ridership that Link now actually experiences. However, PITF is encouraged that Sound Transit has set strongly growing ridership over the next three years, 2013-2015, as an important objective for the region's significant light rail investment made to date.
Sound Transit has television ads encouraging ridership across all of its services: http://vimeo.com/36161525 for example.
Sound Transit is now planning light rail line extensions to open in 2016 serving two stations northward to Husky Stadium and one southward to South 200th in SeaTac. Incredibly, the agency forecasts these three new stations will expand system ridership by over 74,000 riders per weekday. Sound Transit's present day ridership experience compared to earlier pre-opening forecasts does not bode well for this forecast.
Public Interest Transportation Forum thanks Sound Transit for its willingness to share data with us, and thus document a unique record of day-by-day development of light rail ridership in a world class city. All data sheets from Sound Transit are posted at the bottom of this page. The route of Seattle's Central Link light rail from downtown to the main regional commercial airport is shown here on a map that also depicts with dotted lines the forthcoming extensions both to the north and the south. Some history of how this line was developed is here on Seattle Transit Blog.
This frequently updated summary page on Seattle light rail ridership was referenced in a March 3, 2011 Seattle Times article by journalist Mike Lindblom, "Sound Transit ridership falls short of goals." Expressing the attitude of many rail boosters, Sound Transit Board member Dave Enslow shrugged off below-forecast ridership in this article: "He emphasized the rail lines are a 100-year investment." PITF counters that the interurban regional rail system Seattle had 100 years ago was terminated well before it had seen a century of operation.
Return to the Public Interest Transportation Forum home page.
In considering the weekday ridership, note that rail ridership on a new line in any U.S. city typically rises within 18 months to its ultimate level until more stations are opened or the line is extended. William Millar, head of the American Public Transportation Association, noted this pattern in an interview in 2009, reported by the P-I. Tacoma Link, open since 2003, illustrates this reaching of a plateau in the previous chart, red line. As noted earlier, the 2020 forecast -- ultimate ridership -- for the Airport to Downtown segment in Seattle -- considered by itself separately from future extensions -- was 45,000 average daily boardings per weekday. The new, revised Sound Transit trend line noted earlier appears to reach 42,000 weekday average in 2020, but PITF's reading of the numbers leads us to believe that for Central Link the ultimate ridership prior to its extensions north and south may be settling in at less than that. Sound Transit now hopes for 31,100 in 2015, the last full year in which ridership on the Airport to Westlake Center line is forecast separately. Oscillation between 20,000 and 30,000 average per weekday is what PITF now believes likely until the next two stations open for service in 2016. This range has been the case May 2010 through March 2013. As noted earlier, Sound Transit was confident when construction began that a 32,500 weekday average would be achieved by autumn 2011.
Because weekend ridership has turned out to be important for Seattle's light rail, the next ridership chart provided below shows all days. The first, second, third, and fourth years of light rail operation are displayed on the same time axis, with the July 20 revenue service start-up anniversary beginning each year. The fainter daily numbers jump up and down considerably, but to indicate what's happening more clearly, the 14 day moving averages of daily ridership – on all days including the important weekends – are shown in thicker lines The first year, July 2009 to July 2010 is in green. The second year, July 2010 to July 2011 is in red. The third year, July 2011 to July 2012 is in blue. The fourth year, July 2012 to July 2013 is orange. The fifth year is the purple line beginning July 2013, and now covers just through the end of September 2013.
As the chart illustrates, ridership in the second full year of Seattle's light rail service (in red) did not grow as strongly as in the first year (in green). The blue line of the third year tracked pretty much continuously just above the red line of the second year. The orange line of the fourth year started to trend downward after the SeaFair weekend 2012 summer peak moving average reached about 30,000 per day., although the combination of Seahawks and Husky football at Century Link Field near Stadium Station provided a visible surge during September 2012. Fall 2012 showed upward bumps because of the Seahawks despite the downward trend. Ridership bottomed out as 2013 began, and a steady upward ridership trend began that turned into a power surge in the mid-May to mid-June period when the Mariners still had hope of a winning season. Growth continued moderately after that, peaking about the time of SeaFair weekend, and then the fall decline began, shown with the purple line on the left hand upper side of the chart below. The winning Seahawks and the high price of parking provide hope for more light rail ridership in the last quarter of 2013 than ever before seen in that season.
