This article has been updated with improved information.
The original version of this article, posted May 24, 2007, used schedule times from the #5 bus from Spring, 2004, which was during the time period in which the Yellow Line was being constructed, and therefore slowed bus travel times along Interstate Ave.
This revised version (posted June 8, 2007) uses #5 data from the fall of 2000, which is prior to Yellow Line construction and, according to TriMet, represents the highest ridership quarter for the #5 bus. (The general difference is that #5 trip times improve by 3 minutes across the board.)
In the author’s opinion, the overall conclusions have not changed, although the numbers aren’t quite as favorable to the Yellow Line as in the original version of the article.
Any reader comments posted prior to June 8, 2007 should be evalauted in the context of the original article.
In a number of discussions here at PortlandTransport, questions have been asked and debates have arisen regarding the Interstate MAX Yellow Line.
In those past discussions I’ve tried to answer questions about travel times, ridership, quality of service, etc. using easily-found data from online sources. More recently, however, TriMet has responded to information requests to fill in some gaps in that information, which makes better before-and-after comparisons possible.
In these two posts, the first on travel times and ridership and the second on the pedestrian environment (coming soon), I’ve attempted to gather what I know about the Yellow Line, Interstate Ave., and the original #5 bus line into one place.
Take a ride into a land of lists, tables and analysis after the flip…
The addition of light rail to Interstate Ave. has improved travel times along that street and from downtown. However, MAX does not currently extend to Hayden Island / Jantzen Beach or to Vancouver. Riders going to/from Vancouver until recently had to take the #6 bus which did not have travel times as favorable as the old #5 (due to a different operating route and/or transfer waiting periods).
This week is a good time to compare schedules, however: C-Tran began service to the Delta Park / Vanport MAX station on May 13th, allowing for transfers to/from MAX and Vancouver-based bus service. In September, the level of service will be expanded. At that time, the #6 line will be reconfigured to provide service to Hayden Island, but not to Vancouver. (TriMet has provided me with preliminary information about the service changes and this information is subject to change.)
The following chart lists estimated peak-hour travel times (based on the nearest departure to 5pm from available schedules, also see table footnotes) between various origins/destinations using different travel methods.
|Origin/Destination||MAX Yellow Line||Original #5 Bus1||Current #6 Bus – Full Route||MAX / #6 Combo – Transfer @ Lombard2||C-Tran/MAX Combo – May – Transfer @ Delta Park3|
|Downtown Portland4 / Lombard||24 min||29 min||39 min||—||—|
|Rose Quarter Area5 / Lombard||13 min||17 min||25 min||—||—|
|Downtown Portland / Jantzen Beach||—||38 min||51 min||48 min||—|
|Downtown Portland / Downtown Vancouver||—||46 min||59 min||56 min||45 min|
|Rose Quarter Area / Jantzen Beach||—||26 min||37 min||37 min||—|
1Based on Line 5 schedule published September 3, 2000.
2Includes 12 minutes transfer waiting time, apparently the worst-case scheduled wait for this peak period.
3C-Tran Route #44 was used for this table… Routes 41 and 47 also serve Delta Park as of May but only on a couple of runs each day. Travel time includes approx. 8 minutes transfer waiting time.
4The following “Downtown” starting points were used based on the printed schedules: MAX – Pioneer Courthouse Square; #5 – SW 6th & Salmon; #6 SW Columbia between 5th and 4th less 2 minutes estimated offset (based on TriMet online trip planner results) to get closer to other starting points.
5“Rose Quarter Area” for the #6 is NE Grand & NE Pacific.
As can be seen from the table, for trips along the route common to both MAX and the original #5, travel times have decreased by up to 5 minutes (20%) for Downtown-to-Lombard and 4 minutes (30%) for Rose Quarter-to-Lombard.
