When driving, the traction battery provides all of the energy, including charging the 12V; you are not draining the 12V while driving despite the extra "widgets".
You are correct. DC-to-DC basically works like an alternator on an ICE car and it's on whenever you drive the car (and sometimes when 12V needs a top off). I incorrectly statined that the battery was draining during the drive, the 12V load simply subtracts from the DC-to-DC max capacity to charge the battery.
Also, 12V batteries were not designed for very thirsty vampire drain when parked. Calculations indicate that the vampire drain on the 12V is roughly equivalent to a 40 watt incandescent lamp running continuously. No ICE car has that sort of drain on the 12V when parked.
Not sure why you mentioned incandescent lap here, wouldn't any 40 Watt lamp be just as equivalent?
There is s wide variance in MTTF on Model S vehicles,
What does that mean? MTTF is mean time to failure, which is the an average time to failure over many vehicles - so all averaged vehicles share the same MTTF, there isn't a different MTTF per vehicle.
but the statistics indicate that early Model S vehicles had 12V battery replacements about every 12-18 months.
Do you have any statistics on that, or is it mostly anecdotal evidence? If anecdotal, I had a 2013 Model S60 which I kept 2 years before I sold it with the original battery without any problems. It spent most of its life with me connected to a Mobile Connector.
more recent vehicles have been reporting 3+ years MTTF.
Which more recent vehicles are you talking about here, which average failure time to 3+ years? If you took 2015 cars for example, the early ones could report failures almost 4 years (less 2 months), and the latest 2015's less than 3 years (since they haven't been alive for 3 years yet if in December 2015). Now, if you average a set of data with maximum ~3 years, you're not getting 3+ years MTTF. To get a reported MTTF to be 3+ years you'd probably have to be looking at 6+ year history, unless the newer cars you are looking at all failed at 3 years time.
But that does not avoid the fact that the cycle based wear out of even the more recent better deep cycle AGM batteries will eventually bring their demise.
Do you have data backing this? Any studies published, any specifications from battery manufacturers, or any statistically meaningful data proving this?
While I can't state with certainty what the impact of frequently topping off a car battery is, your explanation doesn't really hold up to even basic scrutiny. Tesla bumper-to-bumper warranty is 4 years / 50K miles. Even if the battery dies within that time, you get a new battery for free, instead of spending time and money setting up a battery tender. Even for older cars, what do you think is the return on investment for using a battery tender vs. just keeping the car plugged in (which you probably want to do anyways to keep the large battery happy)?