One Incontrovertible, Unequivocal, Undeniable Fact Which Refutes the Diary

[rjpalmer]

My brother is a doctor of physics and retired patents attorney. I'm not sure he could help straighten out all your latest misunderstandings - even if he had the least motivation to try.

Keep digging yourself a deeper hole, RJ. Unlike my sensible brother, I am shallow enough to find your efforts entertaining.

Yes, Caz.

I believe you mentioned in the past that he was a very accomplished chap, that's why I was confident that he could guide you through the obvious logical errors of Ike's statistical analysis.

Of course, I appreciate why you are hesitant to flip over this particular rock--you are worried that you will be bitterly disappointed and shocked by what he tells you.


Understood. But, if I may, let me just give you a small preview of the sort of thing he might say.

You see, when Ike was pontificating last week, and giving a wonderful demonstration of how not to do statistics, I was sent a private message by a prominent member of these forums, who assured me that my concerns were justified.

But, being curious and fair-minded, he/she also decided to run Ike's analysis past Jeff Hamm. This was my correspondent's doing; I did not ask for it, but I was given permission to reprint the commentary if needed, with the caveat that Jeff had no interest in directly contributing to the 'Diary' debate.

Being a patient fellow, I decided not to reprint the comments, hoping that you and Ike would eventually see the error of your ways, and apologize to those who were abused & mocked for merely pointing out that Ike's methods were entirely invalid.

I know you respect Jeff's logic and abilities from your comments on the geographical profiling threads, so perhaps you'll believe it if you hear it from him instead of me...that is...if you can withhold your laughter at my expense long enough to try to understand what he is saying.


Good luck! If you still don't get it...I suggest that you try your brother or someone else you trust, perhaps down at the local university. I'm sure they can help.

Warmest wishes as always,

R P

• "Hi Roger, with best regards from Mr Hamm……
  
  
  “Ok, RJ's concern that the issue is about the probability of already known events is valid. It skews one's idea of what chance is. Ike's calculation of probabilities with coin flips (his 5/10 heads), is, while a correct calculation, isn't the probability one uses for inferential decisions (which this is). Given 5/10 heads is the most common outcome, for example, the fact that it only occurs 1/32 times, doesn't mean squat, it's still more probable than any other outcome, individually.
  
  For coin flips, a typical example in statistics, one would look for things like "What is the probability of getting some outcome at least, or more extreme, than the target?", so what's the probability of getting at least 5, or more, of a kind out of 10 flips? And of course, that would be 100% (5:5 H:T, 4:6 H:T or 6:4 H:T, etc, it has to come out somehow).
  
  Now, with the floor boards and the publisher, that would be a case of what Ike calls a "named sequence". (i.e. what's the probability of getting the exact sequence TTHHHHHTTH? It's really low, but that's what I just got as I typed this. But, it's a 6H 4T sequence, and the probability of getting 6:4 or more extreme, like 4H7T, or 7H3T or 7T3H, etc, is well over 90%, because it's everything other than 5:5 exactly).
  
  To even begin to try and equate them, one has to consider all of the possible events one would consider as "acceptable" in terms of finding the Diary at Battlecrease (i.e. pulling up floor boards; or a house inspection of any sort; a visitor to the house who may have checked the library; a garage sale; etc). Also, how long before the "publisher event" would we accept? What if the floor boards had come up the day before? Or the week before? and so forth.
  
  If one could decide upon all the possible events, and timelines, that would be viewed as acceptable (like we can for coin flips), then one start to approach a suitable calculation. And, given the fact that almost anything could be turned into a story that is acceptable, the probability of there being "an acceptable story by chance" is near 100%. The fact that this particular story might appear to have some low chance of occurring, is just like my above coin flip sequence.
  
  Had I predicted that very sequence was going to occur, then I might be able to claim something, but the fact I flipped it first, and know the outcome, changes what I have to deal with - not only the currently known outcome, but all of the possible outcomes that didn't happen but that I would have deemed acceptable.
  
  Hmmm, but long for a quote. Anyway, you can pass that on to RJ. I'm sure he'll be able to extract from that suitable points to deal with the conversation in the thread appropriately. I have no problem with him mentioning my name, but I also have no hesitancy in making the prediction that Ike will dismiss it and insist on his post-hoc calculation of specific events”
  
  All the best

[Note: perhaps the best place to start would be to ask your brother to explain why a “post-hoc calculation of specific events” is not appropriate. And let me just add a final note: unless you and Ike (and Erobitha) eventually grasp this, I have precisely 37,577 reasons not to engage in this discussion further. RP]

[caz]

Caroline, so Tony knew Eddie?

No idea, Scotty.

Tony was living on Fountains Rd until his death in August 1991, and Eddie was living at his girlfriend's address in Fountains Rd in 1992. Both used the Saddle, but whether their paths ever crossed I couldn't say.

The question is how much of a coincidence was it, for a Fountains Rd resident and Saddle regular to have been chosen for the diary's provenance [because he was dead and couldn't be questioned], if Mike and Eddie - another Fountains Rd resident and Saddle user - had yet to meet, and if Eddie's only connection with either of them was the fact that he had worked in Maybrick's old house in a different part of Liverpool?

