If a Picture is Worth a Thousand Words, then a Video Must be Worth a Billion

Happy New Year to you all!

One of my New Year’s Resolutions is to become less verbose in my postings (sure, Jim … like that will last).

The following video captures an interesting dynamic that emerged during the first 64 weeks of Flower Potency Testing by Washington’s supposedly PRAGUE Labs.

The Y-axis (vertical) shows “inflation” of reported TOTAL CANNABINOIDS … being on the red line means the lab was using THCmax (and, hence, their “inflation” is zero %).   As one moves up, the inflation increases.

The X-axis (horizontal) shows the value reported as TOTAL CANNABINOIDS (in absolute percentage terms).

The LOCATION of each bubble represents the average values for all FLOWER LOTS tested by each lab during the indicated week.

The SIZE of each bubble represents the “market share of testing” (percentage of all flower tests THAT WEEK that were conducted by each lab).

The video takes about 40 seconds to step through the 64 weeks of testing (through the end of August).  As the video runs, little “tracers” are left behind so that you can see where each lab was, historically.

A shout-out to Statpoint Technologies for continuing to create great visually-centric statistical software.

Please use the comment section to comment and discuss.  I’m not commenting on it … so I expect all of you less verbose types to fill in the void. I will likely be staying silent on the comments (unless direct questions are asked of me and/or I feel a posting needs some feedback).

If I can figure out how to do so legally, I may share some “C” with whoever is best able to estimate (FROM THE VIDEO) the week in which the LCB sent the direction to the Labs that THCmax was to be reported as the TOTAL CANNABINOID level.   Not a lot … just enough for both of us to have a good conversation together.  (Lab people are excluded from this potential contest … LCB people, too, and undercover Federal Narcs).

Note:  This will very likely be my last post on HI-Blog (at least at the highintelligence address).

I’m busily attempting to migrate my new commercial website away from my wife’s non-profit hosting account.  In that process, I lost all e-mails sent to jim@straightlineanalytics.biz between 10am on Tues Dec 29th and about 5pm on Friday Jan 1.

I had some messages from and conversations with people at the labs (and DOH) during that time that I’d like to have access to, so please re-send to me if you can.  Much appreciated in advance.

I will be completing the migration soon  (I hope).  Once I’m done, if you point to www.highintelligence.org , I believe you will be taken to the BLOG portion of my new site (it’s not up yet) at www.straightlineanalytics.biz  .  I’m trying to make sure that the old “permalinks” to each individual posting (this is my 14th) remain functional, but I’m not going to lose sleep over trying to get that specific detail working.

I’m also a bit concerned that the comment threads associated with HI-Blog may go away (that would be a shame).

I’ll likely keep HI-Blog running (comments and all) for the next week or two while I transition everything.  I’ll try to set things up such that attempts to get to highintelligence route the click through to straightlineanalytics.

I also intend to send out a 1-page survey to everyone that has engaged with the site (and other people in the industry that I know) … so if you want to be included, make a comment on this post and/or send me the e-mail to which you’d like the survey sent.  It will ask a TINY amount about what your relationship to this industry is and a bit more about the types of things you’d like to see from me in the future (and the types of things you’d be willing to put some of your very hard earned cash down for).

Look for my update on GARCN on www.straightlineanalytics.biz coming real soon now (update – Traceability data through November just became available Jan 20th . will update GARCN shortly at the new address).

Happy 2016

Check out my poetry page for 3 Seahawks Hawkus I recently posted on the Seattle Times site.

.

25 comments

  1. Jim,

    This is pretty cool, thanks for sharing! I’m going to estimate that it was sent out by the LCB week 47 or 48, hopefully I’m right, I wouldn’t mind having a conversation with you one day over Washington’s finest, to include ours 😉

    Mark
    Wicked Weed

  2. Wow! Would be nice to know which week correlates with the calendar and if the clustering near the end aligns with the fall harvest. Fascinating Jim! As long as the market is buying based on THC I’m not sure naming these labs would be a good thing. Would P/P owners gravitate toward labs giving better results and shy from those that don’t? I know it’s already happening but would it be more pervasive? It would definitely show the most scrupulous labs in relation to their counterparts. Thanks for sharing Dr. Jim!

    1. Thank-you, Aaron. Labelling the weeks by date is too much of a give-away.

      Chasing THC is one thing.
      Chasing FALSE THC is another.

