MUPLUS+   Join Plus+ and get exclusive shows and extensions! Subscribe Today! LEARN MORE  


Advertise here now!


It Was Math That Killed The Conspiracy Theory

Statistically speaking, how many people can keep a conspiracy a secret before one spills the beans? Statistically speaking, how long would it take before that happens? Statistically speaking, can someone explain these statistics? Statistically speaking, of course!

Meet David Robert Grimes, a postdoctoral research associate at the University of Oxford who studies cancer and is a science writer and broadcaster. When callers asked about conspiracy theories involving science and medicine, he decided to find a scientific test to determine their validity. The result is his study published recently in the journal PLOS ONE.

Grimes first created an equation to calculate how long it would take for someone ‘in the know’ on the conspiracy to blow the secret, either intentionally or accidentally. The equation took into account the number of people in on the secret, how long it took for one to reveal it and an estimate on the probability of one person revealing it – 4 in a million. That estimate was based on three real proven conspiracies: the NSA Prism surveillance program, the Tuskegee syphilis experiment and the FBI forensics scandal.

Data on three known conspiracies

Data on three known conspiracies

Dr. Grimes then applied his equation to four alleged conspiracy plots – the manned moon landings were a hoax, climate change is a fake, unsafe vaccinations are being covered up and the cure for cancer exists but it’s being suppressed by drug companies.

Data on four alleged conspiracies

Data on four alleged conspiracies

First he determined the maximum number of people that could be in on the conspiracy before it was revealed: moon landing (411,000), climate change (405,000), vaccinations (736,000), cancer (714,000).

Next, he calculated how long before these things – if they were hoaxes – would be exposed: moon landing (3 years 8 months), climate change (3 years 9 months), vaccinations (3 years 2 months), cancer (3 years 3 months). In other words, if the moon landing really was a hoax, it would have been exposed – voluntarily or otherwise – within 44 months of its occurrence based on how many people had to keep it secret. According to Grimes’ equation, Armstrong and Aldrin’s footprints are up there.

The equation can also be used to figure out how many people could be let in on a conspiracy based on how long it needed to be kept quiet. If you only need to keep the secret for five years, you can tell 2521 people; 10 years – 1,000 people; 100 years – better only tell 125 people.


Not surprisingly – statistically speaking – the less people who know, the better. I wonder how many people know Elvis is still alive?


 TAGS: , , , , , , , , ,

  • Christopher P.

    Basing a model on a mere three known instances makes for an incredibly weak approach. This is especially true when the data for each instance is wildly different. The time to exposure is purported to be six years for both the NSA PRISM project and for the FBI forensics scandal, and yet one has a maximum involved of 30,000 people and the other 500 people. So one had potentially 60 times as many people involved in it as the other, and yet it took the same amount of time for it to get out? And from this we’re supposed to believe there’s a consistent and predictable relationship between the number of people involved and the time it takes for the secret to come out? Utterly laughable.

    Another thing is the numbers used for estimating those in on the conspiracy. He uses peak employment at various organizations to plug into his useless model? What basis is there to believe that such a high number of employees would know enough about a specific cover-up to effectively blow the whistle on it? Again, absurd. And to whom would the whistle be blown? Governments that silence and prosecute whistleblowers? A mass media controlled by giant corporations in bed with government, reliant on other giant corporations in bed with government for their advertising dollars, many of which would be catastrophically affected if certain alleged conspiracies were revealed? Pure fantasy.

    A statistical parlor game is not a scientific test.

  • Jack of All Trades

    The approach to this exercise seems obviously foolish. Using the cancer conspiracy as an example: taking the peak number of employees from a private firm and assuming that they would all be “in-the-know” is a highly flawed approach. It is not logical to assume that you become aware of this hypothetical conspiracy the day you get hired. If I were to have attempted a study like this, I would use 1% of the total numbers listed in this study (i.e. 700,000 employees, assuming 1% are “in-the-know”, that gives us 7,000 people that need to be kept quiet).

    Conspiracies require a group to compartmentalize information. It is inherently flawed to incorporate all people who are associated with a conspiracy into the study.

  • JudgeBot

    Unfortunately, his math is way off and he ignores some important cohorts. Otherwise the Manhattan Project would have gotten out. I mean, it was FULL of people that were against the actual project, AND they tested actual atomic bombs, above the actual ground, producing actual goddam atomic explosions.