Love,

Caz
X

[caz]

Yes, Caz.

I believe you mentioned in the past that he was a very accomplished chap, that's why I was confident that he could guide you through the obvious logical errors of Ike's statistical analysis.

Of course, I appreciate why you are hesitant to flip over this particular rock--you are worried that you will be bitterly disappointed and shocked by what he tells you.


Understood. But, if I may, let me just give you a small preview of the sort of thing he might say.

You see, when Ike was pontificating last week, and giving a wonderful demonstration of how not to do statistics, I was sent a private message by a prominent member of these forums, who assured me that my concerns were justified.

But, being curious and fair-minded, he/she also decided to run Ike's analysis past Jeff Hamm. This was my correspondent's doing; I did not ask for it, but I was given permission to reprint the commentary if needed, with the caveat that Jeff had no interest in directly contributing to the 'Diary' debate.

Being a patient fellow, I decided not to reprint the comments, hoping that you and Ike would eventually see the error of your ways, and apologize to those who were abused & mocked for merely pointing out that Ike's methods were entirely invalid.

I know you respect Jeff's logic and abilities from your comments on the geographical profiling threads, so perhaps you'll believe it if you hear it from him instead of me...that is...if you can withhold your laughter at my expense long enough to try to understand what he is saying.


Good luck! If you still don't get it...I suggest that you try your brother or someone else you trust, perhaps down at the local university. I'm sure they can help.

Warmest wishes as always,

R P

• "Hi Roger, with best regards from Mr Hamm……
  
  
  “Ok, RJ's concern that the issue is about the probability of already known events is valid. It skews one's idea of what chance is. Ike's calculation of probabilities with coin flips (his 5/10 heads), is, while a correct calculation, isn't the probability one uses for inferential decisions (which this is). Given 5/10 heads is the most common outcome, for example, the fact that it only occurs 1/32 times, doesn't mean squat, it's still more probable than any other outcome, individually.
  
  For coin flips, a typical example in statistics, one would look for things like "What is the probability of getting some outcome at least, or more extreme, than the target?", so what's the probability of getting at least 5, or more, of a kind out of 10 flips? And of course, that would be 100% (5:5 H:T, 4:6 H:T or 6:4 H:T, etc, it has to come out somehow).
  
  Now, with the floor boards and the publisher, that would be a case of what Ike calls a "named sequence". (i.e. what's the probability of getting the exact sequence TTHHHHHTTH? It's really low, but that's what I just got as I typed this. But, it's a 6H 4T sequence, and the probability of getting 6:4 or more extreme, like 4H7T, or 7H3T or 7T3H, etc, is well over 90%, because it's everything other than 5:5 exactly).
  
  To even begin to try and equate them, one has to consider all of the possible events one would consider as "acceptable" in terms of finding the Diary at Battlecrease (i.e. pulling up floor boards; or a house inspection of any sort; a visitor to the house who may have checked the library; a garage sale; etc). Also, how long before the "publisher event" would we accept? What if the floor boards had come up the day before? Or the week before? and so forth.
  
  If one could decide upon all the possible events, and timelines, that would be viewed as acceptable (like we can for coin flips), then one start to approach a suitable calculation. And, given the fact that almost anything could be turned into a story that is acceptable, the probability of there being "an acceptable story by chance" is near 100%. The fact that this particular story might appear to have some low chance of occurring, is just like my above coin flip sequence.
  
  Had I predicted that very sequence was going to occur, then I might be able to claim something, but the fact I flipped it first, and know the outcome, changes what I have to deal with - not only the currently known outcome, but all of the possible outcomes that didn't happen but that I would have deemed acceptable.
  
  Hmmm, but long for a quote. Anyway, you can pass that on to RJ. I'm sure he'll be able to extract from that suitable points to deal with the conversation in the thread appropriately. I have no problem with him mentioning my name, but I also have no hesitancy in making the prediction that Ike will dismiss it and insist on his post-hoc calculation of specific events”
  
  All the best

[Note: perhaps the best place to start would be to ask your brother to explain why a “post-hoc calculation of specific events” is not appropriate. And let me just add a final note: unless you and Ike (and Erobitha) eventually grasp this, I have precisely 37,577 reasons not to engage in this discussion further. RP]

I will read your post, RJ, and what Jeff had to say, when I catch up with all the previous posts, but in the meanwhile could I just ask you, for the sake of completeness, to post precisely what information was given to him about the two known events on 9th March 1992: the Maybrick floorboards being lifted and the Maybrick related phone call, and precisely what information was asked for in return, and how the question was phrased? Or was it simply a case or copying and pasting all Ike's posts on the subject, minus the full context, and asking for Jeff to comment?

[caz]

No Ike, I'm not going to give you the answer, and I'll tell you why.

It will leave the impression that this example is relevant, and somehow analogous to your own statistical analysis, and that the difficulty we face is one that can be resolved through simple mathematics.

That would be wrong, because the example you give involves calculating the odds of events that we know happened.

By contrast, you are attempting to calculate the odds of a theoretical event, that you don't know happened, and that is what is leading you into the quagmire...