      I suspect that farmers that are using the inflating labs may not enjoy the benefits of this inflation for much longer.

      There even may be a downside to them, should they choose to continue using the inflators.
      That is, if the inflators are allowed to continue serving the needs of this market without any sanctions.

      To paraphrase Mr. Lynch: “I’m just about the objectivity and PRAGUE-ness, Boss.”

        1. I suspect many of our fine Farmers are.
          This is not a negative reflection on the Farmers … and no-one should feel that way.

          It is, I believe, a reflection on some of the labs.

          No-one can control what they did yesterday.
          We can all, however, influence what we do tomorrow.

  3. Am i correct in reviewing the graph that lab N consistently has higher THC content than any of the other labs?sent

    1. Lab N is consistently inflating more than most (it is higher on the chart).
      The higher THC (TOTAL, actually) is when bubbles are over to the right side of the chart.

      1. Thanks for posting that video, Jim. Cool representation of events. I already knew the identities of 6 labs, but now I know the identities of all 14!!! The video reveals some other interesting things about your data.

        I want to be clear that Lab N has never been inflationary to its customers or to the State database, and let me explain.

        Agricultural testing labs return results to their customers using a document called a Certificate of Analysis. In the case of our industry, the Certificate of Analysis is an official document in all transactions, and certain numbers on that document are also reported to the State.

        Unsurprisingly, the database that the State maintains (that’s the data set you have, Jim) is somewhat ambiguous in a few of the data fields that it records. One of those confused fields is the category called “Total” which the State failed to adequately specify at the onset. The labs in their infancy, were forced to each decide which number on their Certificate of Analysis met the definition of the ambiguous “Total”, and again unsurprisingly, there were a few competing definitions. That’s what your video illustrates.

        When the State finally weighed in, they sent a memo to all the labs in which they declared that available THC was the only thing they were interested in for the purposes of their database.

        I can assure you that Lab N has never been ambiguous about the way it reports data to its customers or to the State traceability, nor have the definitions of words or numbers on Lab N’s Certificates of Analysis ever changed. As soon as Lab N received the memo, Lab N sent an email to all of its customers informing them of the pending change, and then executed that change on the specified day.

        Lab N told its customers then, and I’ll reemphasize it now: the change in the definition by the State affected only the State’s database, it did not affect Lab N’s database or the databases of Lab N’s customers. The only thing that changed was which number on the certificate the State was asking for.

        I’d like to also offer up the observation that while the State is interested only in THC for the purposes of “Total”, consumers (and therefore the P/Ps and Rs) are interested in cannabinoid resin. Most conversations about sales where “potency” is leveraged negotiate on what could be called “Total Measurable Cannabinoid” expressed as a percent by weight. Any cannabis lab worth its salt should be calibrated for THC, CBD, CBG and the acid forms of all three, and that’s more than 95% of all resin in any common marijuana flower. Lab N’s certificate includes all the relevant information including the value “Cannabinoid-Total”, which is what most people care about if they want a summary of resin content.

        The good folks at Lab N are huge fans of transparency. They work hard every day to provide accurate data in a rapidly evolving marketplace, and they’ve walked their profession through huge strides in short time as they continue to further the practice. I bet the same could be said of many of the marijuana labs.

        I commend your efforts, Jim, and I think your heart is in the right place. There seems to be a lot about your data set that you don’t quite understand yet, and it’s led you to miss a lot of the meaning while focusing on shallow features. I think if you were to open your data set up to the public you’d find some informed people out there willing to help you interpret. I like to believe we all benefit from truth in testing.

        Seriously, Jim, if you think your data set has evidence of criminal wrongdoing, then out with it and let’s help the good guys win.

        In the meantime, we at Lab N are going to take your lead and share openly some of the anonymous data that we have. We’ve got a big database that is both detailed and unambiguous. If there’s something you want to learn we are ready to investigate. Come take a look for yourself: http://confidenceanalytics.com/laboratory-data-transparency/

        – Lab N

    2. No. Inflation isn’t what is happening in this graph. Prior to week 49, how “Total” gets calculated was undefined by the LCB, and they authorized the input of that data by the labs using different calculations. What you’re really seeing here is a comparison of labs utilizing competing and legitimately authorized calculations of the ambiguous “total.” They all snap to the line in week 49, not as a result of coincidence, but directly as a result of the LCB finally providing the chosen official definition and associated rules. What you’re seeing is a common occurrence in industry as labs and their regulators work out what makes the most sense as they study an analyte. A basic method is developed, data is compiled, many ways of looking at and interpreting that data are utilized. As the market and industry develop, the regulators help suss out the standardized system. This graph illustrates that process. The more pertinent information gleaned from this graph is how those methods were applied in the context of consistency. A lab that posts with any formula prior to definition standards are fine by me so long as that formula and its resulting data were posted consistently. I worry more about labs who bounce about the graph. Claiming inflation and its connotations of intention is inflammatory at best.

