top of page
Search

What Are The Chances?

Lord Orsam
15 hours ago
21 min read

On 6th October 2026, out of the blue, and with no apparent context, unrelated to anything being discussed at the time, my old friend Roger Palmer started a thread on JTR Forums entitled "Ripperologists Can't Do Math" (he means "Maths" dear British people), saying:



"I've been renewing (sic) various attempts by Ripperologists to calculate probabilities, etc., and I've come to the conclusion that none of them can do math. One is as bad as the next. Cornwell, Edwards, Paley, Barrat, Holmgren, etc. They all suck equally".


This, by the way, is from the same guy who told us on 10th January 2018 that Anne Barrett penned 99% of the diary text but who then told us on 20th May 2022 that he suspected she penned 85-95% of the diary text but who then told us on 15th June 2022 that she might have penned 90% of the diary text but who then told us on 4th August 2025 that she might have penned 80% of the diary text but possibly only 50% but who then told us on 21 May 2026 that it might have been 70% of the diary text. Yes, this is the same person who is now lecturing others on the correct use of mathematics to calculate probability! 😮😆


From his use in his post of 6th October 2026 of the word "renewing", by which I can only assume he meant "reviewing", it could also be said that Palmer sucks at English as well as maths.


But it's odd, is it not, that, for no apparent reason he had suddenly been reviewing attempts of Ripperologists to calculate probabilities? And then he decided to start a thread about it. One which just happened to include me.


Was this post of his just another Gary Barnett-style attempt, as part of Roger's new nutty obsession with me, to make an Orsam hate post on JTR Forums (disguised as a general hate post) where he knows I can't respond? I bloody well think it was.


Anyway, although nothing about Roger's bizarre behaviour these days surprises me, I was nevertheless rather surprised to see my name in that short list of five "Ripperologists".


Apart from the fact that I'm not a Ripperologist, I was baffled as to what Roger was talking about. The only time I've ever calculated probabilities was on one now famous occasion in respect of the Maybrick diary but surely, surely, even though Roger has turned into a raving mad quasi-diary defender, he couldn't have been talking about that, could he?


[Reader: He could!]


When I first read his post, I could only think he had in his mind my dismissive response to his own crazy statistical claim which he made in the comments to one of my blog posts that because there were somewhere in the region of 600 to 1000 barristers practising in London in 1887 this somehow this means that Montague Druitt is likely to be the person who attacked Minnie Cameron on Southwark Bridge in September of that year. That claim is so ridiculous that, when added to his nonsensical estimates of Anne's contribution to the diary text set out above, surely Palmer should be in the list of "Ripperologists" who don't understand "math".


But it transpires that it wasn't that. It transpires that Roger has gone so far down the diary defending rabbit hole that he was, in fact, thinking of something I wrote in relation to the Maybrick Diary in September 2017, more than nine years ago.


We know this because he was asked the next day to clarify what he had in mind and he replied:



"Barrat fell prey to the post hoc data dredging and the Ludic fallacy in the opening paragraphs of his review of Robert Smith's book, coming up with 18 to 1 odds of two events coinciding, which is a meaningless number - white noise. Math cannot be used to make that calculation".


Roger is very coy here in his explanation whereby he claims that I came up "with 18 to 1 odds of two events coinciding". That is, of course, utterly meaningless. In the abstract, there are always 1 to 1 odds, with 100% certainty, of two events coinciding. Two events are coinciding all the time, every single hour, every single minute, every single second of every single day. That's because there are millions, billions, trillions or gazillions (or whatever the actual number is) of events happening all the time. It's absolutely guaranteed that two events will be coinciding at any particular time, if you don't specify what those events are.


What I actually said in my 2017 article about Robert Smith's book was not only entirely correct, and easily demonstrated to be so, but was later independently supported by an expert in statistics, Jeff Hamm, who used the exact same mathematical method as I did.


What I wrote in 2017, in essence, was that knowing that Portus & Rhodes electricians were working in Paul Dodd's house (Battlecrease) for 14 days in 1992, the chances of Mike Barrett having made a telephone call on a day in 1992 that electricians were working Battlecrease were 1 in 18. That is a mathematical certainty.


I can demonstrate this very easily.