Regardless of what Jeff might have had to say [I've not read it yet and I'll wait to see what he was given], I thought Ike was indeed calculating the odds of two very specific known events - the floorboards and the phone call [with the known Saddle connection thrown in just for jolly] - happening on the same day by chance alone, and therefore being entirely unconnected. Which theoretical events was he factoring in? I must have missed those.

If you were entirely happy that there was a 1 in 18 chance of Mike deciding to make that unique call about a diary supposedly by Maybrick, on a day when Maybrick's old floorboards needed to come up, what was all that guff about an entirely theoretical event, involving a chin-wag between the two known Saddle users in advance of the work being done, if it wasn't to see if the coincidence might be toned down or made to go away, with a theoretical connection between these known events, which would keep two other theoretical events afloat - Orsam's Awesome Auction and His Eleven Day Creation?

[caz]

This is a little misleading, isn't it?

"The landlord only said..."

Here's Feldman's account below. It is far from clear whether Feldman even asked the landlord about Eddie's visits to the Saddle, and Feldman is not great about giving the details, or sticking with a coherent chronology. So we don't really know what--if anything--the landlord said about Eddie, or what Feldman asked.

Roger Irrelevant strikes again.

My point wasn't about Eddie, or whether the landlord remembered him. It was merely that the landlord only mentioned lunchtime drinks, when remembering visits by Mike, his father and Tony. All I meant was that I don't know what evidence there is for Mike using the pub in the evenings. [Or, for that matter, Eddie using it at lunchtimes.]

[Iconoclast]

[*]"Hi Roger, with best regards from Mr Hamm……

“Ok, RJ's concern that the issue is about the probability of already known events is valid. It skews one's idea of what chance is.

Dear Readers,

Probability theory is used to analyse experimental data. Group A and Group B consist of individuals matched on multiple criteria (to make them as 'equal' as possible). Group A eats seeds and nuts for breakfast for a year. Group B eats toast for breakfast for a year. They are timed at 3 months, 6 months, 9 months, and a year running over one mile. At the end of the year, the researchers get a whole bunch of data. How do they assess that data in order to see if there is evidence of a difference between the two groups due to their breakfast? Well, we use probability theory (statistics) to crunch the data to see if we can establish the likelihood of observing any difference we see in running times between the two groups by purely chance alone. If we get a probability of 1-in-20 or less (p<=0.05) of chance being the reason for any difference observed, we are legitimised in concluding that our hypothesis (that eating seeds and nuts for breakfast will improve running times over toast, for example) has been confirmed. Confirmed, but not certain. There remains a 1-in-20 chance that chance alone caused the difference. So we do the experiment again, and again, and again, and if we keep getting a significant p value our confidence grows that our hypothesis was correct all along.
Why do I say all of that? Well, because experiments retrospectively analyse known data, and this is not only acceptable but inevitable. How would we analyse the two groups before we have any running data to compare the two with?

Therefore, using probability theory on known events is absolutely rock-solid appropriate. Personally, I would say that it was the best use of probability theory because you know your data has happened.

Ike's calculation of probabilities with coin flips (his 5/10 heads), is, while a correct calculation, isn't the probability one uses for inferential decisions (which this is).

As Caz has guessed, and I will now guess, I think it is highly unlikely that Jeff would have suggested that the 1-in-37,557 chance was inferential if he'd actually read all of the posts. It wasn't. End of.

Given 5/10 heads is the most common outcome, for example, the fact that it only occurs 1/32 times, doesn't mean squat, it's still more probable than any other outcome, individually.

I'm fairly confident that neither I nor anyone else has ever suggested otherwise.

For coin flips, a typical example in statistics, one would look for things like "What is the probability of getting some outcome at least, or more extreme, than the target?", so what's the probability of getting at least 5, or more, of a kind out of 10 flips? And of course, that would be 100% (5:5 H:T, 4:6 H:T or 6:4 H:T, etc, it has to come out somehow).

Jeff's off some blind alley here, probably because he hasn't been properly briefed. I don't think any of my readers imagine I am disputing any of this. I'm unable to even guess at what Jeff thinks he's addressing here but it's definitely not any argument I've made recently.

Now, with the floor boards and the publisher, that would be a case of what Ike calls a "named sequence".

No it ******* isn't!!!!!!!!!!!!!!!!!!!!!

Why on earth would anyone make such an idiotic claim? If this was a named sequence, I'd have asked "What was the probability that the floorboards came up first and the call to the literary agent came second?" or something a bit like that. I'm not going to keep explaining this, but the question is not one of inference nor one of sequence. It is a simple (really, really, really simple) case of basic probability theory. Two things that could have happened (but didn't have to happen, ever) either together or separately from May 12, 1889, finally did happen by March 9, 1992 and - lo - they happened on the same day. Number of times it happened (1) over number of times it could have happened (37,557). End of. No need for RJ or anyone else to run around looking for ways of contradicting this Statistics 101 scenario.

(i.e. what's the probability of getting the exact sequence TTHHHHHTTH? It's really low, but that's what I just got as I typed this. But, it's a 6H 4T sequence, and the probability of getting 6:4 or more extreme, like 4H7T, or 7H3T or 7T3H, etc, is well over 90%, because it's everything other than 5:5 exactly).