However, for riders who regularly go to Jantzen Beach or Vancouver, the trip times via the #6 bus are worse than the old #5, depending on whether it is taken as a direct route from downtown, or a transfer is involved — Worst case: 13 minutes (28%) longer Downtown-to-Vancouver (now a full hour), or 11 minutes (42%) for east-side trips. In the best case, where transfers between MAX and the #6 align perfectly, travel times can be competitive with the old #5, but the likelihood of routinely making such great connections is low.
The new C-Tran connecting service appears to actually improve travel times over the original #5, however this is only true for trips where the transfer wait time aligns well. C-Tran only runs service half-hourly, and there is no mid-day service. A worst-case trip of just missing your transfer and waiting 29 minutes for the next departure clocks in at a whopping 66 minutes… 7 minutes worse than the current #6 and 20 minutes worse than the old #5. On the other hand, a perfect alignment of an arriving MAX train with a departing C-Tran bus 1 minute later results in a travel time of 38 minutes, shaving 8 minutes off of the old #5’s schedule, but it would be unwise to plan a trip around such tight margins. It should also be noted that southbound riders from Vancouver do not face as much of a time penalty for missed transfers because MAX runs 3X more often than C-Tran.
In September, C-Tran service, based on preliminary information, will increase to 15 minutes, which will should make the average trip time to Vancouver better than the #5 most of the day, but service to Jantzen beach (served by the #6) will be mostly unchanged.
Another factor to consider is the average wait time for someone wishing to travel without consulting a schedule. At peak times, the original #5 bus departed downtown every 7.5 minutes (8 departures between 4:30pm and 5:30pm), for an average wait time of 3.75 minutes. The Yellow Line now departs every 10 minutes in the same time period, for an average wait time of 5 minutes. This means an increase in average wait times of 1.25 minutes for the casual rider.
Clearly, the changes in travel time, positive and negative, have created winners and losers… The next section on ridership attempts to quantify just how many riders fall into each category.
Ridership, Part 1: Hayden Island / Vancouver
As shown above, riders going to Hayden Island and Vancouver face generally longer travel times than before MAX (except for Vancouver riders whose schedules align well with the new C-Tran service).
The question therefore is how many riders’ trips are made longer by this arrangement vs. how many non-Jantzen/Vancouver rider’s trips are improved by MAX.
The following table is based on average weekday boarding data provided by TriMet for the #5 bus’s highest ridership quarter:
|Original #5 Bus Route Segment||Ons||Offs||Total Ons & Offs|
|Vancouver||902 (12.4% of ons)||895 (12.5% of offs)||1,797 (12.5% of ons & offs)|
|Hayden Island||886 (12.2% of ons)||881 (12.3% of offs)||1,767 (12.2% of ons & offs)|
|All Other Stops6||5,467 (75.4% of ons)||5,397 (75.2% of offs)||10,864 (75.3% of ons & offs)|
6Based on Line 5 per-stop boarding counts, Fall 2000, excluding stops south of Salmon so that included downtown stops are similar to areas served by the current Yellow Line.
At first glance it appears that up to 24.7% of #5 riders face longer trips today. However, a portion of the above boardings/alightings are for trips between Hayden Island and Vancouver. Such trips are served with a similar schedule today by the #6, will be served by frequent C-Tran service in the future, and thus should be excluded from the comparison. This exclusion of “unharmed” riders can be determined by removing northbound Hayden Island boardings and southbound Hayden Island alightings from the totals, and placing those values into a separate category:
|Original #5 Bus Route Segment||Total Ons & Offs||Adjustments for Local Vancouver/Hayden Island Trips||Adjusted Total Ons & Offs|
|Vancouver||1,797 (12.5%)||Southbound Offs: -196||1,601 (11.1%)|
|Hayden Island||1,767 (12.2%)||Northbound Ons: -261||1,506 (10.4%)|
|Local Vancouver/Hayden Island||—||+457||457 (3.2%)|
|All Other Stops||10,864 (75.3%)||—||10,864 (75.3%)|
From the above table we can see that 75.3% of existing #5 riders saw a trip time improvement after MAX opened, while 21.5% saw trip times lengthen (or saw increased uncertainty due to transfers). The 3.2% of riders who travelled between Hayden Island and Vancouver saw no major change.
