Focus is a completely different story, which conveniently wasn't the story being discussed here.
Back to the, so far all of your arguments seem to basically hinge around peak flow through the turbine housing, and how two small turbos have more flow capability than a single, larger unit. That point is completely irrelevant, as on a single, I have the option of using a larger turbine housing to reach the same equivalent flow capacity, and consequently achieve a similar TIP as whatever you get for your twin sizing. On the examples that have been discussed, I can go anywhere from the smallest T3 housing (.48 is the smallest I'm familiar with) to a T04 housing large enough that a 3.0L would be bouncing off the rev limiter to even spool it at all.
Further, your statement that you don't need to reach peak boost until 4500rpm, while it may be fine for you, is horribly flawed in assuming that is what everyone else would want. Personally I'm not a big fan of a dyno queen; I would much prefer a well balanced system that is useful under everyday conditions, in that it has a reasonable boost threshold, and doesn't choke itself off until beyond the useful rpm range. I'm very much an area under the curve guy, not a peak numbers guy.
A single turbo is inherently more efficient than a twin setup, all other things being equal, that is a fact.
IMHO, there are three key reasons to consider a twin turbo over a single.
1. Package reasons. On many V-type engine applications, a twin setup will allow much better header design. Unfortunately, the contour is not one of these applications, as there is barely enough room between the block and the dash panel for the existing manifold, let alone a turbo for the rear bank.
Fwiw, on my Cobra, I have chosen to run a twin turbo setup for package reasons (mostly anyway) as I can mount the turbos very close to the block on short manifolds, without relocating a large number of components. A single, while quite "doable" on my mustang requires a somewhat compromised header design due to the long assymetric crossover required, and requires relocation of several components that i dont want to move right now.
2. Turbo Sizing. In some cases a proper single size may not be readily available (or at all) This can be the case in many larger engine designs. This also somewhat applies to my own project, as I had several T3's "laying around" to use, and no appropriate T4 handy.
As a side note, my case is an excellent example of considering the entire system when sizing your turbos. In my case, some T3/T4 hybrids would have been an ideal choice for maximum power and efficiency that could be had out of a 302; but since I have no intention of even cracking the valve covers until I'm ready to build a completely new motor, the T3's I do have are capable of supporting enough power to split my block in half (its been done on several T3 assisted 5.0L engines that I know of) So steping up to T4 compressors was completely unecessary. Of course this is all aside fromt he fact that a T4 cold side won't fit between the block and the framerail on my car anyway (heck one commercial kit required a frame notch plate to fit a GT28 in there . . .)
3. This third point is very simple: Pimp factor. It sounds way cooler to say "my car is twin turbo" rather than "I have a single turbo on my car"
Sadly, this is truly the reason most people even mention or think of twin turbos "OMG, he's got TWO turbos, that must make waaaay more power than just one".
Travis, I will give you this, a good bit of your theory is correct (most of it actually, though there are a few things in your posts that I hope are just accidental mis-statements) but your application is flawed, especially in your comparisons.