To even begin to try and equate them, one has to consider all of the possible events one would consider as "acceptable" in terms of finding the Diary at Battlecrease (i.e. pulling up floor boards; or a house inspection of any sort; a visitor to the house who may have checked the library; a garage sale; etc). Also, how long before the "publisher event" would we accept? What if the floor boards had come up the day before? Or the week before? and so forth.

I agree, Jeff! But you should not have been led to believe that that was the issue you were being asked to address! As I have said recently on here, you are welcome to pose any probability question you want. I accept that there are more scenarios which could be factored-in, and if you want to do that in order to increase the numerator and therefore decrease the probability of it happening by chance alone, fill your boots.

But you won't be analysing the simple scenario I have been talking about.

Had I predicted that very sequence was going to occur, then I might be able to claim something, but the fact I flipped it first, and know the outcome, changes what I have to deal with - not only the currently known outcome, but all of the possible outcomes that didn't happen but that I would have deemed acceptable.

Careful, Jeff, you are wandering into RJ-land where things which are known to have happened could only have happened on the day they did regardless of how many times they could have occurred before that. The probability of predicting that sequence is 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5. Funnily enough, looking back after it happened, the probability that anyone had predicted that outcome is also 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5 x 0.5. Knowing that someone did does not elevate its probability from X (less than 1) to 1 (certainty) simply by dint of the fact we now know it actually happened. Same with Maybrick's floorboards and the call to Rupert Crew. Exactly the same.

Hmmm, but long for a quote. Anyway, you can pass that on to RJ. I'm sure he'll be able to extract from that suitable points to deal with the conversation in the thread appropriately. I have no problem with him mentioning my name, but I also have no hesitancy in making the prediction that Ike will dismiss it and insist on his post-hoc calculation of specific events”

The vast majority (to come back to where we started) of probability analysis uses post-hoc calculation. Once I know 37,557 days have passed since May 12, 1889 and those two Maybrick-related things have happened on the same day, that allows me to do a post-hoc calculation. I could have done before March 9, 1992 (the 37,557 would be smaller by every day I did so) or I could do it on or after March 9, 1992. So that's pre-hoc and post-hoc, but either way, it doesn't matter. The statistics are correct.

Ike

[erobitha]

I think Jeff Hamm is a poster who is clearly well-experienced in dealing with statistical data and probability theory. I would be very interested to see if the question posed to him was exactly the scenario Ike outlined above. I have a sneaky feeling Jeff may have been given a slightly twisted version of events. Curious we do not see what exactly was asked. Waters muddied once more.

Welcome to prove me wrong on that R.J.

[rjpalmer]

I think Jeff Hamm is a poster who is clearly well-experienced in dealing with statistical data and probability theory. I would be very interested to see if the question posed to him was exactly the scenario Ike outlined above. I have a sneaky feeling Jeff may have been given a slightly twisted version of events. Curious we do not see what exactly was asked. Waters muddied once more.

Welcome to prove me wrong on that R.J.

Dishonest people frequently accuse others of being dishonest because they assume everyone else is the way they are.

Let me know if you fall into that category, Erobitha.

This is the third or fourth time I've been falsely accused of being dishonest on this thread, which is one of the reasons I know longer wish to engage with the Maybricknicks.

As I explained in my previous post (do you people even read?) it was another poster who asked for an independent opinion--I was not involved.

I was never a part of their discussion, and was only sent the final statement that I reposted above; I made no contribution to any of it, and as far as I know, there was no 'question posed.' He was just asked to read Ike's statistical analysis and determine whether it was valid. He did an admirable job, in my opinion, but clearly Ike is still in denial, as are you.

Anyway, the usual unpleasantness aside, paragraphs 3, 4, and 5, make it self-evident that he FULLY understands the issues at hand (and even refers to 'the floorboards and the publisher') and his critique is direct and to the point.

Let's not pretend that it is merely a 'misunderstanding' that prevents people from seeing the validity of Ike's methods. Ike is engaged in junk statistics, and many people know it. Others appear to be easily led astray by muddled thinking.

[rjpalmer]

I will read your post, RJ, and what Jeff had to say, when I catch up with all the previous posts, but in the meanwhile could I just ask you, for the sake of completeness, to post precisely what information was given to him about the two known events on 9th March 1992: the Maybrick floorboards being lifted and the Maybrick related phone call, and precisely what information was asked for in return, and how the question was phrased? Or was it simply a case or copying and pasting all Ike's posts on the subject, minus the full context, and asking for Jeff to comment?

As I just noted in the previous post, it is self-evident from paragraphs 3, 4, and 5 of his statement that he was acquainted with all the relevant details. Why does it matter if he received his information from reading Ike’s posts? Since you’ve been defending the validity of Ike’s claims, it seems odd that you now seem to be questioning Ike’s ability to have given a concise and coherent reckoning of the relevant issues.

What I can’t comprehend is why you continue to believe that the date of Maybrick’s death, or the date of the last entry in a diary that has not been authenticated—6 May 1889—are relevant to the matter at hand.

Why would they be?

Why is this 37,557-day span relevant? It IS inferential—entirely inferential—based on Ike’s assumption that the diary is the genuine artifact, coupled with his inability to think the matter through from the beginning.

Bear in mind that there is no consensus about where the diary came from, nor how old the diary is.

