>>Are you also assuming your SSTO blows up 5% of the time?  I would.   
-PW
>
>Why?  You are assuming that ground controllers would 
>destroy a vehicle that is designed to be piloted back in
>the event of a problem?  Perhaps you were under the 
>impression that that 5% is due to explosive failure, rather
>than the decision to abort?
See my reply to Mark, I am saying that in an equal comparison, the simpler vehicle will be more reliable.  And I think that an assumption that any rocket can be immune to catastrophic failure (in the near term) is an obvious mistake.