This isn't exactly what you were asking for, but here are some graphs from a different Go AI (Facebook's "darkforest") showing its estimated win probabilities: http://zhuanlan.zhihu.com/yuandong/20639694
Neat! My understanding is that AlphaGo is a stronger player, and so will likely have different estimations than the darkforest. But this is still really cool.
Looked up the strength (of dark forest) and it is claimed to be mid level amateur similar to crazystone and zen. So, the close measurements probably aren't too precise (when they are both around 50%), but the break away is probably about right.
First game, AlphaGo looked ambiguously weak until end of mid-game when it suddenly looked very strong.
Second game, AlphaGo looked obviously strong throughout.
So the probabilities are maybe not entirely wrong.
For reference I am 4k rating, somewhat below darkforest(1d amateur) and so very very far from 9dan pros that AlphaGo is beating. Usually, pro matches are hard to understand but the excellent commentary and abundant community analysis helps greatly.
