Agreeing with Mel


It’s not often that Mel Zucker and I agree.

But I agree that TriMet turning down his request for boarding/deboarding counts on the basis of security is off the wall. In the past TriMet has often provided such counts to my neighborhood when we were evaluating routing options or potential stop re-locations.

Can we have a little common sense please.


32 responses to “Agreeing with Mel”

  1. I’ve got a pending request in with them for and updated version of the spreadsheet they supplied me from FY2005. I hadn’t heard back and was too busy to follow-up… maybe a phone call is in order?

    All that the boarding info will tell you is what the most popular stops are (not too difficult to tell from any cursory examination.) It doesn’t tell you what time of day is the highest concentration, or what demographics of people board there, etc., so as far as “security” issues go, I think withholding such information would be hard to justify.

    – Bob R.

  2. Oops… I should have read the original article fully before posting.

    It turns out that Mel was asking for specific time-of-day numbers at specific locations. In all fairness, I still think that information ought to be available and isn’t much of a security risk.

    – Bob R.

  3. You’re right; all a terrorist would have to do is ride any line during peak hours to figure out the most crowded areas. Given his track record and misunderstanding of Interstate Avenue LRT data when it opened, I’m sure Zucker would not completely understand any new data TriMet gave him. TriMet knows this and I’m fairly sure this is why they don’t want to give it to him.

    If Zucker wants to know ridership during the peaks between Forest Grove and Hillsboro, spend a couple of hours watching buses go by during the peaks; if you see standees, you know the seated load is generally exceeded. Simple.

    As far as I know, also, TriMet has no plans for an LRT extension to Forest Grove; this plan is being pushed by Forest Grove. It should be a streetcar given the relatively low volumes, but that’s another post…

  4. JK –

    The real numbers are not an embarrassment, they are among the best in the nation, and I’ve used the info here many times to debunk misinformation and misconceptions about Portland-area transit. I’ve also used the data to suggest changes and improvements.

    – Bob R.

  5. Given his track record and misunderstanding of Interstate Avenue LRT data when it opened, I’m sure Zucker would not completely understand any new data TriMet gave him. TriMet knows this and I’m fairly sure this is why they don’t want to give it to him.

    That’s not the point.

    TriMet is a governmental agency that, based upon the U.S. Constitution, that government functions by and for the people.

    This is America where people are free to voice concerns about government, and if that means TriMet needs to spit out numbers, it has an obligation to do so. If TriMet is right, then Mr. Zucker can then say “oh, OK, people do want this.” If TriMet is wrong, then TriMet has an obligation to be responsive to the needs of the people.

    BTW – I asked for a copy of TriMet’s recent proposed budget, and have yet to hear from them. Are budget numbers now confidential too? (Never mind that EVERY OTHER GOVERNMENTAL AGENCY IN OREGON) publicly posts such information on their webpage.

    Unfortunately TriMet has a credibility problem brewing. It has no problem giving “good news” but routinely witholds “bad news”. We (the taxpaying public) helps support TriMet and have a duty to be informed about how our government works. It’s kind of hard when our government operates in secrecy.

  6. You missed MY point; I’m just trying to ascertain a motivation, not justify the withholding of information.

    OF COURSE, TriMet should give Zucker the data, even if he is likely to misunderstand it as he did when the Interstate Avenue LRT opened.

    In that case, Zucker forgot that for every boarding, there was also an alighting, and visa versa; thus he underestimated LRT ridership on the new route by 50%. Certainly any misinterpretations on the part of Zucker can be corrected here, since many posters are actually have some understanding of how to interpret transit data. The TriMet folks can also respond to any misinterpretations.

  7. http://portlandtransport.com/archives/2007/04/agreeing_with_m.html#comments
    Michael D. Setty In that case, Zucker forgot that for every boarding, there was also an alighting, and visa versa; thus he underestimated LRT ridership on the new route by 50%.
    JK: Care to provide evidence of that?

    As I understand it, the main difference between his numbers and those of Trimet centered on wether to count boardings in downtown as indicative of the Interstate line’s effectiveness.

    Thanks
    JK

  8. Michael D. Setty In that case, Zucker forgot that for every boarding, there was also an alighting, and visa versa; thus he underestimated LRT ridership on the new route by 50%.
    JK: Care to provide evidence of that?