Some believe that Anne Graham saw the diary in the 1970s—a belief that obviously conflicts with the ‘floorboards’ theory.

David Barrat believes the Diary didn’t even existed on 9 March 1992 (and I tend to agree with him).

Scott Nelson seems to be wondering if Devereux and Lyons might have cooked it up sometime in the late 1980s or early 90s, based on an earlier manuscript.

Other believe it genuinely dates to 1888-89.

Others believe it dates to some other undetermined date in the 20^th Century.

A Liverpool paper claimed that Scotland Yard believed the diary was created between 1982 and 1992 (“in the past ten years.”)

Meanwhile, the earliest confirmed and undisputed day the diary was even laid eyes on happened in mid-April 1992.

So I ask again: why on earth do you believe that the span of years, 1889-1992 (37,577 days) is an appropriate part of any statistical calculation?

Explain your reasoning in a concise manner.

With that in mind, please note the following from Jeff’s paragraphs 3 & 4:

“Now with the floorboards….

“To even begin to try and equate them, one has to consider all of the possible events one would consider as "acceptable" in terms of finding the Diary at Battlecrease (i.e. pulling up floor boards; or a house inspection of any sort; a visitor to the house who may have checked the library; a garage sale; etc). Also, how long before the "publisher event" would we accept? What if the floor boards had come up the day before? Or the week before? and so forth.

Is this not precisely the same thing that I’ve been asking in so many posts?

Here’s what I wrote in #6454.

“As the diary doesn’t say anything about (Maybrick) hiding the Diary at Battlecrease, let alone under the floorboards (in fact, the diary’s last entry seems to point towards the Knowsley Buildings), there are many different ‘events’ that would have qualified as being a “coincidence.”

Can you grasp why he is asking you to consider ‘all possible events’ and why I was asking about ‘many different events’? This is REALLY what we are attempting to calculate—and it is largely unknowable and largely pointless.

If Feldman had learned that a plumber had been in Battlecrease in February 1992, could that not have been enough to tickle his suspicions? Or what if part of the old Liverpool Cotton Exchange had been excavated in January? Would people have wondered about that as a possible provenance?

Similarly, Jeff wonders if a garage sale at Battlecrease would have been enough. Could the diary have been hidden inside an old sofa that had been sold? Or as part of a box of books that the owner didn’t even bother to thumb through before selling?

All of these are merely theoretical—just as the ‘floorboard’ provenance is theoretical. It just happens to be the ‘event’ that was noticed and investigated.

But since we do not know how old the diary is, we are in the unenviable position of trying to calculate the amount of possible events that could have raised suspicions and made people draw ‘inferences.’

Actually, it is an entirely a crazy thing to calculate statistically, which is why Jeff writes… “to even begin to try…”

Again, quoting my Post #6454:

"We don’t know how frequently Dodd had workmen renovating his house. You will scoff at this, but, for the sake of argument, if workmen were at Dodd’s house every day of the week, for an entire year, there would be no ‘coincidence.’ Whatever date Mike first contacted a literary agent that year, it would have coincided with work being done on Dodd’s house. The coincidence would have happened 100% of the time."

I suggest you think that through very carefully.

Work on Dodd’s basement? Renovations at one of Michael Maybrick’s houses? Tearing out a shelf at the chambers of one of Maybrick’s former lawyer’s law offices? The excavation of John Over’s outhouse? An estate sale held at a neighboring house in Aigburth?

As Jeff writes:

If one could decide upon all the possible events, and timelines, that would be viewed as acceptable (like we can for coin flips), then one could start to approach a suitable calculation. And, given the fact that almost anything could be turned into a story that is acceptable, the probability of there being "an acceptable story by chance" is near 100%. The fact that this particular story might appear to have some low chance of occurring, is just like my above coin flip sequence.”

Clearly he knows about the floorboards. He alluded to the floorboards. ‘As for the floorboards…’

As I wrote in Post #6466:

“Until you have actual evidence, what you are really attempting to measure is the validity of Keith Skinner’s (or Paul Feldman’s) suspicions. Or put another way, you are trying to calculate how likely it would have been for a ‘Maybrickian’ event significant enough to have caught someone’s attention to have occurred on the same day that Barrett came forward with the diary.

I see that Ike is now crying foul and is assuming that Jeff doesn’t know the relevant details. He absolutely DOES know the relevant details, and he knows that Ike is looking at the problem from an entirely flawed viewpoint. He even correctly predicted Ike’s reaction.

In attempting to psychoanalyze Ike’s thought process, I’m guessing that the hiccup in his thinking comes from the fact that we actually know that the work on Dodd’s house took place on 9 March 1992, whereas we don’t know about theoretical garage sales, excavations at the Cotton Exchange, carpentry work at Michael Maybrick’s cottage, etc.

This must be the ‘sticking point.’ But Ike has to face the fact that millions of events happen at any given moment, and often it is only our human consciousness that ‘connects’ them—even though no actual connection exists.