THE DEMONSTRATION


Assume that electricians were working in Paul Dodd's house every single day of the year during 1992. What then is the probability of Mike Barrett having called Doreen Montgomery on a day that electricians were working in Battlecrease?


The answer is simplicity itself. It is 100%. It would have been impossible for electricians not to have been working in Battlecrease on day that Mike telephoned Battlecrease.

Bearing in mind that 1992 was a leap year, with 366 days, we can express this as 366/366 or 1/1.


Now, assume that electricians were working in Paul Dodd's house for 183 days in 1992, which is precisely half the number of days in the year. What then is the probability of Mike Barrett having called Doreen Montgomery on a day that electricians were working in Battlecrease?


The answer is again mathematically simple. It is 50%. Also expressed as 183/366 or 1/2.


So now let's assume that electricians were working in Paul Dodd's house for 91 and a half days in 1992. What then is the probability of Mike Barrett having called Doreen Montgomery on a day that electricians were working in Battlecrease?


The answer will not surprise you. It is 25%. Also expressed as 91.5/366 or 1/4.


Now, before we go on, there is something we need to consider. Electricians do not work every day of the year. They do not work on Christmas Day or other bank holidays. On the timesheets presented to us, it appeared to me in 2017 that the Portus & Rhodes electricians only worked Monday to Friday. Further, Mike was only going to have been able to speak to Doreen Montgomery at her London office on a working day.


It therefore makes sense to eliminate weekends and bank holidays from the calculation, which is what I did in 2017.


Once we do that, we are left with 254 possible days in 1992 on which (a) electricians could have been working at Paul Dodd's house and (b) Mike telephoned Doreen.


So let's go through it again using 254 instead of 366.


Assume that electricians were working in Paul Dodd's house every single working day of the year during 1992. What then is the probability of Mike Barrett having called Doreen Montgomery on a day that electricians were working in Battlecrease?


The answer is 100%. Also expressed as 254/254 or 1 in 1.


Now assume that electricians were working in Paul Dodd's house for 127 days (half the number of working days in the year) in 1992. What then is the probability of Mike Barrett having called Doreen Montgomery on a day that electricians were working in Battlecrease?


The answer is 50%. Also expressed as 127/254 or 1/2.


Now assume that electricians were working in Paul Dodd's house for 63 and a half days in 1992. What then is the probability of Mike Barrett having called Doreen Montgomery on a day that electricians were working in Battlecrease?


The answer is 25%. Also expressed as 63.5/254 or 1 in 4.


And so we now assume (as was in fact the case) that electricians were working in Paul Dodd's house for 14 days in 1993. What then is the probability of Mike Barrett having called Doreen Montgomery on a day that electricians were working in Battlecrease?


The answer is 5.5%. Also expressed as 14/254 or Ta da! 1 in18 (actually 1 in 18.1428571429 if we want to be pedantic).


This is mathematical certainty. There can be no dispute over the maths based upon the assumptions we have factored in. The answer is 1 in 18, just as I said in 2017. There was a 1 in 18 chance that Mike Barrett telephoned Doreen Montgomery on a day in 1992 that Portus & Rhodes electricians were working in Paul Dodd's house.


THE VINDICATION


That my mathematics was perfect is proved by the fact that Jeff Hamm, an expert in statistics, independently confirmed my methodology four years later. I stress the word "independently" because he appears to have been entirely unaware of what I wrote in 2017. Here is his Casebook post of 24th July 2021:



We can see that he said:


"However, you seem to think the simplistic examples used to teach probability theory can just be transplanted into the real world they can't because the real world is far more complicated. House work isn't a flat distribution, for example. So if all we're debating is the probability of Battlecrease having floor work donrme on that day then use of a flat distribution will be wrong. Since we don't know how that increase over time function works, we need an estimate for the probability of work around that time.

​​​​​​ That's where the 14 days of work in the past year come in ( I think I've seen that said, perhaps I'm misremembering, but again, it's the underlying principle that matters here). So 14/365.25 is about 3.8%. now, that's a bit better, but there are seasonal changes in work rates, some repairs are more common in spring and fall, for example. We would need to do some research, work out how common heater installation is over the year, etc and we need to factor in the age of the house still for that (is heater installation more probable if the house is old? Etc)...given the rate of work being done at around the time of interest seems to be 14 days out if a year, then the actual probability is more likely to be in the vicinity of 3.8% further adjusted by the seasonal rate of heater installation. Basically, if it is the busy season for heater installation, the the probability will be more than 3.8%, but if it is a slow time of year, the probability will go down. Other factors will also come into play, change of ownership for example

now we could argue over what factors to include, and so forth (we've already agreed that prob of work inreases with property age, so we both agree your flat distribution is wrong) but that seems a strange topic for a forum about JtR?"


