Last week, I wrote about how the new medical marijuana system created under the Cannabis Patient Protection Act (SB 5052) shut out many of Washington State's medical marijuana dispensaries. The law required that dispensaries obtain a recreational license to continue operation, but a loophole in the application process created a black market for dispensary employee pay stubs, allowing new actors to game the system. I also mentioned that the Washington State Liquor and Cannabis Board (WSLCB) blindsided applicants by announcing a cap of 222 new licenses late in the application process.

What I didn't have space to mention was that the cap itself was based on data that even the state contractor that compiled it admits was shaky. Given that the medical marijuana market was, by Washington cannabis data scientist Jim MacRae's estimate, at one point supporting at least 800 dispensaries, and that SB 5052 mandated that the state issue enough new licenses "in order to accommodate the medical needs of qualified patients and designated providers," how did the WSLCB decide on issuing only 222 new licenses?

The decision was based on a study conducted by BOTEC Analysis, a California-based research and consulting firm that the WSLCB hired to determine just how much of Washington's cannabis market was being served by the medical marijuana industry.

According to MacRae, BOTEC's study was inherently flawed and created a flimsy foundation upon which to base a decision that would affect thousands of patients. "They systematically underestimate values commonly known to be higher," he said. "Their results should not be used to inform or guide policy or business decisions."

Indeed, by BOTEC's own admission, their analysis was, at best, an educated guess. "There's not good data [on the overall size of the market] in the past two years, so we essentially just had to make an assumption," said Steve Davenport, lead analyst on the report, when I reached him shortly after the report was released in mid-December.

Given the absence of an organized tax system on medical cannabis, which might have provided a more accurate estimate of sales, BOTEC basically had to call up dispensaries and ask for data. "There's a lot of uncertainty on both sides," said Davenport, speaking of the survey process. "When you ask someone what their revenue is, they might have a certain incentive to be dishonest." That incentive being, of course, to stay on the good side of the Department of Revenue.

BOTEC surveyed 40 stores and used the responses to build a model for average dispensary sales. They then applied that generalized model to about 170 dispensaries, adjusting for total open hours, days of operation, and other factors.

Their final estimate was based on the survey data combined with the data generated by the model, meaning that the estimate for the size of the state's entire medical market was based on iffy data from about 200 stores. Ultimately, BOTEC listed just 403 operating dispensaries in their report, far short of MacRae's estimates. Which is his biggest criticism: The size of the market was based off a significantly diminished data set.

"The use of CURRENTLY OPEN DISPENSARIES as the fundamental proxy by which to 'size' the medical Cannabis market is inappropriate in an environment where dispensaries have been the focus of increased enforcement and closure over the past year," MacRae wrote on his blog in a refutation of the study.

"We're not expecting that it's exactly right," Davenport admitted. "The important thing, too, is that that was just a snapshot in time. Even before the WSLCB issued new rules, a lot of dispensaries were closing down. If we had done our study again a few months later, you would have seen a smaller medical figure."

But just because dispensaries were closing—or had been forced to close—doesn't mean that the medical market was shrinking.

Even more troubling is that the WSLCB knew there were problems with BOTEC's data. In an e-mail sent by Bob Schroeter, the WSLCB's director of public records and support services, to BOTEC chairman Mark A.R. Kleiman and president/managing director Brad Rowe, Schroeter raised several red flags with BOTEC's findings.

"The LCB is concerned that the number of dispensaries BOTEC has estimated is very much lower than all other estimates we have seen without an adequate reason as to why," Schroeter wrote in the e-mail, which was sent in November and obtained by The Stranger. "Equally concerning, the estimate of product leaving dispensaries also appears low and lacking supporting information... The LCB is concerned that without seeing estimates of the number of patients or the amount of cannabis that went 'out the door' for dispensaries, we cannot use this report to estimate the need for additional stores."

If the study was so far off the mark, and the WSLCB knew it, why did they use it? Brian Smith, WSLCB director of communications, declined to comment on Schroeter's e-mail, saying the report was "based on the best available data at that time." Reached more recently, Davenport said that the LCB's concerns had been addressed in the final draft of the report. MacRae, however, was still skeptical.

"BOTEC should have known better," he said, "and it would be easy to suspect that they may have been pandering to a perception that a small estimate was what their client desired. A small estimate made the LCB's job easier. It allowed them to decrease the number of stores that their overworked enforcement officers and non-armed staff have to service."

That's certainly a possibility—but regardless of the reason, it's clear that Washington's medical cannabis patients are suffering as a result. "The results of the BOTEC study are, essentially, meaningless," said MacRae. "Using BOTEC's results (and even assuming them to be usable) will hurt people in 2016."