In this sense, the floorboards being lifted at Battlecrease are just as airy and theoretical and irrelevant as a garage sale that may or may not have happened—until someone proves otherwise. Phony statistics don’t help; they merely cloud the issue. And we know that other airy and theoretical events also triggered suspicion. When it was discovered that Anne Graham worked on the site of Maybrick’s old office, the former Knowsley Buildings, research was done to see if the diary had come from there. What if a workman had a vague memory of an old box of books having been removed? Wouldn’t that have been ‘viewed as acceptable’?? And what if he had heard about Feldman’s upcoming film project, and tried to cash in by inventing a story that one of his mates had found something in a biscuit tin?

Whatever the case, if one wants to view the two events as a startling coincidence, that is entirely their business. It’s just not acceptable to punch in the date of Maybrick’s death and claim, through flawed logic, that we are faced with 37,577 to 1 odds.

It’s bogus.

Note: The above will have to suffice to any future readers of this thread. I can waste no more time on this tomfoolery (pun intended).


FINIS

[Iconoclast]

I think our readers who have been paying attention will instantly recognise that I have acknowledged time and again that other factors could be added to the mix. Obviously we don’t have any specifics, and many of the examples offered in RJ’s latest post are just theoretical and designed solely to make it sound like opportunities to uncover a Maybrick journal were a daily event back in 1992. My point is, if you add in different factors, that is a different probability analysis. Everyone is welcome to do theirs. I’m happy with mine.

RJ’s post makes a rather startling suggestion: apparently, if workmen were at Dodd’s house every day of the week, for an entire year, there would be no ‘coincidence.’ Whatever date Mike first contacted a literary agent that year, it would have coincided with work being done on Dodd’s house. If you recall, dear readers, this is also RJ’s mythical apples argument - apples fall from a tree somewhere constantly therefore you can’t assume that something happening at the same time as one apple falls was associated with or caused by it falling. Obviously, if workmen were at Maybrick’s house every single day - not simply for this arbitrary year 1992 but every day since he died or wrote his last entry into the scrapbook (May 3, 1889) then obviously someone ringing a literary agent with the diary of Jack the Ripper authored by James Maybrick would be absolutely guaranteed to happen on the same day as work was being done on Maybrick’s house. Honestly, if you’re looking for ‘junk statistics’, this is it. RJ’s post takes two separate events which were previously never on the record, and turns one of them (the lifting of Maybrick’s study floorboards) into an event more regular than breathing. No amount of explaining will change this. RJ wants us to believe that the call to the literary agent was inevitably going to fall on the same day as something else which could be linked to a means of finding Maybrick’s scrapbook, and - if we let him - he’ll just keeep on that theme forever as a way of masking a very startling truth - the very best means of finding his scrapbook was the thing we know finally happened on March 9, 1992, as was the call to the literary agent. And - Lord - that doesn’t even start to factor in the implausibility of a member of the team drinking in the same pub as the guy who brought the scrapbook to the literary agent.

There is too much in RJ’s post to properly address right now. I may or may not address some or all of it again later. Please remember, dear readers, that you can analyse anything. It’s up to the individual to make their case (backed up by probability theory or not). But other cases are not analyses of this case, so let’s be definitive: What is the probability that someone (with a known date) would bring a diary of JtR purported to have been written by James Maybrick to light on the same day that Maybrick’s study floorboards were raised (on the record)? It’s 1-in-37,557. That specific scenario is effectively impossible (and we haven’t even factored-in the ‘double event’ of Lyons drinking in The Saddle as well as Barrett). Any other analysis will produce a different odds, but mine can’t because mine is highly-specific and can only be answered one way. If someone tries to tell you that this method is ‘wrong’, or ‘junk’, or whatever, please be wary. They may have ulterior motives in doing so.

RJ’s post says ‘Work on Dodd’s basement? Renovations at one of Michael Maybrick’s houses? Tearing out a shelf at the chambers of one of Maybrick’s former lawyer’s law offices? The excavation of John Over’s outhouse? An estate sale held at a neighboring house in Aigburth?’. These are all possible. Do we know if any actually happened? If we’re guessing that they did, then we’re guessing. If we know they did, we should show it. None of it changes the analysis I have been highlighting. That statistic will always be in the region of 1-in-37,557 (depending upon which day in May 1889 you wish to start on). If you guess that there were, say, 36 other events which were very much the same as Maybrick’s study floorboards coming-up, you could massage the odds down to around 37/37,557 which is a much less implausible 1/1,000, but even then you have a p value which is times times more unlikely than even the one (1/100) that the most demanding of researchers use in their experiments. If you want to count every apple falling anywhere in the world, I would accept that you could massage the odds all the way down until you had 1/1 and absolute certainty. Which, interestingly, is what RJ’s post has attempted to do.

On its simplest level, the astonishing ‘double event’ of March 9, 1992 blows the open mind. RJ’s post wants you to believe that the first part of the ‘double event’ was more or less a certainty. It’s up to you, dear readers, to decide whether you feel comfortable with that or whether - like me - you just look at that event and think ‘Wow!’. It matters not a jot what any of us believe though. The fact remains that the data we know about robustly supports the notion that the scrapbook came out of Battlecrease House on the morning of March 9, 1992, the very day that Fate decided someone should ring a literary agent at long last with the offer of a James Maybrick diary confessing to be Jack.

