I am just conjecturing as I said, but trying to come up with a global explanation, given that the 2009 ducati with a carbon fibre integrated engine chassis looked a pretty good bike, as stoner consistently maintained, repeatedly saying that he rather than the bike was the problem when he wasn't doing so well in the early part of the season. He didn't seem to have any front end problems with this bike, and it looked the fastest bike at least in his hands at the end of the season.
My speculation is that they developed the original 2009 bike in aluminium, built or got ferrari or whoever to build a carbon fibre chassis which matched the characteristics of the aluminium chassis, then said great, they perform similarly. Whatever they did for the 2010 bike, which obviously involved going to the bigbang engine but may also have involved geometry changes, didn't work for whatever reason.
I agree with tom that ducati need to innovate to compete, but the innovations need to be ones that work, and as I also said radical changes like the chassis change would now seem harder to implement with the limited testing by the real race riders available. I have yet to see anything convincing as to why a carbon fibre chassis should be an advantage in a race bike (as opposed to a race car) apart from the ability to change things rapidly, but as I have been arguing this may be a mixed blessing. I have seen arguments from people like furosawa, as j4rno has re-posted, that there are reasons why the carbon fibre approach won't work.