What appeared to be a glaring flaw in the state's bullet train ridership forecasts was actually a "typographical error," an official said Wednesday.
A key number in the ridership model released last week should have read 0.0179, not 0.179, said Jeff Barker, deputy director of the California-High Speed Rail Authority. The mistake was in a document explaining the model, not in the model itself, he added.
The clarification could take some steam out of a brewing controversy over the authority's claim that more than 40 million people per year would ride the planned $45 billion Los Angeles-to-San Francisco line. But critics remain suspicious of how officials arrived at that figure and continued to call for answers Wednesday.
Skeptics of the state's bullet train plan have long scoffed at the ridership projections, but have been rebuffed by assertions that it was based on a state-of-the-art, peer-reviewed statistical model. Last week, they seized on newly released documents showing the model actually used by consultant Cambridge Systematics was not identical to the one the authority published online for public review.
An internal memo showed that the Bay Area Metropolitan Transportation Commission, which financed Cambridge's study, opted not to disclose the final set of revisions to the model. Doug Kimsey, the MTC's planning director, told The Daily News they were considered insignificant.
But one number in particular in the revised model stuck out to Palo Alto investment manager and amateur rail watchdog Elizabeth Alexis. It had changed from 0.003 to 0.179, a factor of 60.
The figure in question was one of dozens used by consultants to come up with official estimates of how many people would ride the planned bullet train. Specifically, it was a constant representing the importance of frequent train departures in people's decisions to take the bullet train instead of, say, a car or a plane.
If it all sounds very technical, it is. But Alexis and others said it could have influenced the authority's decision to run the trains up the Peninsula instead of through the East Bay — not to mention affecting the overall ridership estimate.
News that the number was not accurate didn't satisfy Alexis. She said Wednesday she asked the authority whether the figure was a typo before publicizing it last week, but didn't get a response until Tuesday night. Typo or not, she added, the model merits a closer look before the rail authority uses it for any more planning decisions or federal funding applications.
"The largest single risk in this project is that this model isn't right," Alexis said. "And it hasn't been peer-reviewed" in its final form.
She called for the rail authority to recirculate the model to the peer review panel before moving forward.
Barker said the authority is considering that request. He said the ultimate goal is to get the best ridership forecasts possible, though he has emphasized that they are only rough estimates at this point.
"We want to be as open as possible, so if this is something people are interested in, obviously we'll sit down with them," Barker said.
Meanwhile, a group that has used the new ridership documents to reopen a lawsuit challenging the rail project's environmental certification said it is not backing down.
"We're following up," said Stuart Flashman, attorney for the group, which includes the cities of Menlo Park and Atherton along with several activist organizations. "We've just submitted a public records act request to the California High-Speed Rail Authority and one to the MTC to get the original documents. At this point we're not sure who to trust, frankly."
Flashman added, "We had an outside consultant look at the modeling data, and the consultant identified some other problems beyond that one coefficient."
E-mail Will Oremus at email@example.com.