      1. Thank-you, Bobby.

        Inflation, as I have consistently defined it, is definitely on display in the bubble chart movie.
        Anything above the red line is inflated relative to the explicit comparator (and current reporting standard) of THCmax. Inflation, in this context, is not a value-judgement. It is a fact.

        I will repeat (hopefully for the last time) that I am making no attributions whatsoever regarding intention.
        I have no insight into specific motivations (if any) and/or whether the patterns I am seeing in the data reflect any specific underlying cause (or, for that matter, whether any specific cause even exists).

        When I use the terms “inflation” and “inflator”, I have been referring to the reporting of TOTAL values that are higher than the calculated level of THCmax for the sample in question.

        Some labs appear to have been reporting (for example) “total THC” in the TOTAL column (which inflates the number by a factor of 0.14 times the sample’s level of THCA). Some have been reporting other things (believe me, I’ve spent much more time looking into this than a normal human being should —- perhaps I am not normal?).

        There is, however, one lab that stands out from the pack in that I cannot for the life of me figure out how the 3 tested Cannabinoids (CBD, THC, and that old sour-puss THCA) add up, multiply, or otherwise arithmetically factor into what was being reported as TOTAL.

        The questionnaire that I WILL distribute to all labs that are on my e-mail list will include this as one of the 4-5 questions I expect to ask (I’d REALLY like to know what that one lab was reporting). I believe that my e-mail list covers all labs, but can’t confirm messages are getting through, as I’ve not heard back from all of the labs yet. If any of your peers are not aware of my work, I’d appreciate it if you were to let them know about HI-Blog and my examination of the data that they are producing.

        LAB PEOPLE(S): Please look for the survey as soon as I convince my financial advisor and life-partner that the expense of the survey software is justified.

        One final note … for those who have been looking at the Labbiness Bubble Chart Movie, I’m wondering if anyone else has noticed a tendency for things to be shifting from left to right over time? I’d love it if our resident expert geneticists and horticultural savants (e.g., our very fine Farmers) might chime in on how quickly a “new” industry might be expected to achieve the increases in POTENCY (as indexed by horizontal axis in the bubble chart).

        How many generations of plants and/or fine-tunings of grow process might it be expected to take for the observed increases in “POTENCY” to occur?

        Thanks

        (Author’s note – as this is a comment in reply to a comment, I will not count this against my resolution to be less verbose.

        GO HAWKS! (36 to 6 at the moment, and the Team records are falling like the snowflakes outside)

        Verbose PS: I believe I made a couple of references to THC in previous replies in this thread … I’ve likely mixed up “TOTAL” with “THC” in some of those replies … sorry in advance if that is confusing anyone —- it was just me confabulating what is now required in the TOTAL field (the decarboxylation-adjusted “true” Total value for THC) with what, in the “market’s” mind seems to equate to “POTENCY” and/or “TOTAL”.

        1. Jim,

          I think it boils down to consumer demand. It’s the American way, bigger, better and faster! Marijuana is no different and THC is the only relative metric for Americans to access this mentality. I know as a p/p all of our decisions boil down to consumer demand, (and those consumers can’t get enough of the higher THC strains), so much so that we stopped producing strains because of the THC and increased production for the highest testers. I know our company can’t be the only one reacting to this consumer demand, I know I’ve noticed other growers limiting thier production to higher THC strains. This to me would explain the rapid shift to the right, more production thus more testing of higher THC strains.

          1. Thank-you, Mark.

            I believe the industry would benefit, as would it’s consumers, patients, and (ultimately, society) if effective channels of education (and research) were available to address issues such as these.

            That is one of the reasons that I have put so much effort into supporting the formation of an Agricultural Commodity Commission for Cannabis in Washington. R&D and education are central to such a Commission’s role.

        2. inflation

          1: an act of inflating : a state of being inflated

          The problem with your chosen term is that it implies a discrepancy in a SINGLE value (Total Active THC). Unfortunately, no one consistent value has been reported in the “Total” field for testing results — At least not prior to the official rule change which declared one.

