platonov
Master Poster
- Joined
- Sep 11, 2010
- Messages
- 2,339
"One little chap then had a mishap .."





What a long convoluted post - all the better to slip in a possible withdrawal while accusing me
of getting the maths wrong.
But its nonsense - your stats / maths are not quite right I'm afraid ; indeed on a par with your gastric analysis.
By your earlier post / stats there was a 98% chance MK was dead by 9.00 on your science alone & a 99.9% probability that she was dead by 9.20.
I will happly post it again if necessary and for as long as it takes to put this nonsense to bed for once and for all.
All the above talk of 95 % up to 10.00 or whatever cant make the 98 % figure get any smaller -
& its still 99.9 % by 9.20
Maybe the penny has finally dropped on what the defence docs are claiming ( only claiming - its not scientific fact)
Suddenly all the months of precise calculations with (very tall) lone wolves poring over the literature night after night has been cast aside with a bland statement that 10.00 is the upper bound [& an associated view of a high probability that it was before 9.30pm]
.
As I said, I'm not interested in engaging you any longer. But I thought I'd point out your statistical error in trying to revise my probabilities.
There's only (by my calculations) around 2% probability that Meredith was still alive at 9pm (150 minutes after eating her meal). This is clearly very improbable (1 person in 50), but since we know that she was indeed alive at T(lag) = 150 min, then we clearly accept it.
To make the analogy to human height, lets use an actual study of adult male height in Australia:
http://jcem.endojournals.org/cgi/content/full/86/9/4147
In this study of 413 adult males, the mean height was 174.1cm (5ft 8.5 inches), with an SD of 7cm (2.8 inches). Now, two SDs from the mean corresponds to the 97.7% point - which is a good comparison for the 98% point for T(lag) = 150min.
So to say that Meredith had a T(lag) of over 150min is analogous to saying that an adult male is at least (174.1 + (2*7)) = 188.1cm tall (= just over 6ft2). Most of us know one or two adult friends who are this height or over, so whilst it's uncommon (1 in 50 of adult males), it's not extremely unlikely.
Now let's look at the 4 SD point on the curve. This corresponds roughly to the 99.95% mark. 174.1 + (4*7) = 202.1cm (= 6ft 7.6 inches). The data suggest that only one per 250 adult men is 6ft 7.6 or taller. And this is probably in keeping with most people's life experiences. I don't think I've ever known anyone this tall, although I've seen a few strangers.
So the question is, if we know that an adult male is at least 6ft 2 tall, what are the odds that he's between 6ft 2 and 6ft 7.6, as opposed to being taller than 6ft 7.6. This research would suggest that he's about 95% likely to be between 6ft 2 and 6ft 7.6.
And exactly the same analysis can be applied to ToD. The only relevant question is this: given that Meredith definitely had a T(lag) of greater than 150 mins, what is the probability (given the experimental data-derived curve) that her T(lag) was between 150 and 170 min. And the answer is the same - 95%. But there's still a 5% chance that T(lag) was greater than 170m. However, this 5% almost all applies to a T(lag) between 170 min and 210 min. The probability of Meredith having a T(lag) of over 210 minutes is virtually zero, for practical purposes. And this corresponds to a ToD of 10pm. And I've always put the absolute upper level of ToD at 10pm for this reason, with my associated view of a high probability that it was before 9.30pm.
What a long convoluted post - all the better to slip in a possible withdrawal while accusing me
But its nonsense - your stats / maths are not quite right I'm afraid ; indeed on a par with your gastric analysis.
By your earlier post / stats there was a 98% chance MK was dead by 9.00 on your science alone & a 99.9% probability that she was dead by 9.20.
I will happly post it again if necessary and for as long as it takes to put this nonsense to bed for once and for all.
All the above talk of 95 % up to 10.00 or whatever cant make the 98 % figure get any smaller -
& its still 99.9 % by 9.20
Maybe the penny has finally dropped on what the defence docs are claiming ( only claiming - its not scientific fact)
Suddenly all the months of precise calculations with (very tall) lone wolves poring over the literature night after night has been cast aside with a bland statement that 10.00 is the upper bound [& an associated view of a high probability that it was before 9.30pm]
.
Last edited:
what kind of 'science' is this [ just spotted it now ]