Thanks for the interesting discussion.

I agree that even in long-term scale only half of reasons for BTC price are "external" ones. The second half is caused by mass psychology machinery. One example is rally - positive feedback loop aka self-fulfilling prophesy: nobody sells because nobody sells. Now you gave us the second example of the same self-fulfilling prophecy: miners sell because miners sell.

I have to agree with you: currently it makes sense for miners to sell most of mined coins immediately after mining, to speedup capital turnaround. And it's better to sell OTC, to avoid slippage. But by the same reason, it would make sense for them to include a clause in the contract: "do not sell the purchased coins (below XXX$)". The clause would be very easy to control, thanks to the blockchain. Which makes the price suppression hypothesis a bit less likely one.

As for 2012, I still suspect the MtGox coins. 750K bitcoins is a huge amount, we can be 95% sure they were stolen and 80% sure most of them were sold in the following months. Such amount cannot be sold without a trace and I believe the trace is either the slide from $32 to $2, or the following one-year stagnation. Without this dumping the magnitude of the slide and the stagnation would be much smaller.
 
P.S. It would be interesting to see the moment when miners will switch to "don't sell" mode.