And not just any old ‘someone’, scattered anywhere across the UK or even the entire world, but actually a bloke from Liverpool who drank in the same pub as a member of the team working on Maybrick’s study floorboards that morning. If that doesn’t cork you, nothing will. You are unsurprisable.

Ike

[JeffHamm]

Hi all.

I'm only posting this as I just noticed that people are wondering what "I was given". As the implication is that RJ, or someone else, may have given me a biased presentation of what's been said, I think I'm beholden to clear anyone else of suspicion. Basically, Herlock told me that there were some discussions on statistical issues in the thread, and thought it would be of interest to me as stats is one of the things I do, and lecture on. Apparently RJ had mentioned somewhere that my opinion might be of interest, in part I suspect because the diary isn't one of the areas I follow. So, I read through the posts in this thread as they were presented, and wrote back to Herlock the response above and told him he could pass it on to RJ since he had mentioned me (though I'm not sure if that were in conversation with Herlock or if in the thread itself - I'm afraid I didn't have time to read the entire 440+ pages, so I may have missed that bit). Anyway, I stand by what I wrote, and again reiterate, that one cannot treat viewed data as if it were a prediction, which is basically what it happening here. There's no room for debate on that, it's just simply wrong to do so.

Anyway, in short, nobody presented me with any biased or skewed version. I've read through the posts, considered the arguements and examples used, and in the end RJ's concern that the odds were being calculated as if the data were unknown was, and is, and always will be entirely correct. Dealing with known data and working back to "what are the odds" is fraught with complexity, and the probabilities are not the same as if you started with a prediction and then seeing if that predicted outcome happens. Hence my example with the sequence of 10 coin flips. The odds of me getting that very sequence is 2^10, or 1 in 1024, but to say that's so rare I must not have got it except by design is a flawed reasoning based upon a fundamental misunderstanding of how inferential statistics work. As I tried to illustrate, with coin flips it is easy to work out all of the possible outcomes, and so forth. With events like calls to publishers and floorboards, etc, we're dealing with a far more complex space. We've got a minimum of 3 people (Mike, his wife, and his dead friend that was supposed to be the original supplier of the diary). Any story that begins at Battlecrease and connects with any of those 3 people prior to the call to the publisher would create "the link". Doesn't matter how one gets from Battlecrease to one of those three people, anything at all that can be put forth as a way to transport the diary from Battlecrease to that group of people is a potential explanation (just like all the possible patterns of 10 coin flips are potential outcomes). What are the odds that a story can be created? Extremely high, because there is no reason why the transfer had to happen the same day (in fact, to me that seems a problem as I would think it would take at least a few days for someone to first check over the diary, decide what they wanted to do - book? TV? Movie? Sell to collector? - so calling a book publisher right away seems a bit too fast. But, it's not impossible, just seems improbable). Regardless, give the flexibility of human creativity, it is almost impossible not to be able to come up with some way to tell a story that would start with the diary in Battlecrease and end up getting it into the necessary group. The fact that this particular story is able to utilise the pulling up of the floorboards is neither here nor there. If the floorboards came up the day after the phone call, one could just say they obviously had done some preliminary checks - or if the boards came up a month or even a year earlier, it took him some time to decide to unload it, and if there hadn't been work on Battlecrease, then someone had visited and nicked it, or worked there and took it, or it was found while clearing out something, and so on and so forth. The possibilities of a story are endless, the fact this one appears improbable is like my 10 coins, that sequence is improbable, but I didn't predict it before I knew the information. The floorboards and the publisher call were not things predicted, then looked for, and then found, they were found, and a connection was made because a connection could be thought of. That's not improbable, that's how human creativity works, it's highly likely to happen.

Anyway, as I say, it's not really something I care to discuss as if it's debatable because it's not. I do, however, applaud the idea of trying to apply some objective measures and techniques. Unfortunately, this kind of information is just too complex to deal with post-hoc to really try and put exact numbers to, but the odds of finding a story are almost 100% - humans are clever, we can invent all sorts of stories. It's what we do. I don't for a minute expect I will change anybody's views, though.

And so, perhaps most importantly, I just want to repeat, nobody gave me a skewed or biased presentation. I read a few pages of the thread around where stats and odds were being discussed, and presented my views based upon that.

- Jeff

[Iconoclast]

This must be the ‘sticking point.’ But Ike has to face the fact that millions of events happen at any given moment, and often it is only our human consciousness that ‘connects’ them—even though no actual connection exists.

I think we're back to our old friend pareidolia as an excuse for why Jack-related things just keep happening around James Maybrick. This is not the pareidolia which allows us to see human expressions in inanimate objects (the actual meaning of the word, sadly frequently confused with 'wishful thinking') but the 'pareidolia' which causes us to see associations between events which are not actually linked.

It's interesting that no such 'pareidolia' was employed by the original research teams prior to Keith Skinner's uncovering of the Portus & Rhodes timesheets. I don't recall anyone saying that the wallpaper in Michael Maybrick's old house on the Isle of Mann was refreshed four months before Mike Barrett contacted Rupert Crew, or that John Over's outhouse had new shelves put in thirty-three years before the call to Rupert Crew.

As I recall it, it was only the knowledge that James Maybrick's study floorboards had been raised for the very first time on the record on the morning of the day that Mike Barrett contacted Rupert Crew that alerted the waiting world that there was this elephant in the room (the elephant being statistical improbability).