    As I understand it, the main difference between his numbers and those of Trimet centered on wether to count boardings in downtown as indicative of the Interstate line’s effectiveness.

    Thanks
    JK

  9. While I agree that it is a stupid policy, I highly doubt that TriMet is turning down his request flat… For instance, during the middle of the Cold War, Ken Deyes was working as a researcher, researching uranium mines, (obviously a touchy subject,) and after getting a security clearance and spending a couple months dealing with the bureaucracy, he got the data he wanted. If Mel really wants the data, he could probably get it, although it might be more work to get it through TriMet, than just “on” TriMet.

    Now, obviously, one of the conditions of his access to the data is (probably) that he wouldn’t be allowed to publish it in raw form, and given what I’ve seen of his statistical skills so far, (he couldn’t figure out that 65 mph in a car with airbags is safer than 55 in a car without them,) I’d take any study he created with a grain of salt, but that is a topic for another time.

  10. To respond to Karlock:

    In 2004, Zucker and the people who worked with him right after the startup of the Yellow Line thought there were only 5,000+/- total rides per day on the Interstate Avenue line, and as a result accused TriMet of lying.

    Right after Yellow Line startup, Zucker et al did their own ridership counts, and counted somewhat more than 10,000 total boardings AND alightings along Interstate Avenue from Rose Quarter north–nothing south of that point–which he thought was somewhat more that 5,000 or so total transit boardings. However, the correct total was over 10,000 total boardings (passengers) because about half of total passenger movements were ALSO occuring downtown, e.g., where he and his henchmen didn’t count! There was a small “along the line” traffic, but that was less than 10% of the total.

    Zucker’s odd reasoning was that since in downtown a lot of trips on Yellow Line trains would be within Fareless Square, they didn’t count. But he neglected to remember that the vast majority of those trips would be trips to/from Yellow Line points from Rose Quarter north.

    Based on this, I think Mel’s understanding of transit statistics is marginal, at best.

  11. Michael Setty: Zucker’s odd reasoning was that since in downtown a lot of trips on Yellow Line trains would be within Fareless Square, they didn’t count. But he neglected to remember that the vast majority of those trips would be trips to/from Yellow Line points from Rose Quarter north.

    But those riders were counted if they deboarded or boarded along the Yellow Line Extension. If someone boarded across the street from the Central Library to ride home to somewhere near Interstate and Lombard, he would be counted as actually using the Yellow Line extension. If he boarded at the library and deboarded at the Rose Quarter TC to catch a bus, or to another location downtown such as Kells Pub, that’s a ride that could just as well be on a Blue Line or Red Line, on the same tracks that have been in downtown since the mid-80’s. And that ride in no way justifies the cost of the Yellow Line extension.

    What you fail to acknowledge is that by counting people deboarding and boarding on Yellow Line extension stops, and only at those stops, true use of the extension line is determined.

    Perhaps one day Tri-Met will actually put in sensors to count riders. A few years ago someone on a newsgroup claimed that this was being done, but Tri-Met then told me that these had yet to be installed. People need to get accurate information.

    Bob Tiernan

  12. My understanding is that MAX has automatic people-counters.

    But we’re getting off the topic here. Let’s steer it back to whether TriMet should provide the data.

  13. First of all, those additional (Yellow line) trains downtown might not be there if the Yellow line extension wasn’t built. And remember that the Yellow line poorly serves riders to/from Vancouver or east of I-5.

    But getting back on-topic, this hiding of easy-to-gather information makes you wonder if next they’ll declare that info from Transit Tracker and knowing that a bus or train is about 3 minutes away is sensitive. Or the detour route that a bus is taking. And it does not soothe me that TriMet is taking “secret directives” from the Feds.

  14. And it does not soothe me that TriMet is taking “secret directives” from the Feds.

    I think this has more to do with post-911 hysteria than it does Trimet trying to hide data from people. The idea that keeping boarding data from the public will prevent terrorist attacks is not even the most egregious nonsense from the Department of Homeland Security.

  15. Jason McHuff: First of all, those additional (Yellow line) trains downtown might not be there if the Yellow line extension wasn’t built.

    Bob T: Those trains could have been added
    without spending hundreds of millions on the Yellow Line extension.