Leaving aside for the moment Jeff Hamm's point about the seasonal rate of heater installation, what he did here was to take the 14 days worked by the electricians in 1992 and, forgetting that 1992 was a leap year, used the number of days of 365.25 (although, if not a leap year, he should only have used 365 days) to calculate a probability of 14/365.25 as 3.8%. (which would have been the same result had he used 365 or 366 days).


That is the exact same methodology as I used myself except I removed non-working days from the calculation which (as Roger himself publicly agreed in 2021) was the more correct approach.


So we can see that when Hamm wrote:


"First, it probably is lowish, but more like 3-4% low. It is definately not 1/37000+, and 1/18 I've seen too, which I have no idea what that is based on but comes to about 5.5%, which seems a bit high but is at least closer to the mark."


it hadn't occurred to him that 254 was the correct number of days rather than 365 or 366 but his 3.8% figure, once adjusted to take the working days into account, is the exact same figure, or rather calculated in the exact same way, as my 5.5% figure.


My maths is sound.


And we may also note what Jeff said in a later post the same day:



"knowing there were 14 repairs within the last year gives us a better estimate of the rate of work at the time in question."


And this is the essence of the point I was making. Portus & Rhodes did not only work in Battlecrease on 9th March 1992. They were there for 13 other days in 1992. So if, knowing that Mike called Doreen on 9th March 1992, you want to know how remarkable it is that electricians were working in Battlecrease on that same day, you have to factor in that there were 14 days in total that Mike could have made his call to Doreen during 1992 when there would have been electricians in the property. Those additional days cannot be ignored if you want to talk about probability. At least if you want to talk, as I was doing, about the probability of the Portus & Rhodes electricians doing work in Paul Dodd's house on any single working day in 1992.


THE CONTROVERSY


If the calculation is so mathematically straightforward, why do people keep banging on about it?


Why is it in any way controversial?


Probably because diary defenders claim that the odds of Mike Barrett calling Doreen Montgomery on the same day that floorboards were being lifted in Paul Dodd's house are astronomical and, in their confusion, believe that I was attempting the same calculation as they are.


But I need to make clear that my calculation, like that of Jeff Hamm, had nothing whatsoever to do with the lifting of the floorboards. That's what the diary defenders don't seem to understand about the 1 in 18 number. Once you include the floorboard lifting it becomes an entirely different matter, unrelated to anything I was talking about in my 2017 article.


Not that everyone understands this. Here is Tom Mitchell, thick as a proverbial plank, getting it all wrong (as usual) in 2020:



We can see here that he wrote:


"Lord David Orsam...By the way, is that the same fellow who argued that - although some 35,000+ days had passed since Maybrick's death during which there was not a single piece of evidence that the floorboards in Battlecrease House had been lifted - the odds of their being lifted on March 9, 1992 (the same day that Mike Barrett contacted Doreen Montgomery about his still-wet manuscript) were just 1 in 18!!!!!!!!!!!!!!!!!????????? "


That was a fail of the highest order. As I've already said, I didn't mention anything to do with the floorboards in my 1 in 18 calculation of probability which was all to do with electrical work being carried out in Battlecrease House on that day. Yet in Tom's mind, I was saying that the odds of the floorboards being lifted on 9th March 1992 were 1 in 18. An extraordinary misrepresentation of my position. I repeat that the floorboards have nothing whatsoever to do with the 1 in 18 probability figure.


Gary "the Clanger" Barnett also weighed into the controversy on 22nd July 2021 in his own uniquely stupid way when he posted at Jeff Hamm:



"All way above my head, but I’m assuming that, counter-intuitively, 366/254 could be equally as nefarious as 37,000.


‘O’ dear!"


