Paying customers (app revenue) and eyeballs on screens (add revenue) are the metrics that really matter to a platform's success - not specific device model sales. Operating systems draw app makers, and apps ultimately determine success of a platform. When the Palm Pre was released, the hardware and OS were considered to be as good or better than the iPhone at the time. Palm OS is still being copied today (everyone, including iOS, now uses its multitasking UI). Nobody bought it because of the app ecosystem.
Apple made the choice not to license iOS to other handset makers, so they have to take the good (less handset options = less fragmentation) with the bad (less handset options = lower market share). Apple still does very well because the app and add revenue historically has been the highest. Those gaps are narrowing now. If/when that equation swings in Android's favor, you could see the market share numbers suffer even more. It's a tide that Apple needs to address.
If Android needs 85% market share to finally equal the app and advertising revenue of iOS at 10%... that's not exactly a glowing recommendation for Android. It just shows that a lot of people are using Android... but they spend very little money on average.
Developers know this... and that's why they're not swayed by Android's phenomenal 85% market share.
I think you'll still see developers focus solely on iPhone/iPad even when it's a tiny percentage of the mobile market. (it's happening already)
Hell... it happened with the Mac years ago. There are software developers who only make software for the Mac... even though Macs represent 5% of the computer market.
Sometimes it's just not worth it to spend extra resources to support the biggest platform... when instead you should be focused on the most profitable platform.