    Compartmentalization of information: Let’s say the most bat-shit crazy 911 conspiracy were true, just for the sake of argument. It doesn’t require that the individuals running drills on that day are involved. It doesn’t require the passengers, it doesn’t require the flight crews, really, just your secret police which have already done things that require them to hold their silence. In fact, it doesn’t even require those secret police types to be alive tomorrow.

    When I worked for Sunstrand, an electronics manufacturer in Arizona, we had military contracts that were “kitted out” to different teams. While I may have worked on a death ray, or a time machine, the nature of the compartmentalization of the work ensured that nobody would ever know exactly what it was that they were building. You get orders that deal with your little piece of the pie, and never have a full “big picture” view, so you never know to what larger end your labor contributed.

    Concrete documentation: The Whistle Blowers never blow a whistle on things that there isn’t a record of. Who would believe them. There is a structural barrier that would have kept Snowden from enjoying credibility if there was no documentation, or access to documentation. Imagine the sheer luck of having motive and opportunity converge at just the right point, in his case. It almost DIDN’T happen. You can see how his model starts to break down rapidly when you put it in view of “what has happened in the past”. It stands on assumptions that are not a requirement of the phenomena it purports to model.

    Personal culpability: Notice two things, first, the Edward Snodens were never the actual “trigger men” in conspiracies that get exposed. Trigger Men are much less likely to come forward, because they are culpable in the crime, and generally previous crimes. The trigger men, if they do come forward, do so only on their death bed, and structurally (not conveniently) have no access to concrete documentation. Trigger men also enjoy their power psychologically. Cops killing an innocent kid then covering it up are the perfect example. They often get away with their conspiracy even when there is video footage.

    So let’s put this together and see how it might look: The cops on the scene are doing what the plan required of them, but they are entirely unaware of their role in the bigger picture. (Compartmentalization) They may get an order from higher up to do a single thing out of the norm, under some innocent pretext. Social engineers can talk people into all sorts of things using pretext. This whole thing actually works better when the people scheduling drills on the day of an actual attack (as an example) are NOT in on it, because they will point to a schedule they were handed months ago, and nobody will believe it wasn’t a conspiracy more than them. (Concrete documentation – either non-existent, outside of access, or purposefully misdirecting.)

    We spend billions every year in covert intelligence operations, and from a quick accounting, the vast majority never leaks. We KNOW this happens, because he have an idea of the black budget. Let me repeat that. The very existence of our vast black ops military industrial complex, demonstrates that this model doesn’t function.

    So the model is flawed, as it ignores how past events were recorded, and plausible transmission of activity. Transmission techniques that we know intelligence agencies in fact, are known to use. How does his math fare?

    “Nasa had 411,000 employees”,

    sure, but if there was a hoax, his model assumes that it was discussed at some company meeting? Bad mathematical assumption.

    “But if everyone from Novartis, Pfizer, Roche, Sanofi, Merck and Co, and Johnson and Johnson, to GlaxoSmithKline and AstraZeneca were involved it would amount to 714,000 people and the story would get out within three years and three months.”

    This is a fundamental misrepresentation of who has the power to shut a project down, and how much bosses (even in our jobs) share with those below them.

    To be clear, this isn’t an argument FOR conspiracy. This is an indictment of Dr. Grime’s model. There MAY be a model that can dispel conspiracy theory. This is demonstrably not that.

  • Jack of All Trades

    I posted the same point as you regarding compartmentalization. I read the study and he clearly states the same assumption, that not everyone associated with the conspirators would be aware of the conspiracy. I am still disappointed with the study and its conclusions because of his oversight, using such high numbers.

  • Moor Cow Belle

    “…this isn’t an argument FOR conspiracy. This is an indictment of Dr. Grime’s model. There MAY be a model that can dispel conspiracy theory. ”

    No, there will never be such a model. The idea that you can come up with a mathematical model to somehow disprove collusion or theories about possible collusion… it doesn’t even make sense to think this way.
    There will always be conspiracy theories because there will always be conspiracies. It’s as simple as that. Human beings exist, therefore there will be conspiracies, and therefore there will be theories about possible conspiracies.
    What are we really wishing for here? A mathematical proof that the people who control governments, corporations, intelligence agencies, & military forces don’t ever & could never exploit the general public? There’s a pretty good proof to the contrary. It’s called history.