          For those months where “Total” was reported as Total Cannabinoid by some labs, you have calculated the difference between Total Cannabinoid and Total Active THC and called this inflation. That is disingenuous. That implies falseness where there is none. Total Cannabinoid is one valid metric for assessing marijuana, while Total Active THC is another. One does not inflate the other. They exist side by side as two separate properties of the material.

          You should also know that at least a couple labs in the state calibrate for and report up to eight different cannabinoids. Those would all be included in Cannabinoid Total.

          “I will repeat (hopefully for the last time) that I am making no attributions whatsoever regarding intention.”

          Difficult to accept considering the following:

          * Your use of the words “FALSE THC” imply an attempt by labs to report different THC levels than are present in a sample.

          * Your suggestion that “inflators” might be shut down or otherwise sanctioned indicates your belief that these labs are not in compliance.

          * You state that so-called inflation is a reflection on some of the labs, when all it truthfully reflects is a change in reporting rules.

          *Your response to Wildman’s concern that he may be using an “inflated” lab: “No-one can control what they did yesterday. We can all, however, influence what we do tomorrow.”

          You have clearly attempted to correlate “inflation”
          as you’ve defined it with falseness and wrongdoing on the part of labs. That is either dishonest or misguided.

          1. Adah:

            Thank-you very much for pointing out that things other than those required to be reported by the LCB are being tested for and that, at least for a time, they may have been included in the TOTAL value. That is helpful to know.

            This fact could well explain why I could not figure out what one lab was doing to derive the TOTAL values they were reporting. Thank-you for helping to clear that up for me (I was feeling temporarily inadequate, as I’m usually very effective at reverse-engineering functional relationships between a constrained set of inputs and a deterministic output).

            Is it at all possible that anything other than Cannabinoids could have been included in the TOTAL as well? Brix, perhaps?

            You are entitled to your opinion, and I’m not about to censor it – but I disagree with most of what you conclude and/or imply. I do, however agree with your comment about my using the term “False THC”. I should have used the term “False TOTAL” (“False” in reference to THCmax being “True”).

            Inflation =to think or say something is larger than it really is
            THCmax = the reference regarding “what it really is”

            Even though the LCB chose (I gather) to not explicitly and unambiguously define to our certified expert labs what PRECISELY was to be reported as TOTAL, I sincerely believe there is information hidden in the variability of what the different labs (groups of labs in some cases) chose to report when reporting that ambiguous value. I’m guessing that some of them took to heart the importance of adequately compensating for the different degrees of decarboxylation that occur using different testing methodologies. Maybe they read the AHP monograph. I’m guessing some did not – at least closely.

            Given what is being reported on the standard minimally compliant label in the regulated market, most people equate THC and/or TOTAL with potency. I believe most of us likely believe that is a poor definition, but it does seem to be what many of the customers of this market are using as a proxy for potency.

            Given that some labs use methods that totally decarboxylate (hence INFLATING reported THC values vs labs using methods that do not so completely decarboxylate), I will attempt to call this DECARBOXYLATION-INDUCED INFLATION in the future. I wonder (I’m being disciplined and am not going to go back and check) if any of the labs that were producing higher reported THC values through DECARBOXYLATION-INDUCED INFLATION were, at the same time, adding in up to 8 different cannabinoids (and, perhaps, brix) into the TOTAL value — which would then, ironically, ALSO be inflated with respect to THCmax.

            Seems kinda win/win for high and inflated values if the two were to come together. (Win/win re: “Friendliness”)
            .
            It would be best if you made reference to the comments I was replying to when you pull the words I used replying to those comments out and present them in a vacuum, as soundbites surrounded by your opinions. Opinions to which you, of course, are entitled. I’m a fan of free (and balanced) speech.
            .
            The context in which words are generated is important … just as is a standard reference for the reporting of TOTAL — such as THCmax.

            While I make a concerted effort not to be baited, I will say with some confidence that I do not believe that I have been dishonest in any of this work. I am, however, aware that I may well be misguided in all that I do and that if that is the case I may well not be aware of being misguided. If it helps, please know that I aspire to not being so. You may all feel free to Judge my success in those efforts. I appreciate such feedback. It can only make me less misguided.