Personally, I would not be looking for such tenuous links as houses hundreds of miles away being renovated, etc.. Personally, I'd be looking for events which were located as closely as possible to the very epicentre of James Maybrick's life. So, Battlecrease House definitely, and his office in Knowsley Buildings undoubtedly. And I'd be interested in events there which implied in any way that something hidden could be uncovered. This would allow us to do a different probability analysis. We don't have any definite evidence to work with so we are going to have to include in our hypothesis some guessed events. If I was guessing, I'd say that there would be maybe 37 separate days maximum since May 12, 1889 where something hidden could be uncovered in one of the two epicentres of Maybrick's life. You'll have worked out that I have chosen 37 because my maths is iffy and I know this produces odds of around 1/1,000. Still incredibly unlikely to have happened by chance, dear readers.

Personally, I wouldn't be casting my net further and further and further to try to find other guessed events which I could include so that the odds would keep on tumbling. But you might. RJ's post shows that he most certainly would and indeed has (without the slightest shred of evidence to support it, obviously). If that serves a purpose for you, then psychologically you will be motivated to argue for it. Doesn't make it robust, and it certainly doesn't change the 1/37,557 odds. The only thing - in my scenario - which would change those odds would be one or more other days coming to light (and on the record) when Maybrick's floorboards were raised (initially, I was referring to Battlecrease House only but I would be prepared to accept the day Knowsley Buildings came down so that would half the odds to a still eye-watering 1/18,500 or so).

Remember, we are not talking about the likelihood of someone contacting a literary agent with Maybrick's confession before or after one of these two events (described above) although very close proximity after the floorboards came up would obviously be intriguing - we are talking about the odds of that happening on the same day as one (or even both) of those events.

Ike

[Iconoclast]

And so, perhaps most importantly, I just want to repeat, nobody gave me a skewed or biased presentation. I read a few pages of the thread around where stats and odds were being discussed, and presented my views based upon that.

- Jeff

Hi Jeff,

So we have our expert on probability theory - you lecture in it and I never have (and my background in it isn't sophisticated enough to have done so) so I applaud the fact that we have the type of person we need to help to clarify this.

That said, I will not be budged on this point and I am intrigued to hear your view of the specifics. My point throughout this long and very tedious exchange has been that 37,557 days passed between James Maybrick dying (May 11, 1889) and his floorboards being raised for the very first (and only) time on the record on March 9, 1992. Now - this must surely cork you? - on that very same day, someone rang a literary agent for the very first time on the record regarding their interest in a diary of Jack the Ripper purportedly written (though he didn't realise it on that day) by James Maybrick.

What are the odds of those two events happening for the first time on the record on March 9, 1992?

No ifs, buts, or maybes. No "what about the next day?" or "what about Florence Maybrick's shack being demolished?" or "you have to include future days when the floorboards were raised". If my aunty had bollocks, she'd have been my uncle. But she didn't.

So, what was the cold, simple probability?

Your numerator can be 2 if you want (the raising of the floorboards in Maybrick's study on March 9, 1992, and the unknown day Maybrick's office in Knowlsey Buildings was demolished), but your denominator is 37,557 (given my premises, above).

You or I or RJ or Caz or my aunty with the genital confusion could change the scenario to add in other factors, but please don't do that. Please just address the premises as they are given here. What are the odds of someone ringing a literary agent on March 9, 1992 given that that was the only day definitely on record that Maybrick's study floorboards were raised (allowing if you wish for the certain fact that his office had long since been demolished)?

I'm not trying to be a smart arse here, Jeff. This is Billy Basic probability theory. I would not venture anywhere near the more complex stuff, and thankfully I don't need to.

Ike

[Iconoclast]

Similarly, Jeff wonders if a garage sale at Battlecrease would have been enough. Could the diary have been hidden inside an old sofa that had been sold? Or as part of a box of books that the owner didn’t even bother to thumb through before selling?

All of these are merely theoretical—just as the ‘floorboard’ provenance is theoretical. It just happens to be the ‘event’ that was noticed and investigated.

And any or all of these things could have happened. We don't know, and it doesn't matter in the context of the current discussion around a very specific and basic probability analysis.

And - for the record - lifting the floorboards in Maybrick's study for the first time on the record since he died is as smoking a gun as I personally would ever have hoped for (if ever I had). It doesn't prove that the scrapbook came out of Battlecrease House, but - Lord - the chances of that happening by chance alone on the very day someone finally contacted a literary agent!

And that's all before we get into the probability that that someone would conveniently live in Liverpool (rather than, say, Portsmouth or Inverness or Barcelona) and that a member of the Portus & Rhodes team would drink in the same pub as that someone. Wow! Imagine the size of Liverpool (if you don't want to imagine the size of the world or the UK or even just Merseyside) and imagine the population of Liverpool in 1992 (I think I previously produced a number of over a million inhabitants), and then imagine how comfortable you are that Maybrick's study floorboards were lifted on the same day a guy who drank in The Saddle first flagged-up the possibility that he had the diary of James the Ripper?

Jeff?

[Iconoclast]

FINIS

Finis my arse.