Probability And Accuracy Of Reports

[Guest JiggyPotamus]

I came across a page on the bigfootencounters website called "Bigfoot From a Physics Point of View," and near the bottom of that page the author calculates the following probability: based on the assumption that 99% of bigfoot reports are false, either being hoaxes or misidentifications, we can take a sample of 100 of the best reports and figure out the probability that they are all false by raising 99%, or .99, to the 100th power. This yields a result of .366, meaning there is a 36.6% chance that all of these 100 reports are describing something other than bigfoot. This also means that there is a 63.4% chance that at least 1 of those 100 reports is truly describing a bigfoot. Or that the witness saw what they claimed to see. Allowing for 99% of reports to be wrong about what they supposedly saw is quite generous in my opinion. I wanted to see what would happen if we plugged in some other numbers using this same probability formula.

 

I know the BFRO database has over 4,000 reports, but I will use that number even though there are probably many thousands of other reports that have been filed by another organization, or are in someone's private research database. Not to mention the sightings that have gone unreported over the decades. 4,000 will suffice for this example since we know there are at least that many reports.

 

So again assuming that 99% of reports are wrong about seeing a bigfoot, for whatever reason, the chances of all of these reports being wrong are about 3.4 x 10 ^ -16. That's a crazy large number, which would mean that there is well over a 99% chance of at least one report being accurate. So the chances of all reports being wrong is small.

 

So let's make it a little more difficult. What if instead of saying that 99% of reports are wrong, we say that 99.9% of reports are wrong? Well in that case there is a 1.8% chance that all reports are incorrect in what they describe. Meaning there is a 98.2% chance that at least one report actually describes a real bigfoot.

 

So making the number of incorrect reports even larger, let's say that 99.99% of people out of these 4,000 are wrong about what they think they saw. Even though some cases have multiple witnesses. So using this larger percentage we see that there is still only a 33% chance that ALL reports are incorrect, meaning there is still a 67% chance that at least one report truly describes a bigfoot.

 

And remember that the more reports we have, the smaller the chances are that all witnesses are describing something other than a bigfoot. Say there are 10,000 sightings, which is more than feasible, considering the BFRO has about 4500 or so, and I know of other databases that were collected over the years. So we would maybe be doubling the known reports to account for unreported sightings, which is being quite stingy in my opinion, as there are probably many more sightings than this overall, reported or unreported.

 

It would not be worth it to even calculate based on 99% of reports being wrong, because again the chances will be astronomically low. So we will start with 99.9% being wrong, again with 10,000 sightings as our number. The result is about 45/10000, meaning the chances of all reports being incorrect is about 4.5 x 10 ^ -3. I think that means there is about a 99.55% chance that at least one report is correct.

 

So let's assume that 99.99% of people are wrong about seeing a bigfoot. In that case there would be about a 36.7% chance that all reports are wrong, or that 63.3% chance that at least one report is correct. Again, this is assuming that 99.99% of reports really are not describing a bigfoot. If we go even further and assume that 99.999% of reports are inaccurate about bigfoot, THEN there is a 90% chance that all of these reports are wrong about bigfoot. That still would leave a 10% chance of 1 sighting being accurate about seeing a bigfoot. But to assume that even 99% of people are misidentifying or hoaxing is preposterous.

 

My question to the community is whether anyone knows of a more in-depth method that could be used to calculate this type of thing? Or maybe we could come up with some differential equations to model something about sasquatch. Anything really. Accurate data is hard to come by though, which makes it extremely difficult. I have seen the differential equations used to model population dynamics, but it is quite hard to do with bigfoot without making a lot of assumptions. But I suppose it is possible to use a wide data set and then compare the results, and we maybe could glean some type of information in the long run. But aside from that, what about just predicting the probability of sightings being correct or incorrect? Any more detailed way to go about something like this, possibly taking into account more variables?

[Bonehead74]

Of course, for your calculations to mean anything, one must presuppose that bigfoot exist. If bigfoot do not exist, then it makes no sense to calculate the chances of a reported sighting not being a mis-id or hoax. The probability would be zero.

In other words, the existence of bigfoot does not (and can not) depend on any sort of statistical analysis of any number of anecdotal sighting reports. Instead, the ability to determine the potential validity of said reports using your (or similar) calculations is wholly contingent upon the actual existence of at least one bigfoot creature. The assumption, a priori, that bigfoot exist, is required for any such theorizing to have merit.

[Guest DWA]

Not only this, but a strictly non-statistical 'gut' evaluation, based on actually reading the reports, is necessary.

 

The presumption of 99% false positive is not really based upon anything but a similar 'gut' evaluation, based on zero actual assessment of the reports.  Having read most all of them, I for one can tell you that almost without exception the reports fall into one of three categories:

 

1) Hospitalizable mental malfunction;

2) Big fat total deliberate lie; or

3) What the witness says the witness saw.

 

Given that so many witnesses were driving or handling heavy weapons, 1) can be ruled out as a reasonable alternative for almost all of them.  Given the consequences of being known or suspected to have filed a bigfoot report, so can 2).

 

That's a far better assessment than the one fueling the 99%.