The "O" in "'O' dear!", in case you missed the allusion, was an unmistakable reference to Orsam, so the Clanger was clearly suggesting that I had come up with a "nefarious" number. The Clanger here obviously had in mind the fact that I had mentioned in my 2017 article that there were 366 days in 1992 of which 254 were working days and, like Roger Palmer is now, was obsessed with trying to prove me wrong about something, anything. But his post was completely misplaced because 366/254 wasn't a probability calculation of mine at all. As I said in response to the Clanger's post my It's a Lottery! article published on the old site on 18th September 2021:


"God only knows why the Clanger has taken those two numbers, expressed them as 366/254 and then compared them to 37,000. They are wholly unrelated."


As I also said at the time, "The Clanger is a nincompoop."


And one thing that is most odd about Roger's recent JTR Forums post is that, despite having posted repeatedly on Casebook about where Tom Mitchell has gone wrong in his claim that the odds of Mike calling Doreen on the same day the floorboards were being lifted in Paul Dodd's house are one in thirty seven thousand - about which I entirely agree with Roger - he doesn't include Tom Mitchell in his list of Ripperologists who don't understand maths. Funny that. 🤔


ROGER'S TWIST


Roger's sudden criticism of my maths ability is even more astonishing bearing in mind that Roger addressed my probability calculation in a Casebook post dated 24th July 2021 in reply to Jeff Ham who had mentioned the 1/18 number (not, I believe, from having read my article but from others having mentioned it on Casebook). There was, in that post, nothing from Roger about "post hoc data dredging and the Ludic fallacy", nor any serious criticism of the mathematical basis of the calculation. Here is that post:



We can see that in 2021 Roger literally defended my approach, saying:


"All he was trying to express is that it would be somewhat unremarkable for the owner of Battlecreate to have work done on his house on any random day - the same day when a person might call a literary agent".


Now Roger did say that the calculation was "open to the objection that he used a limited sample - the year 1992" and suggested that I would agree with this. But I do not agree with this. Roger has failed to understand that what I was calculating in 2017 was the probability of Mike's conversation, which was known to have been in 1992, having taken place on a day when electricians were working in Paul Dodd's house, knowing that electricians were working in Paul Dodd's house for 14 days in 1992.


What I asked in my 2017 article was:


'What are the chances, do you think, of electricians working at Battlecrease on the very same day that Mike was telephoning a literary agency in London...'


I then explained:


"The timesheet evidence revealed in Robert Smith's new book, '25 Years of the Diary of Jack The Ripper: The True Facts' shows that there were electricians from Portus and Rhodes working in Battlecrease for 14 days during 1992 (i.e. 9 and 10 March, 9, 10, 12, 15 and 16 June, 15, 16, 17, 18, 19, 20 and 21 July). There were 366 days of the year in 1992 (a leap year) but electricians wouldn't have been working during weekends or bank holidays.  If we exclude all weekends and bank holidays, that leaves 254 days in 1992 on which the Portus and Rhodes electricians could have been doing work in Battlecrease.


If they had only been in there on one day of the year, that would have meant the chances of it being 9th March 1992 would have been one in 254 (not in itself utterly incredible) but, as they were working for 14 days during the year, the correct chances are 14 in 254 otherwise expressed as 1 in 18." 


It wasn't a calculation designed to say that there was a 1 in 18 chance of Mike having made a telephone conversation at any time in his life on a day that there were electricians working in Battlecrease. That would have been ridiculous. If anyone read it that way in my 2017 article they simply read it wrong and misunderstood it. If Roger read it that way in my 2017 article he simply read it wrong and misunderstood it. But there is no excuse for Roger because he would certainly have read me say in my It's a Lottery! article published on my old website on 18th September 2021 (bold added):


"...the essence of what I was doing was calculating the chances of electricians working in Battlecrease on any single working day during 1992  (which, in this case, was 9 March 1992 but it could equally have been 11 March or 15 October) so that it didn't even matter what else happened on that day.

 

My entire point was that if you are attempting to calculate the chances of electricians working in the house on ONE day during the year you absolutely cannot ignore the fact that they were working in Battlecrease for a total of 14 days in the year.  If you do, you lose sight of the chances of them working in there on 9 March 1992 (because, for example, as we know, they were ALSO working in there on 10 March 1992)."

 

I also stated in the same 2021 article:

"We already knew that Mike telephoned Doreen on 9 March 1992, so what I was demonstrating was the chances of electricians working in Battlecrease on THAT specific date within the year....With the electricians working in Battlecrease for 14 days, then, on the basis that they would only have been working on weekdays, my calculation was mathematically correct."