            Has anyone else noticed that the big brown lab CONTINUED TO INFLATE FOR ABOUT 5 WEEKS to the tune of over-reporting TOTAL by about 3.5% after almost everyone else THAT WAS INFLATING had come in line with LCB guidance to report THCmax?

            I know that this may be misguided on my part, but It almost seems as if they were having difficulty accepting that the old way of reporting was dead (or supposed to be dead, at least). Perhaps their processes were such that changing to the newly unambiguous requirement to report THCmax was difficult or otherwise impractical. It’s not like there appears to have been any cost for this non-compliance (not that I’ve been looking for any).

            Was there? You seem to know a good deal about this stuff.

            In looking over the comments that this post/series has received (thank-you all for participating), a few have come across in a way that evokes images (at least to my potentially misguided self) similar to the process of mourning. I’m almost wondering if some of the commenters are mourning the death of the old way of reporting.

            I’ve seen DENIAL.
            I’ve seen (just now) a touch of ANGER.
            I’ve seen a bit of BARGAINING.
            I’m glad that I have not yet seen signs of DEPRESSION (and hope that I do not).
            I’m expecting to see signs of ACCEPTANCE (eventually … maybe when I look at the updated data later this month the signs will be there).

            I hope that I see acceptance. Acceptance is a good thing.

            I welcome further input from all.

        3. From my farming prospective. The guys with big money, outdoor grows, under experienced growers etc. this group will take 1-2years to dial down a method, find their market, get good at what they do. Of course that took me much longer with out the Internet info that is currently available and the legal movement in a few states. Established experienced ussually indoor growers are killing it. Yields are increasing with potency in a factors. Mostly in terpenoids, and other canabinoids than the top 3. Yields are starting to average 3-4lbs a light. Really expect to get better as horticultural growers meet cannabis growers and technology is shared.

          1. Thanks, Dutch.

            Folks … listen to this guy. He virtually swept the awards for solventless extracts at the first Terpestival … using product that he (and his little fishies) grew.

            I’ve had the pleasure …. and a pleasure it was.

  4. were it not for the tracers, i would have a high level of “whisky tango foxtrot” as it is i did anyway. but, then i watched it again & it started coming alive for me. they do bounce right to it @ wk 49. it’ll be interesting to see who the straight shooters are. honesty is an important quality to me ~and PRAGUEness

  5. I have heard that a certain lab “guarantees’ they will provide the highest THC levels. I assume that is Lab N and Lab L is not too far behind. Watching the chart it looks as if Lab N was pulled from the stratosphere like a dog on a chain back to the base line.
    I am having a tough time determining the blue ball – is that B,G,J or M? Also getting a good share of the market dollars but staying much closer to base line (even dipping a bit mid-way).
    I’m going to guess around week 54 the max came into being.
    I’ll bring the herb to the discussion!

  6. Ditto on 49. More interesting will be to slow it down so I can more carefully track the labs. Maybe add lab identifier to the bubbles?

    1. Steve … I’ll do that. You are not the only one having difficulty discriminating shades of color (particularly with some of the smaller bubbles). Expect to see an updated video later today/tonight (the SeaHawks turning Cardinal flesh into turds is taking priority right now).

      Thank-you for your suggestion. (and vote on “magic week”)

  7. I absolutely love your blog and find almost all of your post’s to
    be just what I’m looking for. Does one offer guest writers to write
    content for you? I wouldn’t mind composing a post or elaborating
    on a lot of the subjects you write about here. Again, awesome blog!

    1. Thank-you for sharing your kind thoughts.
      — for now, just use the comments (avid readers can handle long stuff — or they’d be gone already).
      However, many people I respect that I know have my back have advised me (thanks, Gals) to summarize my thoughts more effectively and to distill key points to “bullets” or efficiently-articulated soundbites.

      Use the comments now … keep them shortish, and if it looks interesting, I’m open to collaboration.

      Thanks …. Go Hawks .. it’s only a 2 score game … GIVE THE BEAST A CHANCE TO FEED. It is alleged that the BEAST loves Panther flesh … that is why it now hunts in Seattle.

  8. I just wanted to let everyone know that Wildman (Aaron) was the first to correctly guess that week 49 was the week in which the WSLCB sent direction to the Labs to begin reporting THCmax when reporting “TOTAL” in the Cannabinoid (Potency) test.

    The specific date of the letter to the labs was Monday, May 11, ,2015 … the very beginning of Week 49.

    The prize will likely be awarded at the event in Monroe in mid March.

    Jim

Leave a Reply

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