    Jason McHuff: And remember that the Yellow line poorly serves riders to/from Vancouver or east of I-5.

    Bob T: The line wasn’t built or designed for serving people, but I’d still like to look into what lines could be feeder lines (other than those on N Lombard). But I think that those living in Northeast close to K-worth etc probably head south to the Blue Line or where they can catch a Red Line train. However, N-S routes aren’t numerous.

    As for the date, of course Tri-Met should provide it. But the info overall is available in ways they don’t make a point of. Ridership, for example, has not gone up in the past few years. Tri-Met admits this in their audit w/o actually coming right out and saying it.

    Bob Tiernan

  16. Bob T.

    [personally directed remark removed]

    Let me see if I can make this clear, again. I am doubtful I’ll succeed. Perhaps this is “off topic” but I think its important that readers understand this most basic type of transit data.

    First, for EVERY transit trip on ANY transit line, there is ONE boarding and ONE alighting. The previous Interstate Avenue bus carried 6,500 daily passengers; that is, a total of 6,500 boardings and 6,500 alightings, or a total of 13,000 passenger movements.

    If 100% of these bus riders were headed to downtown Portland south of the Rose Quarter station, that means there would be 3,250 boardings and 3,250 alightings DOWNTOWN in the area south of Rose Quarter, e.g., across the bridge. THIS ALSO MEANS 3,250 total boardings and 3,250 total alightings from Rose Quarter north to the end of the line, e.g., 6,500 total. In reality some of these people stayed along Interstate Avenue and transferred to/from the Blue and Red Lines, but this fact doesn’t change the point (perhaps only 80% were to/from downtown–you don’t know for sure for any given transit line without a transfer analysis or O-D survey, separate animals from a boarding and alighting count.)

    In a similar vein, Zucker et al counted 10,000+ total LRT boardings and alightings from Rose Quarter to the end of the new line. This is a rather significant increase from the previous bus service. Even though they didn’t count boardings and alightings south of Rose Quarter across the bridge, particularly shortly after startup when the counts were made.

    Another way to consider this is that for the roughly 90 trains in each direction operating south through Rose Quarter and arrving northbound from downtown, there was about 50-60 people per train. Well, for each of the 50-60 people per train who passed Rose Quarter southbound, your counters did indeed count their BOARDINGS but not their ALIGHTINGS downtown. Similarly, for the 50-60 people who arrived at Rose Quarter northbound, your counters did indeed count their ALIGHTINGS from Rose Quarter north, but not their BOARDINGS downtown!

    To get a count of “actual use of the line,” you have to double the total counts for each boarding and alighting for those passengers who didn’t begin and end their trips on the segment between Rose Quarter and the end of the line.

    That is, TOTAL boardings and alightings were over 20,000, including those getting on and off downtown. You didn’t need to do total counts including downtown to know overall patronage (10,000+) because YOU KNOW there is one boarding AND one alighting for each passenger trip.

    Counts downtown would have given you two additional pieces of information: (1) ons and offs at each downtown stop; and (2) the Yellow Line trains’ share of MAX trips made within Fareless Square.

  17. This is a rather significant increase from the previous bus service.

    How do we know that; TriMet isn’t releasing that data!

    Since it appears that some people have the data and some people do not, I think that is inherently unfair (TriMet serves ALL people of the region, not just those that it subjectively chooses to). So:

    Let’s see the ridership on Line 5 dating back to 1990 until the route was terminated, and from Interstate MAX from then to the present.

    TriMet should be able to easily and effortlessly provide daily average (weekday, Saturday and Sunday/Holiday) boardings (since we know there is one boarding and one alighting for each line, all we need is boarding numbers) for each year. We don’t need day-by-day or month-by-month breakouts. Is that too much to ask for?

  18. How do know that?

    Because TriMet has to collect ridership data, including boardings and passenger miles by mode, consistent with the statistical validity standards spelled out by the National Transit Database (NTD) mandatory annual reporting of the Federal Transit Administration (FTA) if they want to continue to receive Federal funds. If TriMet, or any other transit operator, fails to meet these requirements, three things happen: (1) they put a prominent mark on the given transit operator’s data as “Questionable,” (2) the FTA may hold up funding; and (3) in extreme cases where the statistics have been questionable and “dinged” in the Tri-Annual Review of transit operators (making sure regulations and requirements are met), no grants move at all until deficiencies have been corrected to the satisfaction of the FTA.