Where did the #5 Vancouver/Hayden Island riders go? After the #5 was replaced by MAX, the #6 route was changed to add service across the Columbia.
Counts on the #6 (Fall ’06) show comparable boardings to the #5 (Fall ’00):
|Route Segment||#5 Boardings||#6 Boardings|
This suggests that the Vancouver riders, despite the longer travel times, did not elect to switch to another mode of travel, however some Hayden Island riders may have been lost. (Besides trip times, reasons could also include economic, employer incentives, parking, car availability, ability or desire to drive, etc.)
What about with regard to new ridership? The latest Winter, 2007 averages for the Yellow Line show 13,040 weekday boardings (a gain of 80% over the complete #5 ridership, and gain of 138% over the corridor common to both lines. That could put the percentage of people with longer trip times vs. the people benefitting from MAX in the 11% range, depending on who the new riders are and the nature of the new trips.
But just how are those new MAX riders accounted for? Are they all from Fareless Square? The next section attempts to answer some questions about current Yellow Line ridership.
Ridership, Part 2: A Look at Where People Board
Some critics have questioned whether the increase in Yellow Line ridership can be attributed to boardings/alightings occurring within the downtown Fareless Square area. Downtown, the Yellow Line shares trackage with the current Red and Blue lines, and if the criticisms are true, such service could possibly have been provided by simply increasing the number of Red or Blue line trains. (In 2009 when the transit mall reopens, the Yellow Line will shift to a North-South mall alignment, sharing tracks with the new Green Line.)
The following table explores where people boarded the #5 bus vs. where people now board the Yellow Line (where the routes were common), along with the total number of riders in each segment. I’ve put the Rose Quarter TC station outside the list of downtown Fareless Square stops, even though it technically is a fareless stop as well. The reason for this is the assumption that Yellow Line boardings at the Rose Quarter are mostly for the purpose of making transfers in the Transit Center… the primary destinations in the area (Rose Garden Arena, Memorial Coliseum, Oregon Convention Center) would only generate significant Fareless Square traffic during events, and only for riders originating from downtown. However, this Rose Quarter data is listed as its own line item so that readers can make their own determination.
|Route Segment||Yellow Line Boardings||#5 Bus Boardings|
|Downtown||4,550 (38%)||1,661 (30%)|
|Rose Quarter TC||1,285 (11%)||880 (16%)|
|Remaining N. Portland Stops||6,028 (51%)||2,926 (54%)|
|Totals||11,863 (100%)||5,467 (100%)|
The above table addresses one aspect: The ridership growth on the Yellow Line is distributed roughly proportionately across the entire line, although of course the Hayden Island and Vancouver Ridership dropped to zero (much of this ridership was shifted to the #6 line). The segment entirely outside fareless square maintains over 50% of the boardings.
But still, just how many Yellow Line rides are fareless? Fortunately, it is not difficult to arrive at an answer… Unlike the Red and Blue lines, the Yellow Line terminates within Fareless Square and requires all riders to deboard at the end of the line.
For inbound trains, we can count all downtown boardings as fareless (riders must deboard somewhere within Fareless Square) and compare with total alightings. For outbound trains, we can count all downtown alightings… by definition these riders got on the train in Fareless Square. The following list breaks it down, again under the assumption that most Rose Quarter boardings are in fact transfers.
- Downtown Outbound Boardings: 3,596
- Downtown Outbound Alightings: 819 (22.7%)
- Downtown Inbound Alightings: 3,731
- Downtown Inbound Boardings: 954 (25.6%)
This tells is that approximately one quarter of all downtown boardings/alightings are truly fareless, or less than 10% of the total Yellow Line ridership. Clearly, the bulk of the ridership growth has not come from the downtown portion of Fareless Square.
