A ratio you never see is the price of gold vs. the dollar index. Probably because the price of gold itself is thought to represent dollar strength (or weakness), so by dividing the price of gold by the dollar index, you'd appear to just be amplifying any up and down movements in the price of gold. Moreover, the natural ratio of gold nuggets to baskets of foreign currency in the earth's crust is, as of yet, unknown.
On the other hand, gold is far from a perfect proxy for dollar strength. For example, lately, the dollar has been rallying vs. other currencies, but gold has been holding its own. If someone were to tell you that gold will soon be at $2000, but the dollar index will be back at 100 (e.g. because of a Euro collapse), then holding dollars and holding gold would be equally profitable (+33%). So, the decision to trade in your federal reserve notes for gold should be based (to some extent) on the ratio between the predicted performance of the two.
Well, look no further. Let's start with silver, because it makes for a more compelling long term chart. One way I've begun to read charts is by first drawing a regression line to fit long-term data, and then creating new lines with the same slope to look for significant-appearing boundary points. That approach seems less subjective to me, for if we draw channels and trend lines based solely on local maxima and minima, we may be over-valuing outliers. Moreover, if you believe that the PM markets are pervasively manipulated, as I do (to quote Jim Rickards: "they don't consider it manipulation, they consider it their job"), then a regression line will capture overall market trends in a way robust to short-term hijinx.
Using this approach with silver, we appear to be at an important crossroads. On the 8 year chart below, we see that the $SILVER:$USD ratio bounced off the blue line twice before breaking support in 2008; it then took 2+ years to get back to the blue line, wherein it became resistance (Nov, Dec, & Jan 2011). The ratio then finally cracked through in February of this year, and bounced off once (during the May massacre). Now, it's right about there again. I'm currently triple long dollars (UUPT) and double long gold (DGP). If we break through the blue line, I'd say it's time to sell silver and gold, and buy dollars. And vice versa, if the blue line proves again to be support.
The $GOLD:$USD ratio is at a similar crossroads. Unlike silver, this ratio doesn't exhibit any steady long term trend. However, the three-year chart is an obvious rising wedge, cleared in April. The top line has been support once (May massacre), and it's back there again. Note I have included Bollinger bands instead of standard envelopes to account for the increased volatility owing to covariation of gold and dollar strength (15 day, 2 sigma parameters capture all the points quite nicely). Note that we are also right at the lower Bollinger band, which in the past has been a bullish sign for gold.
So, let's see what happens over the next few days.
*I feel compelled to inform our readers globally (as Eric "Leading the Witness" King might say) that my market musings are merely those of a dude thinking out loud, who's basically waiting for online poker to become legal again so he can return to his preferred method of gambling. In that spirit, please draw your own conclusions, and thanks in advance for sharing them with me.