 

There can, therefore, be no room for confusion. As I've said, I wrote those things more than five years ago. Roger, who reads all my articles, must have read it. So he cannot claim in good faith that he didn't understand what I meant in my 2017 article.


I should add that the calculation I did is not affected in any way by whether 1992 was an average year for workmen or not. It was a calculation which only applies to 1992 because Mike made his telephone call in 1992.


It is true that the calculation is based on the assumption that the 14 days on which the electricians were working in Battlecrease were perfectly random and that there is no particular time of the year in which Portus & Rhodes electricians were more or less likely to be working the house, so that seasonal factors were not relevant. This is something that Jeff Hamm mentioned. He was right to say that there could be factual elements, over and above the pure mathematics, in the same way that a weighted coin would distort the mathematical probability of a coin toss producing a head (or tail) fifty percent of the time. The fact of a weighted coin, however, would not change the basic mathematics which calculate a 50/50 probability based on a non-weighted coin. But I think it's wrong that Jeff includes as a seasonal factor the question of when storage heaters are likely to be installed. The storage heaters don't have anything to do with this question. It's irrelevant what type of work the electricians were doing in Battlecrease. The only question is whether electricians were in Battlecrease on the day that Mike called. They could have been doing any sort of electrical work. My own assumption is that electrical work occurs all year round at about the same rate. It's probably different for, say, plumbers, but I'm not sure there is a seasonal factor for electrical work. If there is, then, absolutely, Jeff would be right to include it as a factor affecting the probability. But that doesn't have any effect on the mathematics of the calculation, which is what Roger was claiming I couldn't do. Whereas, in fact, the mathematics is sound and solid.


We can see that Roger concluded his 24 July 2021 post by saying:


"I suspect that 18 to 1 is closer to the mark than 37,577 to 1."


While sympathetic to what I was saying, this statement reveals Roger's failure of understanding because my calculation of 18 to 1 bears no relation whatsoever to Tom's 37,577 to 1 which relates to the lifting of the floorboards.


It gets worse though because 22 days earlier, Roger had posted this strong defence of my calculation:



"Ike, with all due respect, and very little is due at this point, the above comment exposes your ignorance of statistical analysis. You simply do not understand what statistics are, or when they should be used, nor how Orsam was using them, nor their limitations.


All Lord Orsam was calculating is how likely it would have been for an electrician to have been working at Dodd’s house on any given workday in 1992 (workdays being the only days Mike could have called Crew Literary Agency). That’s it. You may not have liked his exercise, but that’s what he was calculating. As such, of course he had to count all the known days electricians had worked for Dodd that year--before during and after. That’s how probabilities work. Hell, it could have been a lot worse. If we were to learn that Dodd had more electricians working the following November/December, long after the diary had been found, they would have been counted, too. And, come to think of it, Orsam could have counted all days that ANY type of workman was on Dodd’s property, just as long as he admitted that that is what he was doing, since the diary doesn’t say anything specifically about hiding the diary in a place that only an electrician would look."


Somehow Roger managed to understand what I was saying perfectly in 2021. Bizarrely he seems to have forgotten it in 2026. What could possibly have changed in the meantime?


He did then add a postscript telling Tom Mitchell how he should have attacked my calculation (but getting it wrong for the reasons I've already stated):



"Ike, had you really wanted to attack Lord Orsam’s statistical analysis in any intelligent way, you should have said. “Hold the phone! Foul! 1992 was not a typical year for Dodd. It’s not EVERY year that he has electricians in! It’s misleading to state that it was 18 to 1. He had no electricians in 1991, for instance.”


But, really, mate, he was being somewhat easy on you. Barrett having first called Crew on March 10th would have been just as much of a coincidence as March 9th (and some say it would have made the timing far more credible), so the odds really should be reduced to 9 to 1. And since the diary doesn’t say Jack Shite about floorboards, nor even hiding the diary at Battlecrease, it would have been appropriate to add in other visits by other workmen who could have led to suspicions about the diary having been unearthed—plumbers, etc. And other buildings significant to Maybrick’s life other than Dodd’s. You claim this is unfair, but the very vagueness of the diary forces us to do so."


