Jump to content
SweatyYeti

Patty's Arms And Hands

Recommended Posts

Backdoc

Basically if you take an average you will arrive at a consensus.

 

EX: Let's take Patty and measure areas of her extremities by multiple many different people.  Lets say that arm or leg is measured  in a generic unit of measurement as  7,7,7,8,7,8,7,6,8,8,7,7,8,7,8 then there is a consensus or average that is probably closest to the truth.  As long as someone does not insist the number is 3 or 4 when clearly is it somewhere in the 7-8 area, then the math always works out.

 

That is why when there are Elections there various polls out there. They can't all be 100% right.  Some will show one candidate winning while more will show the other winning, the state polling averages will tell you who will win in nearly all cases pointed to the fact the majority of the polls will tell us who the winning candidate will be in most cases.

 

With Patty, we end up with a general consensus of "The forearm is longer than the upper arm." or vice versa.  (whatever the issue on length and size to be argued).   The exact numbers are not as important as the general average. 

 

Seems there is a good history of using the ARM to THIGH ratios as a basis for comparing people and apes. This is not something someone made up out of thin air.

 

Now those averages can still be plugged in to the Formulas offered by some on the BFF  using the UPPER ARM/ UPPER LEG and they will tell us nearly the same thing as any one exact measurement of the upper arm and upper leg.

 

What the skeptics like to do is take some little issue and say, "well, you measured from this spot to this spot so you are not accurate" The skeptic will try to argue one specific attempt and say "that number you say 18 is really 16 then your conclusions are not accurate!  You measured from this spot on this picture of a blurry shot of Patty!  You measured wrong"

 

 

Backdoc

Edited by Backdoc

Share this post


Link to post
Share on other sites
MikeZimmer

Backdoc

...

"What the skeptics like to do is take some little issue and say, "well, you measured from this spot to this spot so you are not accurate" The skeptic will try to argue one specific attempt and say "that number you say 18 is really 16 then your conclusions are not accurate!  You measured from this spot on this picture of a blurry shot of Patty!  You measured wrong" "

 

 

They do say that I guess. I am more interested in having solidly conducted and documented scientific research, which is what I believe some are doing here. It is possible to make things pretty tight. I would like to be able to say to the real skeptics that a sound procedure has been followed and your objections are without merit, and convince them that the arguments are sound. The other type of skeptic I try to ignore.

 

 

Your suggestion of averaging many people's measurement might be a good one. There would have to be a written protocol to follow, as you suggest. I was hoping that our resident Photogrammetrist, and others acting in that role, can come up with unbiased measures. With crowd sourcing, we might end up getting measurements from a wide range of human subjects as well.

 

 

The assumption here seems to be that the errors will be evenly distributed around the true value, and the mean will give us that. However, if we are all using a flawed protocol, we may end up with a mean which is biased in some unknown direction.

 

These are interesting questions, which I will have to think about.

 

Will consensus yield accuracy? That is a question I have been wrestling with in another context.

 

I did some work on seeing how different size error bars would affect the humorofemoral index, Other things being equal, a plus or minus error of 5% on the measurements still gives a good result. A 20% error is pretty horrible. I suspect, but do not know, that with good technique, it is possible to get a 99% probability confidence interval of plus or minus 5%. To use that error figure, I  subtracted 5% from the measure for the humerus, and added 5% to the measure for the femur. This had the effect of moving the humorofemoral index a bit closer to the human range, which was my intent. The standard deviation still showed that the probability of it being a number from a human was extremely small.

 

I do not know if there is an objective way to ascertain a confidence interval and error values when measuring one subject. Is there a foolproof way to determine where a joint is? Is there a foolproof way to determine if a limb is foreshortened. if there is, the residual error should be extremely small.

 

Where is a good statistician when you need one?

Edited by MikeZimmer

Share this post


Link to post
Share on other sites
MikeZimmer

 I realize people fight to defend their position, rightfully so, but sometimes one only looks weaker arguing something as easily seen and understood as bone lengths. I personally accept the premise that the bones of the subject in the film are decidely out of human range, until someone, or I can see evidence to the contrary, or explains or shows a workable way to simulate the same bone lengths with a person in a suit.

