Adapting to an Algo-driven market

Adapting to an Algo-driven market


What do we mean by algo-driven markets?

From my studies of the algorithms that started dominating trading around 2008, I understand they have the following features:

  • Reactive to trades, not proactive, i.e., if no buying or selling, the algo is silent.
  • Automated, therefore consistent.
  • Rules-based, but the logic is not reversion to the mean logic that human investors apply to market prices. It is the opposite, i.e., if humans are buying, the algo amplifies the buying until the buying becomes exhausted. Same for selling.
  • Not artificial intelligence. The algo’s only intelligence is the rule that once a herd starts to run it continues in that direction until it becomes exhausted, so as the leader of the herd the algo front-runs emotion. After selling is exhausted, there will be net buying trades, in which case the algo soon reverses and front-runs buyers’ emotions in that direction. So, the algo is a scheme of pump and dump in both directions. Regrettably, it is also a scheme in which the stock exchanges pay for the algo order-flow because it generates more trading for their shareholders.

Regulators don’t understand that today’s algos work like the bucket shops of old except worse in that these operators encourage the client to act emotionally and hence make mistakes while their computer systems are programmed to take the opposite sides of trades (with less losses than gains) and then don’t even send most of their trades to be cleared. The owner of the algo only has to net out the buys and sells at the end of the day. So, 5.05 million shares of buying and 5.10 million of selling results in settling only 50,000 shares today (and only 10,000 if tomorrow’s trading resulted in a net buy of 40,000 shares). So, how is this different than a bucket shop?

Few people in the world have studied algorithms more than my associate Harp Singh. Frustrated at first, his insights of the past 12-18 months have started making him profits again. He called me this weekend to make a huge call on the market, based on a combination of what he believes is happening in the global economic picture (the precipice of high inflation caused by Trump’s misguided economic policies) along with the power of the algorithms that now control capital market prices.

Harp has been focused daily largely on the copper market, both the commodity and major Copperminers, Freeport-McMoRan (FCX), HudBay Minerals (HBM), First Quantum Minerals (FM.TO) and Lundin Mining LUN.TO).

During our conversation on Saturday he opined that I had not been wrong about recommending a buy trade in HudBay Minerals (HBM) back on July 5 — just early — because net selling at the time was still being pushed by the algorithms that now control market prices.

I wrote about HBM in that article:

Many of the important fundamentals we look at are strong. At the current price of US$5.60, PE (ttm) is just 6.5, dropping to just 5.1 with anticipated earnings for the next 12 months. EPS were up +26.2% Y/Y and are anticipated to grow +41.6% next year. Price-to-sales and price-to-book presently are just 0.89 and 0.68 respectively. Being impressed with these figures, we ran a Discounted Cash Flow analysis which shows a Fair Value of US$9.28 using 15% discount factor and 9% CAGR for EPS.

Since my trade recommendation, the price of HBM has plunged -35.5% in just 100 days. But, professional analysts still rate the stock extremely highly, the P/E is an astounding 4.17 and Forward P/E a minuscule 3.31 while earnings have grown +26.2% this year. Price-to-sales and price-to-book ratios today are just 0.62 and 0.43 respectively, which are ridiculously low.  RSI 7 is now down to a rock-bottom 10.6 on the Daily and 23 and 24 on the Weekly and Monthly. These are strong Accumulation Zone numbers based on RSI — although other technical indicators I use are not so positively indicative of a cyclic reversal at this point.

Obviously, the Copperminer industry stock outlook is highly dependent on Copper prices. The December 2018 Copper Futures price today is $2.67/pound, down from Friday’s $2.748/pound. The 2018 highs traded in the $3.25 range, although for the past three months the $2.67 level is also an average, or slightly below average, and there is no technical indication that 2018 highs are going to be exceeded or even close to being reached anytime soon. So, it’s fair to say that today’s Copper and Copperminer prices seem attractive only to long-term investors based on fundamentals and perhaps to reversion-to-the-mean long-term technical traders.

Before getting back to Harp’s market call on Copper and the Copperminers, I want to state a strong opinion that market controlling algorithms today have not only blown up the capital of Value investors, as widely discussed, but they have done the same to technical traders of most stocks in most sectors.