Had Tom decided upon the line of attack suggested by Roger it would have failed because, as Jeff Hamm understood, the calculation I did relates solely to the number of days worked in 1992. Had Mike Barrett made his call in 1991 it would have been a completely different situation but he didn't. He made the call on a day in 1992 which is why the 14 days worked by the electricians in Battlecrease in 1992 is mathematically relevant.


That wasn't all, though, because after the Clanger's disastrous attempt at trying to mock my imaginary "366/254" claim, Roger posted:



"I don't think Orsam was seriously suggesting that the odds of Dodd's floorboards coinciding with Barrett calling a literary agent was 18 to 1.


People chose to interpret it that way, but I think all he was pointing out is that it isn't all that strange for a man to have work done on his house. We know of 14 days in 1992 when this happened at 'Battlecrease.' And since the diary doesn't specifically mention floorboards, any work at Battlecrease could have triggered a provenance tale."


It's funny ain't it, how five years ago Roger managed to understand quite easily what I was saying which was that one has to take into account the fact that electrical work was being done in Battlecrease for 14 days during 1992, and wasn't only just done on one day, yet, this year, he has suddenly lost his mind and now thinks that what I was saying was all entirely mathematically wrong?


To me this suggests that Roger is posting about me now in pure bad faith with ulterior motives, suffering from a severe case of Orsam Derangement Syndrome, exactly as the Clanger did.


THE PROBLEM WITH ROGER


A very disturbing pattern is emerging with Roger. First he attacked Michael Banks on JTR Forums, while also attacking me for no apparent reason. As I set out in great detail in my recent blog post We need to talk about Roger, his arguments were sloppy, inconsistent, all over the place and entirely contradictory of things he had posted in the past. Then, in a massive shift from his normal posting behaviour, he completely stopped posting on the forums about the Maybrick diary with one sole exception being this extraordinary diary defender friendly post:



Heaven only knows why he concluded his post with the words "Okay, Mike. Thanks" unless he is going through some sort of mental health crisis. Was he talking to the late Mike Barrett? Or to the living Michael Banks? Who knows. It's weird and creepy either way.


The rest of it was another out of the blue comment unrelated to anything anyone else was discussing and was completely absurd. As we can see, he wrote:


"I think these are my favorite lines from Barrett's sleazy confession:


"I felt Maybrick was an ideal candidate for Jack the Ripper. Most important of all, he could not defend himself."


In confession where he implicates two dead men who can't defend themselves---Tony Devereux and Billy Graham--Barrett admits that he finds it efficacious and desirable to accuse a dead man who can't defend himself."


I call it absurd because the main person Mike accused of forging the Jack the Ripper diary was his wife, who was very much alive and able to defend herself (although what she actually did was refuse to talk about the diary). So what was the purpose of this diary defender friendly post?


And it was certainly recognised as diary defender friendly by none other than the Chief Diary Defender Caroline Morris who replied on 26th August in a very friendly way to say that, "Mike played a similar game in 1993 when he swore an affidavit that Tony had really given him the diary."


Well done, Roger. He seems to have lost all interest in challenging the nonsense posted by Morris and Mitchell on JTR Forums which, until the "Diary Forgery" thread was mercifully closed, was occurring on a daily basis. All he seems to be interested in these days is attacking Orsam. That's it. That's all he wants to do now. Incredible.


And that, in my view, was the sole purpose of his odd post about "Ripperologists" and maths. My name was fourth in his list but it was really the only name he was interested in.


It's crazy.


As I discussed in We need to talk about Roger, I don't know if the guy is still fuming about the fact that I disproved his silly claim made in his 2010 Ripperologist article that "..it is clear from Lushington's letter that Robert Anderson had inquired about the feasibility of sending a man to North America and was further asking if he could use the extradition of the prisoner to Roland Barnett to do it." As I pointed out in my later 2015 article, after having discovered Robert Anderson's hitherto unseen letter in an obscure file in the National Archives, Anderson had made no such inquiry (or further inquiry). So maybe someone should start a JTR forums thread about how "Ripperologists" cannot be trusted to speculate about the contents of documents they haven't seen. Or perhaps he's not bothered about that but is still seething from my mild disagreement of his apparent belief that Montague Druitt is likely to have been the person who attacked Minnie Cameron in 1887. It's got to be one of those two things. Whichever one it is means that Roger is a very troubled person who is sadly raging about stuff he shouldn't be in the slightest bit concerned about. But he's obviously so annoyed by something I've done that he's turned into a quasi-diary defender who, Clanger style, now wants to go over everything I've ever posted and try to find flaws in it. Who knows what he will be reviewing next? Will he be telling us that "one off instance" might be a Victorian expression? Nothing about Roger will surprise me any more.