 

Some people are interested in solving a problem, some people are interested in truth, and others ...? In any case, this could easily be about as hard as hard science can be, but not all people get it yet. It is a lot closer to hard science than a lot of the physical anthropology I have read (though I do love that field).

Share this post


Link to post
Share on other sites
SweatyYeti

SweatyYeti, you have computed a humorofemoral index of 91.8. The human humofemoral index lies in the range of 69.8 to 72.8 according to one source (referenced in a previous post). The standard deviation for humofemoral index for humans lies in the range of 1.79 to 3.41, depending on which study you read. All of these numbers were based on small sample sizes, which is really too bad. There may be better numbers out there; the trick is to find them.

 

The smallest difference from the list of possibilities is 91.8 - 72.8 = 19.0, based on a sample of ten Caucasian males. The corresponding standard deviation was 1.62. This difference represents 19.0 / 1.62 = 11.7 standard deviations. Using standard probabilities for standard deviations, the likelihood of this being human is vanishingly small.

 

 

Nice work with the numbers, Mike. :) That 91.8 figure sure does appear to be outside the range of human.

 

And, even if it wasn't completely outside of it...the odds would be very slim that Roger just happened to have employed someone who was a 1-in-a-million freak. 

 

 

 

There are other factors that need to be considered to make it tight - are the measurements correct? There are a number of sources of error in any measurement - did you successfully control for error? Some things that come to mind are reliable specification of the endpoints, allowing for foreshortening, image quality, unconscious bias, .... . Of course, the critics might attack you on any of these points.

 

 

I haven't gotten into a formal 'error analysis' yet....but I have considered the range of possible error involved in making these measurements. 

 

For example....in the measurement of Patty's humerus...I took a very conservative point for the upper-end of it...just above the armpit fold, highlighted by the blue arrow...

 

Bob-PattyHumerofemuralIndex1_zpsa948b1b7

 

 

The upper-end of her humerus may be even higher than where I've placed it...but it certainly isn't lower. One way to tell it can't be lower, is by comparing the vertical distance between the shoulder-joint and the chin. 

 

That distance appears longer on Patty than it does on Bob. So, how could the 'shoulder-joint/upper-end of the humerus' actually be any lower than where I have it placed? 

 

 

 

I had an uncle who may have finished high school, I am not sure. He did field work, wrote and had papers published in Archaeology, and one is still used as a university text apparently - a citizen scientist to be sure.

 

 

 

That's pretty cool. What Papers did he have published?? 

 

 

 

 

I don't measure stuff from images - don't know how - but am interested in seeing this work moving forth with "citizen scientists" at the helm, especially since for a mainstream scientist to embrace this research is a career limiting move.

 

 

 

I'll do whatever I can, along that line...    :popcorn:

 

I would like to present this line of analysis to a younger person in the field of Anthropology...but it would be tough to get any 'takers'....considering the negative effect it would most likely have on their career. 

 

 

 

 

 

Unfortunately, I don't seem to be able to include tables in these posts, or I would give the numbers in tabular format.

 

 

I hope you can find a way to post those tables, Mike. I can try to help you with that, via PM...if it's just a technical issue. :)

Edited by SweatyYeti

Share this post


Link to post
Share on other sites
DWA

Nice work with the numbers, Mike. :) That 91.8 figure sure does appear to be outside the range of human.

 

And, even if it wasn't completely outside of it...the odds would be very slim that Roger just happened to have employed someone who was a 1-in-a-million freak.

Would have been a LOT more than that. Patty is subtly but distinctly outside the human range on multiple markers. So it would be 1/1,000,000 X 1/1,500,500 X 1/2,000,000 X...Would any bigfoot skeptic like to calculate the size of the net that broke cowboy cast to find that freak? Would have cost him millions. It is almost impossible now, with the global reach of the internet. When the world moved at telephone/letter speed? The film hasn't even begun to pay the postage yet.