But what about the Rose Quarter? For argument’s sake, let’s assume that all boardings/alightings at the Rose Quarter station which could possibly be fareless are in fact fareless. For inbound trains, this means all Rose Quarter boardings, and for outbound trains, this means all Rose Quarter alightings. Here’s the data for those two values:
- Rose Quarter Inbound Boardings: 116
- Rose Quarter Outbound Alightings: 386
Clearly only a relative few riders are taking advantage of free Yellow Line trips to/from downtown and the Rose Quarter. At the very, very most, assuming none of these riders transfer to a bus, they represent 4.2% of the total ridership. Worst-case (the term “worst”, of course, assumes you believe legal fareless riders are a bad thing), fareless riders make up about 14% of all Yellow Line ridership, but if the Rose Quarter rider profile is in keeping with downtown percentages, the total figure is lower, closer to 10%-11%.
Ridership, Part 3: The Weekend
TriMet, compared to other transit agencies, has a higher rate of weekend ridership. Does adding light rail to a corridor significantly boost this weekend ridership?
The following table compares the ratio of weekday to weekend ridership for the #5 bus and the Yellow Line:
|Line||Weekday Boardings||Saturday Boardings||Saturday Ratio||Sunday Boardings||Sunday Ratio|
|#5 Bus (Fall 2000, All Stops)||7,314||5,916||81%||4,260||58%|
|Yellow Line (Fall, 2006)||11,863||10,663||90%||6,920||58%|
|Yellow Line (Winter, 2007)||13,040||10,660||82%||6,850||52%|
I’ve included two sets of numbers for the Yellow Line, one from Fall and one from Winter. The numbers available thus far may indicate that weekend ridership is weather-sensitive. The Winter weekend MAX ridership proportions are about the same as the #5 bus in the fall, but the fall MAX ridership for Saturdays is higher, within 90% of normal weekday boardings.
Based on the data and analysis above, the following statements can be said about the Yellow Line:
- Travel times improved by 20-30% for the corridor served.
- 21.5% of the original #5 bus riders saw an increase in travel time and/or were inconvenienced by transfers.
- 3.2% of bus riders saw no significant change (rides purely between Jantzen Beach / Vancouver)
- Most (75.3%) bus riders benefitted from the decrease in travel time.
- The increased travel times for some riders did not necessarily lead to the loss of those riders.
- The majority of the new Yellow Line ridership is not caused by Fareless Square
- Between 10% and 14% of Yellow Line rides are fareless.
- Utilization of the Yellow Line by place-of-boarding is roughly proportionate to the original #5.
- Weekend ridership may indicate light rail boosts Saturday performance, but less so on Sunday.
September 3, 2000 Line 5 Schedule (Adobe Acrobat .PDF)
Fall, 2000 Line 5 Weekday Ridership (Adobe Acrobat .PDF)
Fall, 2000 Line 5 Saturday Ridership (Adobe Acrobat .PDF)
Fall, 2000 Line 5 Sunday Ridership (Adobe Acrobat .PDF)
Fall, 2004 Line 6 and Yellow Line Weekday Ridership (Microsoft Excel Spreadsheet .XLS)
Fall, 2006 Line 6 and Yellow Line Weekday Ridership (Microsoft Excel Spreadsheet .XLS)
Winter, 2007 Yellow Line Ridership Summary (Microsoft Excel Spreadsheet .XLS)
C-Tran Route 44 Map / Schedule, May 2007 (Adobe Acrobat .PDF)
Line 6 Map, May 2007 (Adobe Acrobat .PDF)
Line 6 Schedule – To Portland, May 2007 (Adobe Acrobat .PDF)
Line 6 Schedule – To Vancouver, May 2007 (Adobe Acrobat .PDF)
Yellow Line Schedule – To Expo Center, May 2007 (Adobe Acrobat .PDF)