I try to use another model and for me my point still on the whole holds, even with a larger number of coins "sold yesterday" (but it means something slightly different, see below).
I played a bit with the 5-minute candles on Binance of yesterday. If you aggregate green candles as "buys" and red candles as "sells", then regarding volume during the 7 hours of the main dump (~9.2% down from 58500 to 53800 between 7PM and 1AM UTC) then the relation between the volume of red candles and green candles was slightly above 2:1.
So we could guess: If we assume the rest of the day the relation between buys and sells was in equilibrium (the market moved largely sideways before and after the dump in a "stair" pattern, and a quick look at the candles seems to confirm that), then we had a relation of 1.29 to 1. This means: For each coin bought, 1.29 would have been sold.
Or in other words: 56% of the coins traded this day were "sold", 44% were "bought". The total volume were 912000
BTC, and 56% of these are thus around 510000. This means something slightly different as in the other calculation, because it includes the coins sold by short-term traders. We can paraphrase it to: 110000 coins more were sold than those that were bought.
Let's now go back to the Goxcoins. And let's assume the absolute worst case: all 140000
BTC are sold a single day. This means about 8 billion worth of USD per day are added to the sell side.
Let's assume that all other conditions stay the same - no more dip fishers, no more panicking weak hands. If we assume that at equilibrium (sideways market) we have a volume of 30 billion or 526000 coins on weekdays (1:1 relation), this would drive the volume from this "equilibrium" to 38 billion.
Yes, this would cause a quite big drop probably, because the relation between buy and sell orders would be a bit higher than yesterday's 1.29:1, it would be close to 1.5:1. Thus, I'd expect a dump in the region of 15-20%.
But the scenario is quite unrealistic, because MtGox distribution has already begun but only a fraction of the coins are paid out each day.
So rather than the weak hands, why not blame the disappearing into the void of the "cheap coins' gang?
Even the most perfect equilibrium needs just one tiny push at the right moment and it turns into chaos.
Where you're right is that in certain conditions a single larger dump can trigger other dumps, mostly when there's too much leverage in the market. However, afaik liquidations in yesterday's dump were only worth 600 millon USD, which were about 1.2% of the total volume.
But it's exactly this market dynamic people need to be conscient about. These dumps would be the result of either excessive greed (liquidations) or excessive fear combined with shorting. It's not caused directly by the supply surplus but by mass psychology.
They didn't sell and we went to 53k out of sheer panic, what do you think will happen when they do?
Well today some were already distributed by MtGox ... and the price even recovered a little bit. The dump was earlier than the start of the BTC distribution, afaik. It came after the 47k coins were moved from the cold wallet.
That's exactly what I want to line out with this thread. If there is panic/FOMO, then the movements are probably much larger than as a consequence of a simple sequence of large orders. Of course this also works the other way around, that's why I'm actually less bullish than most about the ETF "inflows" and also about halving effects.