He is, of course, invited to reply to this post in the comments section and to explain to me, and the readers of the post, why my mathematics is incorrect but, hopefully, without the gobbledigook of post hoc data dredging and the Ludic fallacy, clever though that might make him sound.


SOME PROPER MATHS


What is the probability of Roger Palmer continuing to be obsessed with Lord Orsam and consumed with irrational hatred towards him? Answer 100%


What is the probability that we will ever learn the true reason why Roger continues to be obsessed with Lord Orsam and consumed with irrational hatred towards him? Answer 0%


Who said I can't do math?


And it's maths for fucks sake.


LORD ORSAM

19 November 2026















 
 
 

8 Comments

Rated 0 out of 5 stars.
No ratings yet

Add a rating
Guest
3 hours ago

Jeff's original post, emphasis added in bold:

 

JeffHamm, 07-24-2021, 11:49 AM

 

Ike, what you seen to think are admissions are what I've been saying all along. Glad to see you finally comprehended. However, you seem to think the simplistic examples used to teach probability theory can just be transplanted into the real world they can't because the real world is far more complicated. House work isn't a flat distribution, for example. So if all we're debating is the probability of Battlecrease having floor work done on that day then use of a flat distribution will be wrong. Since we don't know how that increase over time function works, we need an estimate for the probability of work around that…


Like
Lord Orsam
2 hours ago
Replying to

I dealt with this in detail in my blog post, Roger. I'll repeat what I said because you don't seem to have read it with any care or attention.


Firstly, I think Jeff was wrong to ask how common heater installation is over the year (in a part of his post you haven't highlighted), because the actual electrical work being done is irrelevant, but I also said that I don't believe there are seasonal changes in electrical work being done across the year. Why would there be? It's not like plumbing. So I just don't think it's a factor. But there is also a key part of his post which you've quoted but failed to highlight. This bit:


“given the…


Like

R. J. Palmer
3 hours ago

Hi David,


I've now had a chance to study Jeff Hamm's original post, which can be found on the "One Incontrovertible" thread on Casebook, dated 12 July 2021 at 11:49 a.m. I'll attach the relevant passage to a second post after making a few general observations.


You wrote that your calculation was subsequently "independently supported by Jeff Hamm," who supposedly used the exact same mathematical method. But Hamm explicitly states that he has no idea what the 1-in-18 figure was based on. He merely observes that the figure (5.5%) mentioned earlier in the thread "seemed a little high," although closer to the mark than 1 in 37,000+.


Hamm's own calculation began with a different figure, 14/365.25, and he stressed that…


Like
Lord Orsam
2 hours ago
Replying to

No, Roger, Jeff Hamm never read my original 1-in-18 calculation. I already made the point in my blog post that he came to the exact same result independently of me, once adjusted for working days. But he knew about my 1-in-18 calculation because other people had mentioned it on the boards. Had you read his posts (and mine) more carefully you would have spotted that he said in his post of 24 July 2021:


"First, it probably is lowish, but more like 3-4% low. It is definately not 1/37000+, and 1/18 I've seen too, which I have no idea what that is based on but comes to about 5.5%, which seems a bit high but is at least closer to the…


Like

Guest
9 hours ago

Oh, I'm totally composed, David. You seem to suffer under the misconception that you somehow "rattle me" and that I'm in state of breathless anxiety which isn't even remotely true. I like to proceed slowly and read with care your claim that Jeff endorsed your statistics was stated very earlier on in the article so I wanted immediate clarification because I found it shocking that a trained statistician would do so. I'll have to go back and re-read Jeff's entire posts and see the full context. As such, be warned that I may need to respond on Casebook to give Jeff a chance to clarify himself (if need be). I know you won't be able to respond there b…


Like
Lord Orsam
8 hours ago
Replying to

If you're so composed, Roger, how did you miss the fact that the very source references you asked me for are contained within the blog post? You haven't explained it. I can only assume that the red mist descended and you raced through my blog post at top speed without reading it properly.