    The Feds are sticklers for accurate NTD data, particularly revenue vehicle miles, revenue vehicle miles, total boardings and annual passenger miles because this data goes into the various formulas distributing rail and bus funding to each urban area. It is in TriMet’s interest to ensure that the data is accurate so they continue the money flow and get their maximum share within the formula factors.

    If the apparently rather marginalized transit critics of Portland, Oregon, don’t get this, then I give up trying to explain anything; why bother?

  19. Michael, how do YOU know that ridership on I-MAX increased over the 5-Interstate line, when TriMet refuses to provide that information?

    TriMet refuses to provide information upon request and they refuse to publicly state it, so either we need a source-check on the data (and if TriMet will release data for this line, then they need to provide it for every other bus line and MAX line too), or TriMet is obiviously hiding something, and that they are selectively releasing that information because it portrays them in a good light – but releasing the same exactly data for other routes would not, so they are claiming a false reason to withhold it.

  20. when TriMet refuses to provide that information?

    It doesn’t sound like that is the information TriMet is refusing to provide. I thought it was specific boarding information by time and location, not aggregate information for the whole route.

    It also doesn’t sound like it is TriMet that is refusing to provide that information, but that it is being prevented from providing it by the federal government.

  21. So, Erik, you think TriMet is lying? Why don’t you just come out and say so?

    From what I’ve seen, there is absolutely no evidence to indicate they were lying–rather that is what some transit critics think! If you want “direct” evidence that “I know” TriMet isn’t lying about additional ridership on the Yellow Line, it is from the peak period volumes COUNTED BY ZUCKER ET AL, which was consistent with a daily volume of 10,000+ riders, consistent with a rather typical, mundane pattern of LRT ridership observed on many other LRT routes in other cities. TriMet peak ridership on that line, as counted by Zucker et al was completely consistent with the norms observed in other cities. So is the train capacity they offer during peak hours.

    If you have specific questions about what those examples and LRT norms might be, email me and don’t bore the readers here. My associate Leroy Demery and I have gone over data from many systems in extraordinary excruciating detail in our recent transit research.

  22. Again, STATE YOUR SOURCE.

    I want to see the specific numbers, the sources for that data, and be able to view it.

    Otherwise the evidence is simply hersay – why should I believe you? Why should I believe TriMet? Because TriMet knows best?

    What is so hard about providing the actual data? It seems to be that we can run around in circles but ultimately TriMet as a government agency as a responsibility to provide open access to data, with some restriction. Do I seriously have to file a FOIA request to get this data?

    ODOT is a leader in providing open and honest information; their website is extensive with traffic and budget information, construction plans, even maintenance manuals – all on the PUBLIC website. I notice that TriMet even went so far as to remove their bus fleet information from their website…because they don’t want people to know how old the busses are, or that a large number of their busses are 40′ busses despite having routes that consistently run overcapacity?

    I don’t think TriMet is lying. I think TriMet is withholding the truth, because it might portray them in a negative light.

  23. Erik, WHAT IN THE WORLD is the difference between the thinly veiled accusations that Trimet is “lying,” or they’re NOT but they ARE “withholding the truth???” That perhaps the Portland transit REALITY doesn’t correspond with YOUR unique idea of “the truth?”

    Perhaps they don’t want Zucker to distort the information–a valid fear, methinks–but NOT a justification for obsfucation over publishing the information. So I guess I actually agree with you on this, but disagree that they’re trying to “hide” “the truth” whatever your conception of that is. A thin reed, indeed.

    FOR ONE LAST TIME, I’ve already said I AGREE that TriMet should release the data–I’m not aware of any other transit agencies that have “secret directives” from TSA not to release such basic data that anyone with two eyes can go out and obtain if they’re willing to spend a few hours–S.F. Muni has released such detailed on and off data but no one seems to care, for example.

    Why are you ragging on me?? TriMet has the data and I’ve told you why their summary data–like total Yellow Line ridership–appears to be quite accurate. I based my comments on Zucker’s ridership count data from 2004 and what I know about peak ridership patterns relative to overall daily ridership on LRT lines!!!! Ask Zucker for his information!

