TechCrunch has relayed a new NPD report tracking “market share” in the fourth quarter of 2011. It shows a “market share” of 43% for iOS to 47% for Android. I put “market share” in quotes because it’s listed as “smartphone” market share, and it’s not clear whether or not iPod touch and iPad, or the various Android tablets, are (or should be) included.

What interests me much more is the by-model breakdown of sales according to NPD. There are currently three iPhone models available — iPhone 4S ($199), iPhone 4 ($99), and iPhone 3GS (free) — and they occupy the top three slots in that order. This may not be an entirely new phenomenon, particularly not for Apple products, but I still find it remarkable. It contradicts what most people keep insisting is conventional wisdom: that cheap and/or free models will always steal volume from premium models. Despite price and variety pressure from various Android models, and even pricing pressure from other iPhone models running the same exact software, the iPhone 4S was still the top-selling smartphone in the United States (according to NPD, at least). And this report is only for October and November — it doesn’t include Christmas.

You can’t explain this with the usual hand wave of “marketing.” If people are opening their wallets for Siri because of the latest commercials, for example, it doesn’t explain why the iPhone 4, without Siri and $99 more than the free 3GS, is #2. At face value, I wouldn’t think Siri is worth $200 to the average consumer, and I definitely wouldn’t think “speed and a Retina display” are worth $99 to them. Surely Apple Retail staffers are pushing the higher-end models, but I still think the average person will gravitate towards free, especially if nearly all the features are there. (Related: my parents now own an iPhone 4 and an iPhone 4S. I had very little to do with it.)

Even though the 3GS finished “last” among the iPhones, it still outscores any individual Android model, which is again remarkable to me. And it shows that despite what many Apple observers like to say, Apple does in fact care about market share — at least for now. I can’t imagine Apple looks forward to qualifying and performance testing new software releases on two-year-old hardware that isn’t significantly improving the bottom line, but that’s what’s happening. (At the moment, iOS 5.0.1  still performs very well on my old 3GS). More than one high-up person has decided that moving these units at any cost is a current priority. The impact on the user base, and the psychological impact of three iPhones on top instead of two, are apparently worth the trouble.