I'm not really sure what you're asking.
Armajaro trading arm bought something like 240,100 tonnes of cocoa for around £650 million, or ~£2,700 per tonne, in July 2010. They then sold the cocoa beans "later that year".
Unfortunately, July 2010 was the peak of the market (it's almost as if trying to buy 7% of global production drives prices up), and they were paying above the odds even then. And prices immediately tumbled off a cliff, dropping to a low of ~£1,8200 in November 2010 (it's almost as if trying to sell 7% of global production drives prices down!), before rallying briefly in 2011, then proceeding to briskly tumble to a low of ~£1,330 per tonne in 2012. (They've since recovered, but they've yet to break £2,000.)
So yeah, obviously Armajaro lost a bunch of money on that trade, precisely because of the volume they were working with. By actually taking delivery, they were relying on prices going up, but prices went down, right as they were needing to sell. Except that because they were buying and selling hundreds of millions of pounds of cocoa, they themselves were a leading cause of the markets moving against them. The same thing happened in other famous market corners, for example when the Hunt brothers tried corner the global silver market (and lost a staggering amount of money). It's no accident that there are no famous successful market corners.
Incidentally, I'm not sure what the graph you linked is meant to be, but here's a good graph of cocoa prices over the relevant period: http://www.indexmundi.com/commodities/?commodity=cocoa-beans...
> whether it's possible that the price was artificially lowered to sell on a lot of commodities for $1 and avoid tax? Does that seem at all possible?
Not following. If you're asking if it's possible if Anthony Ward sold a key part of his trading empire to a hated rival for $1 not because he'd lost a ton of money trading cocoa and the thing was basically worthless, but because he'd made a ton of money and it was actually hugely valuable? Then the answer is no, of course not.
> As a complete novice to the field, could you explain why it's not a good plan in general to seize large proportions of a commodity, for example in these two positions with an increasing price and a stable retail demand for the end products?
First, cocoa doesn't have a steadily increasing price, and while retail demand might be stable, the supply situation is very volatile due to political and economic instability in the growing regions. Cocoa prices were bouncing around like crazy, and while volatility is good for smart traders trying to make bets about price directions, it's bad for people trying to deal in the physical commodity, because what happens if you get stuck with 240 thousand tonnes of cocoa purchased when the price crashes?
Second, even if cocoa was a safe commodity with stable supply, demand, and price, you can't "seize large proportions of a commodity"; what you can do is start buying it. And the more you buy it, the less stable the price is going to be, because the demand is no longer stable due to some wannabe Bond villain is buying up huge amounts of it. The price will skyrocket. So when you ask:
> why it's not a good plan in general to seize large proportions of a commodity
Because you'll pay a large premium over the fundamental market price. Like, by definition. And buying anything for a large premium over the market price is never a good idea, unless you value it well over the market price. But Armajaro wasn't a producer; all they could do with the cocoa is sell it. Which, inevitably, they'll do at some discount to the market price. Nor would this trade have made sense if Armajaro had known that prices were going to spike due to an external shock; then they should have bought options, not actual physical cocoa.
Bottom line: Buying very large amounts of a physical commodity in order to profit from price movements is a terrible idea.
Edit: The largest purchase of cocoa ever was done in 1996 by...Anthony Ward, working for Phibro at the time. He bought 300,000 tonnes, but Phibro lost money on the deal when prices moved against him. Shocking! Although he is reported to have made money on his 2002 trade. Then again, the 2002 trade was only 5% of the market, or 200,000 tonnes, smaller than the other two. (The Guardian says 15%, but you can't trust the Guardian with numbers, and multiple other reports confirm the lower figure.) Or in other words, Ward has demonstrated the ability to make money on small trades, but lose them on bigger ones.