Edited by DWA

Share this post


Link to post
Share on other sites
MikeZimmer

Would have been a LOT more than that. Patty is subtly but distinctly outside the human range on multiple markers. So it would be 1/1,000,000 X 1/1,500,500 X 1/2,000,000 X...Would any bigfoot skeptic like to calculate the size of the net that broke cowboy cast to find that freak? Would have cost him millions. It is almost impossible now, with the global reach of the internet. When the world moved at telephone/letter speed? The film hasn't even begun to pay the postage yet.

Yep, one of my proposals was to use multiple ratios and compound probability, as you have done. So, unless someone with severe body deformities was in the suit - hey that must be it!

Edited by MikeZimmer

Share this post


Link to post
Share on other sites
MikeZimmer

 

 

SweatyYeti: That's pretty cool. What Papers did he have published?? 

 

Mike Zimmer: Here is one: http://www.ontarioarchaeology.on.ca/publications/pdf/oa7-1-donaldson.pdf I Googled "Willam S. Donaldson" archaeology ontario

 

SweatyYeti:I hope you can find a way to post those tables, Mike. I can try to help you with that, via PM...if it's just a technical issue. :)

 

Mike Zimmer: The comment system does not seem to accept the HTML tags for displaying tables. They look good when you first paste them in, then the tags seem to be removed after publishing.

Share this post


Link to post
Share on other sites
MikeZimmer

An article I wrote over the last few days (one of two). Polite suggestions for improvement are welcomed.

 

 

Metrics and Investigations into North America’s Cryptic Primate

This is a short article that I wrote to help clarify my thoughts about a film shot by Roger Patterson and Bob Gimlin of Yakima Washington in 1967. The film shows a large figure, apparently a female primate, powerfully striding down the bed of Bluff Creek in Northern California.

This film has been analyzed by numerous people with various credentials, many of them relevant academic credentials. My reading of the results of this analysis are that the creature is a real primate. It is called by many names, each native dialect had a different one. Now in modern time, in Sitchamalth, I prefer Sasquatch.  It undoubtedly had a different name in the local dialect, but I will have to research that, since I do not know it.

You can see a good motion stabilized copy of the video at: https://www.youtube.com/watch?v=MICxS6kEzUA

 

Metrics Work Summary

What investigators have accomplished in the study of Patty Images:

  • They have been able to make relative measurements of Patty’s frame and show that various derived metrics are unlikely to belong to any human.
  • They have been able to show that it is physically impossible to create a suit with these proportions, fit in a human, and have the human joints bend at a place consistent with the measured proportions.
  • The obvious conclusion is that there is no possibility that Patty is a human in a suit. This clearly implies that Patty is a real creature, almost certainly a bi-pedal primate.

This conclusion is supported by numerous other lines of evidence, much of it independent of the film of Patty.

 

The General Approach to Patty Metrics

 

Continues at http://viewfromsitchamalth.wordpress.com/2014/12/22/metrics-and-investigations-into-north-americas-cryptic-primate/

Edited by MikeZimmer

Share this post


Link to post
Share on other sites
MikeZimmer

Measurement, Metrics and the Patterson-Gimlin Film – Anticipated Sceptical Objections


A Technical Note


 


There are several initiatives underway to use measurement and ratios to establish the authenticity of the figure in the film taken by Roger Patterson and Bob Gimlin at Bluff Creek California, as a real creature, and not a human in a suit. I think that it is important that the research be conducted in such a way that it is unimpeachable in its rigour. To this end, I have tried to establish key areas where a legitimate sceptic could object to the findings of the effort, and show how it is necessary that these objections be met in the very design of the initiative. I have listed my thoughts on the issue in this document.


 


I did not want to write a book, so unless you have already some familiarity with the topic, and with statistics, it will not be useful to you.


 


I do not wish to patronize or insult anyone doing such research, for they are surely ahead of me in this effort. I have never tried to measure an image, and would not be sure where to start. If this document reminds them of things they may have forgotten, or brings out points that they may not have considered, I have done something positive. If it encourages others to embark on this line of research, I have also done something positive.