    BTW current TriMet reports of Yellow Line ridership is around 12,200 daily boardings, which is completely consistent with everything I’ve seen, including Zucker’s counts from 2004.

    If Zucker wants to know directly about current TriMet ridership on the 57 from Forest Grove, he can get the transit report recently completed on the topic from that city’s website, which I downloaded several weeks ago.He can also stand on the roadside for 2-3 hours on the road between Forest Grove and Hillsboro to get a pretty accurate idea of peak hour loads on that route.

    BTW, everyday TriMet riders obviously “know the truth” about service. If particular lines are overcrowded then that ought to be rather obvious to anyone who rides the bus. Do you take transit at all ever, Mr. Halstead?

  24. So, you’re special, and you can have the data.

    However TriMet won’t release the data to Mr. Zucker, or me. (I’m still waiting for a copy of the proposed FY2008 budget that they won’t release.)

    Sorry, but until TriMet starts acting like a democratic government and not like a Communist “democratic” government, I see a serious credibility problem. You can tell me a million times some number, but until I have access to the data you could just as well tell me that 250,000 people ride MAX a day, and I have no proof of such. Or 30,000. Who knows? TriMet does, but why won’t they release it?

    By the way, isn’t this Portland, Oregon? Where is the outrage over TriMet’s inability to function as a democracy?

  25. Erik Halstead:

    You can tell me a million times some number, but until I have access to the data you could just as well tell me that 250,000 people ride MAX a day, and I have no proof of such.

    Bob T:

    There are too many people who do that. One was
    saying several years ago that Tri-Met had infrared
    sensors on all MAX trains for accurate counts, yet
    Tri-Met at that very time admitted that only a
    single car had one. The person I’m thinking of
    even claimed this on his web site and in an
    Oregonian article. We need to stop this
    disinformation.

    Bob Tiernan

  26. Erik:

    I NEVER got any data from TriMet, other than the totals they announced in their press releases. Zucker’s 2004 counts were consistent with the totals announced by TriMet. What part of the fact that a strong transit critic’s data was consistent with the transit system being criticized don’t you understand?

    END OF THREAD. I have better things to do that arguing with opinionated anti-transit “true believers” who obviously cannot believe their own empirical data, particularly when the standard statistically valid methodology for evaluated said data is explained to them ad nauseum, particularly when said methods are the SAME regardless of the transit system.

  27. When there are two people who take the same data, and both twist the data to suit their own needs, and the raw data cannot be extracted through either source (or from a third party, or from a neutral party) – I tend to believe neither.

    I can only assume that since TriMet refuses to provide the raw data, I cannot believe any party with regards to Interstate MAX ridership vs. the 5-Interstate bus ridership; and therefore something is being hidden.

    I’m starting to think that George Bush and Dick Cheney are running TriMet.

  28. [personally directed remark removed. The non-personal gist is that Mel does not believe other perspectives on his yellow line counts are correct.]

    Tri Met’s numbers and our counts were in agreement with one exception. We counted the actual Yellow line. Those are the stations from Albina to the EXPO center. That’s what the money for the yellow line was spent on.

    The trips from the Rose Quarter west are part of the East side line built between 1983 and 1986. All of those passengers were being served before the Yellow line was built.

    There are no more people using transit in the corridor north of the Rose Quarter.

    The National Transit Database accepts whatever the transit agency submits. It does NO auditing or separate counting.

    The new rail cars came with counters. The older trains had to be retrofitted with counters. The counters record everyone that gets on a train or bus (a boarding) and everyone that gets off at every station. The difference between the ons and offs provide us with the load that is on the train or bus at any given point.

    The withholding of information was discussed at the last Tri Met Board meeting this past week. Hanson said he wrote Homeland Security in D.C. and they responded by saying they do NOT interpret the law as pertaining to the requested ridership information. He said the local (Salem) Homeland Security person (Jerry Hill) had told them otherwise. He did not say that Jerry Hill takes orders from Washington. When Paul Knobel (Senator Gary George’s legislative aide) spoke to Jerry Hill on the telephone, Hill told him it was up to Tri Met.

    I have heard of no other agency censuring this information, including NYC the juiciest transit target.

    Mel

  29. Meanwhile, I’m still waiting for TriMet to release its budget to me, that I asked for over a week ago.

Leave a Reply

Your email address will not be published. Required fields are marked *