Let me help you with Jeff Hamm. The probability he independently said he calculated was 3.8% or 1 in 26. That was the EXACT SAME mathematical calculation as I did, except for the fact that he included weekends and bank holidays in his calculation whereas I removed them to produce a probability of 5.5% or 1 in 18. Even if you disagree with the premise of exclu…


Like

R. J. Palmer
10 hours ago

"What I actually said in my 2017 article about Robert Smith's book was not only entirely correct, and easily demonstrated to be so, but was later independently supported by an expert in statistics, Jeff Hamm, who used the exact same mathematical method as I did."


Hi David -- Before I demonstrate the flaws in your understanding of statistics and where and how you went wrong (which can be easily demonstrated), can you give me a citation for the above claim so I can track it down, because I very much doubt that Jeff Hamm endorsed the legitimacy of your equation, nor do I have any memory of him doing so. Of course, it is possible my memory is flawed. …

Like
Lord Orsam
10 hours ago
Replying to

Roger, it's all set out in the blog post in which I've reproduced, in full, Jeff Hamm's posts of 24 and 25 July 2021 (plus pasting the relevant extracts) together with a number of your own relevant posts from July 2021. I find it unfathomable and inexplicable that you could have read my blog post yet missed them. Can I suggest you take a deep breath, compose yourself, and, this time, slowly and carefully, read every word of the blog post. If you do this, I rather suspect that your bravado about how you are going to demonstrate "the flaws" in my understanding of statistics will evaporate but, if not, I'll be here and interested to read all about it.

Like

©2024 by Orsam Music for Bubblegumelicious.

Created with Wix.com

The Sagar Saga: Finding the Missing Ling

In January 1905, former City detective inspector Robert Sagar, who had just retired, apparently gave interviews to reporters of four London newspapers in which he spoke of his knowledge of Jack the Ripper. 

Sagar 1.jpg

The articles containing these interviews, while very similar (indicating that they must have taken place at the same time) are all, nevertheless, a little bit different from each other and, in October 2020, Chris Phillips prepared a helpful comparison of the four reports featured side by side, which can be found here [Chris has now updated them to include the Evening News here].  He had, however, only located three reports from London newspapers, but was aware that there must have been a fourth report because such a report (different from the other three) was carried by some American newspapers, hence he included the Seattle Daily Times (of 4 February 1905) as his fourth report.

I have, however, now located the additional report in a London newspaper.  It was in the Evening News of Saturday, 7 January 1905.  It's not identical to the Seattle Daily Times report but it's close enough, especially in respect of the bit about Jack the Ripper, that, for comparison purposes, what Chris has labelled the Seattle Times report, can be regarded as the Evening News report.  There are, nevertheless, some important differences between the two which I will be discussing. 

So the four "interview" reports with Sagar that we have, are:

1. City Press of Saturday, 7 January, 1905.

2. Evening News of Saturday, 7 January, 1905.

3. Morning Leader of Monday, 9 January, 1905.

4. Daily News of Monday, 9 January, 1905 

A quick glance through these reports reveals that one of them is, curiously, very different to the others.

While the reports in the Evening News, Morning Leader and Daily News all contain quotes from Sagar, the City Press does not.  Its report is written purely in the third person, with no hint that its reporter had ever even spoken to the former detective.

The Morning Leader on the other hand tells us that, 'To a "Morning Leader" representative Mr. Sagar related some of his experiences'.  The Daily News is even more explicit as to when its reporter spoke to Sagar.  Hence, we are told that Sagar spoke to 'a representative of 'The Daily News on Saturday'.

That would seem to make sense.  All four reporters spoke to Robert Sagar on Saturday, 7 January 1905, right?

WRONG!

That's impossible. 

The City Press newspaper was published early on Saturday morning, so that it could not possibly have interviewed Sagar on Saturday and carried a report of the interview in its Saturday edition.  Here is the proof that the City Press was published on Saturday morning in January 1905:

Sagar 2.jpg

I'm a paragraph. Click here to add your own text and edit me. It's easy.

I'm a paragraph. Click here to add your own text and edit me. It's easy.

I'm a paragraph. Click here to add your own text and edit me. It's easy.

bottom of page