I can't help but to be unsurprised. Instances of Apple not exceeding predictions are few and far between.
Confirmation bias, really. Most of the stuff not predicted on is fairly hum-drum. (Nobody's excited about cinema display forecasts.)
However, I'll admit to laughing at Morris and Gruber when they estimated 8, 9 million as incredibly overblow predictions. (Both my parents and I will be buying one come revision 2.)