As shown below, in winter the ridership averages have occasionally bumped into the low winter levels of previous years.
The alignment and stations of Seattle light rail line this year are the same as last year, with the terminus points at Westlake Center in downtown and the Sea-Tac Airport south of the city limits, and eleven stations in between. In February 2010, King County Metro reduced bus services that ran parallel to light rail in the same corridor and deployed more bus trips to serve light rail stations. In 2011-12, King County Metro has had enough funding to maintain bus service to and from light rail stations and thus support Sound Transit's goal of increasing ridership.
PITF has received some questions about seasonality of transit ridership. One question has been, does the drop in light rail ridership mid-summer through Thanksgiving in both 2010 and 2011 follow a usual pattern of transit seasonality? The answer is no, based on the observation that monthly ridership on Metro bus from 2002 onward has always been greater in October than July. The falling ridership pattern on Link Light Rail exhibited in 2010 and 2011 after August is more akin to the ridership pattern on the Seattle Monorail, the Seattle Lake Union streetcar or the former Seattle waterfront streetcar all of which demonstrate peaks of ridership in the summer compared to the rest of the year.
The Mariners, the Sounders, the Seahawks, University of Washington, Sound Transit, and Port of Seattle all promote light rail use for attending baseball, soccer, and football games. Despite low ballgame attendance during the 2012 losing season, Mariners baseball with frequent games has been a big contributor to Central Link light rail ridership of all the SODO stadium sports. The photo shows a 2010 scene on a Link rail car before the opening day game.
Seattle Seahawks games also bring a surge of light rail ridership. The chart just above shows two high points on two Seahawks game days with two red dots, one from 2011 and one from 2013.
PITF has compared light rail ridership on game event days vs non-event days during the four months of June through September, 2012. Light rail weekdays without games average 26,843, while weekdays with games average 30,500 -- higher by 3,658, a gain of 14%. Spread across all weekdays, the Stadium boardings boost up the daily average by about 1,900. On weekends and the two holidays July 4th and Labor Day, the difference is more dramatic. The weekend/holiday average ridership on days without events came in at 19,675. On weekend event days, including the SeaFair weekend, the average was 24,650, higher by 4,975 per day, or 25% above non-event weekend days.
One way of understanding the importance of weekend game day attendance to light rail ridership is to look at the relationship between the average weekday ridership and total annual ridership, a number called the annualization factor. The annualization factor reflects the fact that forecasts involve calculations that come up with daily ridership, and that daily ridership cannot sensibly be multiplied by 365 days per year to get annual ridership. This is because not as many people use transit on weekend and holidays as use it on normal work days. In its forecasting work, Sound Transit expected this annualization factor to be 304. Before seeing the Link Light Rail experience, the agency forecast 26,600 average weekday ridership in 2010 and multiplied by 304 to reach the annual forecast of 8.1 million. Instead, Sound Transit experienced 7 million annual riders with a weekday average ridership of 21,000. You have to multiply 21,000 by 333 to get a 7 million annual number. This means that Sound Transit light rail is now seeing a greater reliance on weekend riders than was expected earlier.
Furthermore, as noted in the July 2012 draft of the Before and After Study, "Central Link experiences the most ridership during the summer months, due to an increase of airport travelers during the cruise ship / tourist season, and due to events in downtown and at the stadiums."
One irony arises from the historical fact that the formerly forecast daily average of 45,000 per day was based on a system plan that did not even include Stadium Station and its game day riders. Stadium Station was at first to be deferred until after construction of the Initial Segment was completed in 2009, but that decision was reconsidered in 2005 and station put back into the construction plans for opening in 2009.
