UPDATE: Silver Buys Filled [Includes Levels]
Cover pic: where we hope to be be sitting in 6 months
Algorithms
Just a reminder that algorithms are to be played just like any other inflexible trading style. The question is how.
(Pardon typos, as we are on road)
What
While we are taking no position of consequence, our group has over 50 years of collective experience in markets. Instead of hating algos, we've decided to
play them. More on that thesis below.
We are buying silver on spec with expectations of a 6 to 12 month hold if profitable and a 1-3 month hold if not.
(Click chart for live prices and charting.)
Why
A macro MA system we used back in active portfolio management to give bias for short term trades hit up on Aug 31.
Why Not Gold?
Same signal hit up in Gold, but Gold has already pulled away from the risk/ reward level.
How Expressed
The position will be expressed with no leverage. In this way, we can quantify our VaR without depending on typical stop losses. The position will be limited to buying liquid Silver coins and SLV.
What is the Var?
The capital used is 100% risk capital. On a monthly basis a decision will be made to keep or liquidate the position. If the position is liquidated, a new assessment to potentially go short using the same criteria will be used.
The monthly liquidation criteria will initially be MA based. If profits incur, the monthly decision to remain long or sell will be more volatility based.
Some Observations influencing the choice:
We acknowledge that Gold has been outperforming Silver on the upside and attribute part of this to Gold's remonetization process. This will likely continue, but it does not change our risk/ reward discipline. If it had not rallied last night, we may well have been putting this on in Gold.
For all we know, Silver will outperform at the end of a precious metals rally as it has in the past.
Further, Silver has the added potential of a bonadide short squeeze which, if it hit would cause backwardated spreads and reduce cost basis as we rolled our position back monthly.
Also: Gold has historically much more downside volatility than upside, meaning rallies are like escalators up, sell-offs are like elevators down. Numbers bear this out, but it is important to note Gold seems to be more "spikey" to the upside on events than in the past.
We acknowledge that the dynamic of gold is changing and spoofs lower may be fewer and further in between whilespoofs or algo stop runs may increase to the upside. But this is not new, and may be cyclical.
Silver, meanwhile has as much history of being squeezed higher (sometimes legitimately with player "assistance" ) as it has lower.
Expectations:
Monthly profits should average north of 1.5% of we are right and 0.5% if we are wrong. Past performance being what it is and all, no guarantees.
Why now?
Historically this system was used to facilitate top down bias for short term treading. It makes sense, just as it did during the time Securities Analysis was written, that when time frames have become so short in securities, and the trade is beginning to be crowded.. that one go the other way.
We are now seeing algos front run each other at night. To us, the trade is getting crowded and longer time frames should be used for our trades. Let volatility be opportunity and not risk.
That is not to say short term Vol based trades are done for us. In fact those are better now. When they are wrong, we see it much faster than in the past.
In this way, the algos are doing us a favor on reducing our hold times and giving us the ability to move on quicker to other opportunities. We view volatility trading more relevant than ever.
Aspiring Robot Killer
Silver is unique in that it gives the illusion of continuous liquidity. But it is more prone to liquidity gaps than other products. And it is during these liquidity gaps that the algos (human spoof replacements ) operate at their own risk in the liquidity roach motel of markets.
And algos don't account for exit liquidity. Humans do. And unless a human is scaling volumes properly on his algo, he will pay the price on the other side. Like pollsters recently learned, averages are meaningless when you have little data.
Just as all these predatory algorithms were derived from old floor trading manipulation schemes fueled up to "11 ", so now we must use their inflexible patterns against them.
Like a broker every day who comes in at noon to buy, it is the automation and operational patterns that are our first line of analysis against these tin can algos written by geniuses but monitored by janitors.
Other Trades Being Considered
As of Friday close:
-Sell NG 3.06
-Sell CVS at 77
Read more by Soren K.Group