Edited August 27, 2014 by DWA

[Guest DWA]

And remember that the more reports we have, the smaller the chances are that all witnesses are describing something other than a bigfoot. Say there are 10,000 sightings, which is more than feasible, considering the BFRO has about 4500 or so, and I know of other databases that were collected over the years. So we would maybe be doubling the known reports to account for unreported sightings, which is being quite stingy in my opinion, as there are probably many more sightings than this overall, reported or unreported.

 

I think a two-for-one is extremely stingy.  Ten-to-one is, to me, extremely conservative.   I have never figured out what to think of the mindset that believes that the actual reports encompass all possible encounters, authentic or not.  Most of us know that given any phenomenon, a substantial number of people experiencing it wont file a report, and logic dictates that this will be a significant multiple of reports filed.

[hiflier]

Hello DWA,

 

I certainly agree. And it's not just with Bigfoot either as you can imagine. Proof of existence is indeed needed for figures to have meaning and not appear subtly disjointed because of an undercurrent involving a subject that so far for nearly everyone has been so elusive. There certainly are enough folks here and around to PROVE sasquatch. That, IMO, should take center stage. Until then? Meh.  

[Guest Divergent1]

I think it's feasible to develop a tool with weighted characteristics that could give you a percentage regarding the reliability or validity of a report. This could give a group the means to decide where to focus field research. Relying on gut instinct alone is always a 50/50 chance. 

[BobbyO]

Jiggy, you've seen one haven't you ?

If so, you will then know that the probabilities and numbers then extend in that this animal does exist so therefore people are seeing them and when people see them they, sometimes, report them to Sasquatch research organisations who then, sometimes, put them out there in the public domain.

When you know that these animals exist after seeing one, the law of probability changes where this subject is concerned as you take what bonehead said and throw it out of the window after turning it full circle and understand that these report databases highly likely consist of a hell of a lot more legitimate reports than illegitimate ones, especially the actual visual sightings.

There is no way in the world I'd spend the amount of time I do adding reports to the SSR and analysing x and y if I hadn't seen one and the worst thing about it all ?

I have never ever talked about my sighting to virtually anyone and certainly no research group with it added to a public database so my one is nothing but a memory to me and me only.

That's wrong of me I know, but it's just what it is and I can't imagine it will change any time soon,

I might add it to the SSR one day though..

[Doc Holliday]

I have never ever talked about my sighting to virtually anyone and certainly no research group with it added to a public database so my one is nothing but a memory to me and me only.

 

 

good idea, and probably wise to keep it that way....... little if any good can come from it ,imo, especially outside these walls.

[BobbyO]

I think I'm wrong for not reporting it to be honest Doc as if everyone were like me, there'd be no reports anywhere to analyse etc like I'm doing myself which when you think about it, is ridiculous.

That doesn't mean I'll be reporting it anytime soon though however, rightly or wrongly.

[Guest Rex]

BobbyO, you are far from alone.. as I never reported any of the run ins I had either to any "database" holders.. Nor have any of the other folks I know of who have seen them.

 

the actual reports are but a fraction of the sightings, fact.

[Guest Divergent1]

I wonder what other things never get reported?

[Trogluddite]

I think I'm wrong for not reporting it to be honest Doc as if everyone were like me, there'd be no reports anywhere to analyse etc like I'm doing myself which when you think about it, is ridiculous.

That doesn't mean I'll be reporting it anytime soon though however, rightly or wrongly.

If it was in Pennsylvania, New York, or West Virginia, let me know....    ;-) 

[MIB]

I wonder what other things never get reported?

 

Not sure what you're suggesting. 

 

Do you mean only other critters, or other events as well?  

 

For instance, only a percentage of rape victims come forward because they'll have to face their attackers in court, have to deal with insensitive cops after a traumatic experience, and have a social stigma attached that frankly victimizes them yet again.   So, is that what you're asking about?

 

MIB

[WSA]

First and foremost......again.....i was told there would be no math. But o.k. let us acknowlege that, whatever they say, the highest and best use of sighting reports are only a springboard for follow up. If there is not that, then they are indeed just stories. They are a standing question that those with blood in their veins should always pursue if they want to qualify as serious about this subject: What is this?

[Guest DWA]

Exactly.

 

I keep seeing the 'skeptics' trying to have their cake; eat it too; and a free cake factory.  They say "you can't prove all these witnesses wrong!"  Then they say "you can't prove any of these witnesses right!" Which is a wonderful excuse for sitting on one's hands and learning nothing new.  Do y'all really like that sensation when your hands go numb that much?  Other stuff goes numb too, ya know, when that's the approach.

 

The BFRO and NAWAC and John Green got those databases going for a reason:  scientists catalog and classify and look for commonalities, and databases (contrary to the true belief of 'skeptics') turn stories into data.

 

It's like you said on another thread:  science is hard and expensive, and for that reason goes where the fruit is low and the picking easy.  Scientists know that the way to go about this is not to waterboard witnesses in The Never Ending Search For Truth!!!!  It's to rack and stack, and see whether commonalities justify looking for what is causing this.

 

Because scientists know that the great human tendency...is not to lie.  Or be mistaken about something clearly seen.