 


Below, I list some of the key considerations, in my view, that a researcher must be prepared to address. Others may have a different list.


 


more at: http://viewfromsitchamalth.wordpress.com/2014/12/22/measurement-metrics-and-the-patterson-gimlin-film-anticipated-sceptical-objections/


Share this post


Link to post
Share on other sites
Trogluddite

...  for they are surely ahead of me in this effort. I have never tried to measure an image, and would not be sure where to start.....

 

It's simple - get a piece of paper.  Hold it up to your monitor.  Mark the two points you wish to measure ....

 

OH WAIT, WRONG THREAD!!

 

(MZ - this is an attempt at humor and not at your expense. Not sure how much you follow some of the other soap operas, so I felt obliged to point that out.) 

Share this post


Link to post
Share on other sites
MikeZimmer

It's simple - get a piece of paper.  Hold it up to your monitor.  Mark the two points you wish to measure ....

 

Dang, why didn't I think  of that! :maninlove:  Journal submission tomorrow, wanna collaborate?

 

Share this post


Link to post
Share on other sites
MikeZimmer

From  https://viewfromsitchamalth.wordpress.com/2014/12/23/fitting-a-humerus-bone-to-pattys-upper-arm/

 

I may be all wet on this, but this is the way I have put things together so far.

 

Mike Zimmer

 

 

Fitting a Humerus Bone to Patty's Upper Arm

 

 

Is Patty is a real creature, or a man in a suit? Efforts are underway by members of the Bigfoot Forum to establish the answer. These folks are using metrics to establish their case. In this effort, we need to be able to measure the lengths of bones, various skeletal members, in order to produce metrics. To do this, we need to be able to determine where the endpoints of the bones are. We do not have the information necessary to measure true dimensions, but we can compare various bones in terms of relative lengths. Using this approach, we can compute various useful standard ratios.

 

The humerus, the upper arm bone, is one of the key bones to be investigated. We need to be able to fit a human humerus onto the upper arm of the Patty image, Since we are using an assumption that Patty is a man in a suit and testing this premise, we do want to fit a human humerus onto the image in the right place. This means with the correct end-points, scaled correctly, rotated correctly and oriented correctly.

 

This should be a simple problem, and if we were asked to do it on an image of a human that was of high resolution, we almost certainly would have little contention about the placement. it would not be a political issue.

 

If we had multiple trials, with one or many raters, we would get some variability in the placement and the numbers. We could graph it on X and Y coordinates, we could compute statistics, and undoubtedly do other interesting things. We could crowd source the effort even, if we could find a willing crowd.

 

We can reduce this effort to two simpler problems:

 

1 – locate the correct position for the shoulder joint

 

2 – locate the correct position for the elbow joint

 

 

 

Once this is done, measurement is trivial.

 

 

 

There are complicating factors of course.

 

The bones are covered in soft-tissue, so there is some imprecision due to an inability to actually see the bones. This goes without saying, but it is the complicating factor par excellence.

 

If the image is in fact a person in a suit, then the suit itself will distort the image, as will any air gap between person and suit, if such exists.

 

There is another claim that the suit had shoulder pads. The shoulder pads would further distort our expectation of where the human bones should lie.

 

There are some broader considerations.

 

1 – any measurement should be applicable to an image of a human, as much as to the Patty image. So, methods could be tested against various pictures of people, and the techniques refined and calibrated for accuracy.

 

2 – there is always variability in numbers from measurements, so assigning endpoints will be subject to variability as well. There is a definitional problem first of all – what point on the bone is going to be our reference point? Then, we will have variability in placing it, generally with repeated measurements by multiple people, but even by one person there will be variability over different attempts. When we can't see the bone end, but are trying to estimate its location, there will be more variability. This can be described statistically. If we are dealing with a plane, as in an image, we have variability in two dimensions, call them X and Y. Means and standard deviations would be the typical statistics, but these must be extended to both the X and the Y axis The measurements can be visualized as an X-Y scatter-plot.

