Lord Mance made the following comments in the Supreme Court about findings of fact in causation cases.
48 Nor do I accept Mr Flaux’s submission that Richards LJ was lured, by a process of elimination, into accepting as the probable cause the least unlikely of a range of possibilities all of them unlikely. That was the error the House identified in the approach taken by the judge at first instance in Rhesa Shipping Co SA v Edmunds (The Popi M)  1 WLR 948..
49 Richards LJ summarised his conclusions as follows  1 Lloyd’s Rep 279 , 299:
“79. Looking at the matter overall, it seems to me that the judge did overstate the factors telling against employee theft and understate the factors telling in favour of it. I consider employee theft to be a much more likely explanation than the judge found it to be. Perhaps more importantly, I regard as implausible and improbable the explanations of accidental loss to which the judge referred when concluding that accidental loss was more likely than employee theft. ….
81. My conclusion does not depend on the separate loss of the Axxis package, but I accept the submissions by Mr Reeve that the loss of the Axxis package adds to the improbability of other possible causes and makes employee theft all the more probable.
82. I have borne very much in mind the observations of Brooke LJ in Laceys Footwear (Wholesale) Ltd v Bowler International Freight Ltd  2 Lloyd’s Rep 369 with which I am in respectful and total agreement, as to the need for a properly rigorous approach to the available evidence. It is the evidence, properly analysed, which in my view leads to the conclusion. That is also why I disagree with the judge’s description of the claimants’ case as ‘too speculative’.”
50 I find the reasons given by Richards LJ for reversing the judge compelling.
None of the possibilities mentioned by the judge in para 66 affords any plausible explanation of the disappearance of the three packages, still less of all the four that were due for delivery to L&A on 26 July 2002.
In their joint memorandum the two experts were in fact agreed that the possibilities of loss, missorted, mislaid and damaged/thrown away/sold at auction, under head I were each “less likely than others”, in view of the sighting of the packages…correctly stacked and labelled on the spur shortly before loading.
None of these possibilities anyway offers any comprehensible explanation for the disappearance of three (or in fact four) large and valuable packages.
The possibilities, mis-delivered and delivered in error, under head II run up, as previously stated, against the inherent implausibility of three or four separate packages due for delivery to L&A all being innocently misdelivered on the same day without any DIAD signature being obtained from anyone.
The possibilities floated before the judge (but not even mentioned by him in his para 66) under head II, labelling issues, are remote in the extreme for the reasons given in para 27 above.
As to head IV, theft, the joint memorandum categorised all the possibilities as “less likely”, except for those involving a UPS driver or employees, and the judge found that, if the packages were stolen, it was probably by a UPS employee.
Inevitably, any systematic consideration of the possibilities is subject to a risk that it may become a process of elimination leading to no more than a conclusion regarding the least unlikely cause of loss. But, as I have said, I do not consider that Richards LJ fell into that trap. I share, without hesitation, the view which he formed overall that theft involving a UPS employee was shown on a strong balance of probability to have been the cause of this loss.
Spacing added for ease of reading. See other cases listed under this tag.