The twin incident stats would be even worse had it not been for the fact that most
enroute engine failures (which should mostly be a non-event) are bound to be unreported.
Normally it's TBM salesmen or single engine fanboys that spout this crap. We can manipulate stats all we like. One could argue that any given piston engine, well maintained, has the same chance of failure. If one does go I'd rather have a spare one on the other side.
I am no TBM salesman but I think you have a basic misunderstanding of stats. It is a fact that the fatality rate (or any other incident rate that you care to pick) of SE turboprops is around several times lower than that of piston twins - regardless of how the piston twins are operated.
Emotionally, the warm feeling from a 2nd engine is fully understandable, but the numbers do not support it.
If one went on the basis of a concurrent failure of two unconnected systems being extremely improbable (which it certainly is) then twins of any sort should never go down as a result of loss of propulsion.
Yet they do.
If one takes out fuel system mismanagement (which is arguably 100% pilot error, no matter how confusing some of them are) you are still left with plenty of ways to go down.
And if you are crossing terrain which rises above the SE ceiling of the twin (for a significant distance) then you have a 2x (min.) chance of crashing into it, than with a single. Admittedly that is an unusual case in Europe (the Alps take only ~ 45 mins to cross, N-S) but twin owners do like to use mountains as a common example of twin advantages.