 

3 – we may be able to crowd source the measurements, that is, recruit a number of volunteers to make measurements that we can pool.

 

4 – any placement of a joint needs to conform to the bending at the joints observed in the Patterson-Gimlin film. This actually constrains the placement of the points in a good way.

 

5 – We should not restrict ourselves to only one technique. We will find that by employing various techniques, will be able to either converge on one solution, or at least find out why different approaches give different solutions. Once that determination is made, it should be possible to modify the approaches to yield convergence on one result.

 

Approaches

 

Taking all of this into account, and looking at some other work which has been done with metrics, I think that there may be the following approaches:

  • Intuitive

  • Explicit Reference Points

  • Rotational

 

Others may come up with more.

 

Discussion

 

Intuitive

 

The intuitive approach would consist of people trying to estimate the placement with naive pattern recognition, and no explicit protocol. I would expect that with a human figure, you would get a reasonable average with a lot of people contributing results (see crowd sourcing above), but that the variance would also be large, and the corresponding standard deviation large as a consequence. It is interesting to speculate as well what the results would be with a padded suit.

 

 

Explicit Reference Points

 

With explicit reference points, there would be an explicit attempt to make use of other elements of the figure to serve as goods for better placement. At least two members of the Bigfoot Forum have endorsed such approaches.

 

In the thread on Patty's arms, Gigantofootecus has sketched out a technique based on making measurements starting from the eyes of the figure. As a certified Photogrammetrist, he has a professional understanding of how to proceed in measuring images. I do not yet follow it, but trust that it has merit. See http://bigfootforums.com/index.php/topic/4782-pattys-arms-and-hands/page-2 post 23.

 

In the same thread, SweatyYeti has employed a slightly different set of reference points to pinpoint the shoulder. He discussed these in a reply to one of my posts. I am still digesting his material, but you may review it at: http://bigfootforums.com/index.php/topic/4782-pattys-arms-and-hands/page-59 and post 1164.

 

Rotational

 

In http://bigfootforums.com/index.php/topic/4782-pattys-arms-and-hands/page-2, Post 39 SweatyYeti shows an animated GIF that shows a full arm swing of the left arm or the creature. Writes SweatyYeti:

 

“So, I put together another animated-gif...(using 'Copy 8 Scans'...courtesy of Bill Munns)....that show Patty's right hand, in the Frames around Frame 352... â€

 

Although I have not attempted this, it seems that it might be possible to use the rotation at the shoulder joint apparent in this series of images to pinpoint the centre of rotation. Experimentation would be needed to determine if this was a practical technique. It is possible that it has already done, and I am behind the times.

 

I suggest that one or both of these approaches might work.

 

If we put reference points on the upper arm, bright dots for instance, near the shoulder joint, we could track the motion of the points throughout the swing. Just as a long-time night exposure towards the Northern sky will reveal the moving tracks of the stars and the stationary position of the North Star, the dots might reveal that there was a relatively stationary centre. Anyway, this is a suggestion, and others with more experience with images may deem it impracticable

 

Alternatively, it might be possible to draw a line along the length of the upper arm extending past the shoulder. This could be done for multiple images from an arm swing. When superimposed, they might all cross in a fairly constrained area. This would narrow down the position of the rotational centre. Just where in the arm the line was drawn would change the results. I suggest that some experimentation might reveal an optimal placement. Again, this is a suggestion, and others with more experience might say that it is unworkable.

  • Upvote 1

Share this post


Link to post
Share on other sites
xspider1

Good post, MZ.  I bet this kind of geometry could help pinpoint the location of Patty's Shoulder and Elbow w-o too much problem:

 

Share this post


Link to post
Share on other sites
MikeZimmer

Good post, MZ.  I bet this kind of geometry could help pinpoint the location of Patty's Shoulder and Elbow w-o too much problem:

 

Thanks,

 

I will look at this in the future. Now, bedtime, and tomorrow - get ready for X-mas.

 

Regards

Share this post


Link to post
Share on other sites

×
×
  • Create New...