For the 2011 revised forecast, Sound Transit had 8.3 million annual riders equivalent to 25,000 per weekday average. The ratio implies an annualization factor of 332, which means the agency is learning from and embracing its real world ridership experience. The 2012 budget forecast for Seattle light rail is 8.4 million annual riders equivalent to 25,455 per weekday average, revealing an annualization factor of 330, coming down a bit but still way above 304 because weekend ridership is so important. The 2013 budget forecast for Seattle light rail is 9.2 million annual riders equivalent to 27,900 weekday average, indicating the annualization factor being held at 330. The actual annualization factor revealed in 2012 is 335, which indicates Sound Transit is sandbagging its 2013 forecast to be below what the actual is likely to be, given the weekday forecast is accurate.
PITF co-founder John Niles critiqued Sound Transit's light rail ridership forecasts in an essay he wrote for the online news blog Publicola, June 1, 2011, followed up by a TV interview of him and Sound Transit Board Member Larry Phillips by Seattle Channel's C.R. Douglas broadcast on City Inside/Out June 10, 2011, and available for web streaming. Shortly thereafter, in a Seattle Transit Blog posting on June 22, 2011, a Sound Transit staffer described in technical terms why the segment of light rail in Southeast Seattle, that is, Rainier Valley, was "most definitely underperforming compared to projections."
All train rides were free over a year ago on the opening weekend July 18-19, 2009, and the ridership estimate counted by hand for these first two days of light rail was 92,397, with a machine count of 66,792. These two free days -- with boarding levels similar to the 45,000 per day forecast for 2020 even if further line extensions were not completed by then -- are not included on the charts. Regular adult fares are $2.00 to $2.75, depending on distance traveled, up by 25 cents as of June 1, 2011, an increase that was expected to dampen ridership by 1.2%, according to the Sound Transit analysis results from 2010 posted here. Passenger loads during the opening fare-free weekend showed Seattle's light rail certainly has the capacity to deal with the original 45,000 forecast ridership in a day.
It's been pointed out to PITF that back in 1996 when Sound Transit was winning its initial voter approval for taxing authority, the agency forecast that Seattle light rail by 2006 would have 105,000 riders per day. Of course, as of 2013, citizens of the region know that the agency is now planning to take until about 2021 to complete and open the system from NE 45th to S 200th that is supposed to carry that many people.
The pie chart shown next provides a comparison of Link Light Rail ridership for a strong recent month, July 2012 with the ridership for other modes in the same month. King County Metro clearly dominates the transit patronage picture. The comparison is somewhat unfair, because light rail and other rail services run on a single line, while the Metro Bus network covers all of urbanized King County. Central Link can be considered a single line in the King County Metro service territory.
On the other hand, the $2.5 billion Sound Transit capital expenditure over a decade to construct and buy vehicles and equipment for the light rail line has vastly exceeded the approximately $100 million per year capital expenditure for Metro Bus equipment and facilities over a comparable time period.
The next chart shows how monthly light rail ridership in Seattle evolved in its first months of life compared to ridership on the Tacoma Link streetcar and the South Lake Union streetcar.
As one can see, Central Link light rail carries far more people than the streetcar lines shown, and is also showing very wide variations from month to month. The summer tourist bump is somewhat visible on the graph line for the Seattle streetcar of the present day (green line to the right), as well as very visible on the former waterfront trolley car, now discontinued (green line to the left).
The ridership forecast made by Sound Transit ten years ago for this line as a justification for Federal funding was made station by station. The 21,400 total daily passengers making up the ridership gap below forecast as of last summer are shown for each station in the next chart. The Airport Station is not included because its ridership is already above the 2020 forecast by about 2,000 per day!
The stations from Tukwila to Beacon Hill can be considered the commute shed for Seattle's light rail, and the daily rider boarding shortage shown below for this series of stations comes to 9,900 per day. The resulting 21,400 daily ridership gap below the 2020 forecast of 45,000 is by simple arithmetic not being made up from the 1,900 daily weekday average ridership gained in summer from Stadium Station plus the 2,000 unexpected daily boardings at the Airport.
