Fanciful platform predictions

As a powerful member of the blogosphere, I got an interesting emailed press release Wednesday predicting the 2010 “North American” smartphone market share:

	2009	Share	2010	Share
BlackBerry	23.2M	49.2%	28.0M	43.0%
iPhone	10.9M	23.1%	13.8M	21.3%
Android	4.6M	9.7%	12.3M	18.9%
Microsoft	4.8M	10.2%	4.7M	7.2%
Palm	1.4M	3.0%	3.1M	4.7%
Symbian	1.5M	3.2%	2.1M	3.3%
Others	0.8M	1.7%	1.0M	1.6%
Total	47.2M	100.0%	65.1M	100.0%

The forecast is by Canalys, the London-based mobile phone consultancy. Of course, as for other mobile phone market share estimates, it’s for new sales rather than installed base.

To summarize their forecast: the market is growing rapidly at nearly 38%, but Android’s sales and share will explode while RIM and Apple will lag Android and the market. Android is expected to match the (current) BlackBerry numbers by 2013. (Other numbers are intriguing: For Palm, was 2009 awful or is the broader distribution of webOS going to save Palm? The “other” residual sets a cap on the share for LiMo, Bada and other misc. Linux versions.)

This is not the only prediction for rapid growth by Android in 2010. A Goldman Sach prediction for 2010 global share has BlackBerry and iPhone flat, while Android rises to 12%.

My question: how do you predict market share without knowing the handsets to go with it? Many (including me) predicted the Nexus One would be a great hit, but it (by one estimate) sold only 174,000 units in its first 10½ weeks — 1/6 the rate of a Droid or iPhone. Perhaps it’s because it was only T-Mobile (before AT&T, Verizon and Sprint), or perhaps it’s the expensive unsubsidized price.

Still, like the AppleTV and Windows Vista, major companies do introduce products that turn out to be duds. To me, it seems like a 167% year-on-year growth — going from half the size of the iPhone to nearly the same — requires more than just a proliferation of models. It also relies on some big hits, like the Droid. And it probably relies on smartphones being sold without data plans, which I think is coming but not for another 2-3 years.

I can’t speak to share, but I think Android will do well to crack 10 million “North American” handsets this year. (How many of these in Canada? Obviously less than 10%).

Even at these levels, Android would pass iPhone here in 2011. But I’ll never again underestimate Steve Jobs’ ability to pull a rabbit out of the hat, so it’s conceivable that a 27% iPhone growth is low — particularly if the iPhone makes it beyond AT&T. If Steve doesn’t find that next rabbit, then that would be bad for the AAPL growth multiple — its P/E is around 21, above HPQ & MSFT but behind INTC & GOOG.