After the big move down from 2 decade highs a few days ago, gold rebounded for a couple of days and got back up from $792 to $820. This is fairly typical in the early stage of a potential price collapse, to have 2-4 day bounces after a major down day. However, to confirm the new downtrend, it is essential that the market then has another big down day quite soon after the pullback, and moves to a convincing new low. With today's price action in gold, I think we may be on the verge of seeing this. From a high of $819-820 overnight, the market has shown surprising weakness, falling significantly and now trading around the $800 level. The lows of this down move so far were made 2 days ago at $791.70. If the market today or tomorrow breaches that level decisively, moving to say $785 below, then I think it could trigger a precipitous decline in gold. Given that we have record gigantic speculative long positions in gold, according to the last COT report, there is the potential for huge waves of margin call and stop-loss selling. Every trend-following CTA, every gold bug, is long up the wazoo in gold, and a decline of historic proportions is the last thing they are expecting. When they all come to sell at once, there will be a severe deficit of buyers and the price will go down like a lead zeppelin. Targets to the downside would be next support levels at $730, and then the major level around $685-690 where gold first broke out 2 months ago, or even down to the prior range lows of $640-650. If we remember the last severe gold decline in 2006, that moved from around $730 down to $550ish, a 25% decline. A similar move here would result in a fall to around $635. As for risk control, there is a really nice close stop just above today's highs of $820. So you can risk $25 and potentially make $70-150 in the next 5-6 weeks on this trade. Alternatively, the out-the-money puts for Jan and Feb look pretty appealing right now as well. Market action that would prove me wrong would be a move to say above $825, or a failed attempt to break the low i.e. if it doesn't move below $785 within the next 2-3 days, then bounces hard to say $810 or higher, then it would look like a convincing support level and the trade would no longer be on. I would still have on a few gold puts, but on much smaller size, as the odds of the trading working will be much lower.