As another source of insight, consider pre-Link weekday average boarding counts for three King County Metro lines covering part of the same or parallel corridors: Route 48: 13,800, Route 7: 11,000, Route 194: 4,800. Link daily ridership beats any one bus line, but of course a comparison should be based on changes in ridership across the entire portion of the network as reconfigured after the train line opens, for the following reason:
General transit operating philosophy in bus-rail combined systems is to feed as many bus lines as practical to rail stations in order to deliver bus passengers to what is expected to be a faster, higher capacity mode. Some one-seat bus rides become bus and rail journeys with a transfer during the trip.
Depending on the routing and frequency of feeder buses, as well as the route, frequency, and capacity of the train, a transit journey after the advent of rail may or may not be faster and more comfortable for a particular customer than the all-bus predecessor. It is the aggregated response of the entire market to the changes brought by a new rail line that makes for success or failure of a project like Link. This effect can be assessed through looking at ridership trends for bus plus rail in the entire corridor that light rail serves.
PITF's overall estimate of light rail riders as of Fall 2010 who were former bus riders was approximately 60% in the Airport market and the Rainier Valley market served by Link, based on raw ridership data collected by Metro and Sound Transit in Fall 2008 and Fall 2010. The drop in bus ridership divided by rise in rail ridership (from zero) came out to be 55% for routes to SeaTac Airport and 67% for Rainier Valley to downtown. This calculation suggests that 55 to 67 percent of rail riders are former bus riders.
Sound Transit has conducted a "before and after study" of Link light rail comparing Fall 2008 with Fall 2011. It has not been published yet, although a draft is in the hand of FTA for a review. This study is a requirement of Federal Transit Administration as one condition of the $500 million construction grant awarded in 2003. A copy of the draft has been obtained by PITF and is being examined for additional insight.
King County Metro did its own study in the fall of 2010, and from this study the drop in bus ridership on parallel Metro bus routes that occurred simultaneously with the ramp up of rail ridership can be extracted.
Note: Sound Transit states that the daily readings charted on this page are estimates and subject to revision.
Through field observations, PITF estimates that Sound Transit is using photoelectric "beam" passenger sensors above the eight doors on twenty of the first 62 rail cars to be put into service and then extrapolates to all the cars on all runs during a service day. This way of counting passenger rail boardings is standard in the U.S. transit industry. Following further revisions, numbers similar to the above will be Sound Transit's official report on passengers served provided to the public and U.S. Government. Sound Transit's quarterly ridership reports to the public are posted here.
Transit also compiles
boarding and exiting counts of customers at the various light rail
stations. February 6, 2010 through February 4, 2011 is
Station boarding & exit data recv'd May 2013: Feb. 5, 2011 to Feb. 17, 2012 Feb. 18, 2012 to Feb. 15, 2013
Using the data for Summer 2012 and Fall-Winter 2012-13, the drop from the light rail summer ridership high can be observed at the station level. The largest drops occur at both ends of the line, Westlake Center and the Airport, and at Stadium Station. Ridership actually rose a bit at Beacon Hill and Mount Baker, probably because of student ridership.
When all the stations are in view and clustered into geographic groupings, as below, the seasonality of ridership is evident, although the patronage at the five residential neighborhood stations in Seattle between and including Beacon Hill to Rainier Beach shows no seasonality and ongoing growth.
Photo of the automated passenger counting electronic eye on Link rail cars number 101 to 110, about a third of the first cars put in service. As of October, 2010 cars numbered 111 to 135 do not have these counters installed. PITF estimates that cars numbered 136 through 145 of the next 27 cars have counters installed, but observed cars 146 and above do not. Total boarding counts are extrapolated from numbers recorded on the cars where the counters exist.
Return to the Public Interest Transportation Forum home page.
Last modified: November 24, 2013