Common sense dictates that the person or group named Satoshi Nakamoto would never trust someone else with the means of moving his (their) stash, so does that not lead to a conclusion Mark Karpeles must be Satoshi, since he would of course trust himself?

Even Craig Wright or Dave Kleiman seem a lot more believable to me than Mark for a Satoshi role. The amount is virtually the same: 648,000 BTC declared missing from MtGox  vs. 650,000 BTC withdrawn from the Seychelles trust by Craig.

As mentioned in the LRB article http://www.lrb.co.uk/v38/n13/andrew-ohagan/the-satoshi-affair, Craig was planning to use these funds to pay for his "research and development stuff." It seems natural that an exchange like MtGox would be chosen to trade them all for fiat. Craig claims to have lost a "large wallet" when the exchange went down.

While it seems counter-intuitive since we were first told MtGox had a ginormous number of users, yet by May 25 of this year only 24,750 creditors had ever bothered to files claims: 9,863 of these were bitcoin-only claims, and 7,952 bitcoin claimants were subsequently approved by the trustee, yielding an average of 25.4 BTC per claimant.

The worldwide number of users, at least those of whom could pass muster with the trustee, is a far smaller one than early media accounts of the collapse led everyone to believe--amounting to merely a few thousand people instead of tens or hundreds of thousands.  Given this much smaller world of actual depositors who had funded accounts of any significance and verifiability, then is it too big of a stretch to believe that the missing 648,000 BTC may have belonged to a very small and select group or perhaps even to only one person?