Market breadth has narrowed so sharply that S&P 500 gains have been made on the strength of a very few stocks. In fact, 303 of the S&P 500 are losers YTD. Three-quarters of the S&P 500 have either lost money or gained less than +7.0% YTD. However, by joining the buying herd in tech stocks like Apple, Microsoft, Amazon and Netflix plus some in the Pharmaceuticals (Merck and Pfizer) and Military-Industrials (Boeing), those investors made large gains.

The fact is that most stocks have been losers, driven down by the selling herd, exploited by algos. Copperminers, which should be beneficiaries of strong global economic growth, have been just one of the myriads of losers. However, a growing global economy absolutely needs copper.

My point is that with the impact of algos in the market, any stock or industry group that on balance is getting sold is being destroyed and I don’t care whether the tools used for trading are fundamentals- or technicals-based. On the other side, many stocks or industry groups that on balance are getting bought are being priced at seriously ridiculous and often totally insane values.

Of the top 12 S&P 500 performers in the market this year to date, that is of the ones that have earned a profit, the average P/E is about 75. Netflix is at 101 and Amazon at 149. Even their Forward P/E’s are 71 and 65 respectively. Reminds me of GE years ago into a bear market which traded straight up from a post-split $3 to $33 from 1995 into 2000, a time when analysts were on Financial TV telling the audience that the GE P/E of 60 was A-OK. Well, it wasn’t and just because today’s high flyers have P/E’s in the stratosphere does not make them investment worthy. Market prices can change very quickly. Algorithms will accentuate the changes. They will cause fast markets that quickly overwhelm human capacity to cope.

Investors and traders today have responded as such to algorithm dominated markets where prices have become a dice roll. They no longer want professional commentary or analysis. They simply want people like me to give buy or sell recommendations only. If we are right, the audience sticks. If not, the internet phenomenon has led to thousands of others who make predictions for them to turn to. The notion that ‘markets are marketing’ becomes an even more apt description.

To that end, I want to report that Harp Singh now opines that with follow-through this Monday and Tuesday from what he observed from last Wednesday through Friday, he anticipates the start of a six-month rally in Copper prices to commence this week from a current price of $2.67 to a $3.80-$4.10 high level.

Harp also provided me a six-month road map for HudBay Minerals (HBM on the NYSE) starting from US$3.61 soon to go to the $5.37-5.50 level, as he predicts, before backing down to the $5-$5.25 level, then up to about $7.00, back to $6.30-$6.50 before going to $8.50-$9 (to sell 1/3 initial position). He would buy the position again in the $7.50-$7.75 level and sell the 1/2 initial position up at around $10 before he anticipates it would drop to $7.75-$8 where he would buy the position back. Consolidation then would occur for 30-45 days before the final move to a high of $12.

Two roadmaps by a very aggressive investor

That’s quite a detailed road-map. And would be remarkably prescient if it turns out to be accurate.

In other words, Harp sees the market algo reversing this week as investor funds continue to come out of the favorites of the past few years and move into over-sold stocks and industry groups like in Copper. The algo, he believes, will create the same incredible gains as the market experienced losses over the past year in many stocks like HBM, FCX, FM.TO and LUN.TO.

In mid-January this year, HBM hit a high of $10.25. Today’s $3.61 represents a -64.8% loss. But if HBM were to recover to $10.25, the gain from here would be +184%. Harp’s projected gain, given a high of $12, represents a +232% gain.

Can the algo do that? Well, crazier things are happening. In the case of Fossil Group (FOSL), that billion-dollar market cap stock is up +203% YTD and that includes a drop of -34% from its high four months ago.

I have made my point about algorithms. Harp agrees. Today’s market controlling algos are stock exchange volume generators that accentuate the trading on the upside and the downside depending on the current mindset of human participants in the market. It’s now a game of deceit that regulators have allowed to happen and seem oblivious to.

As the expression goes, proof is in the pudding. So, let’s watch HBM and hope that Harp follows up with running commentary. Let’s also watch for similar moves in other major Copperminers, FCX, FM.TO and LUN.TO.

As for me, I am a long-term oriented investor who holds HBM in my portfolio and one who believes the stock is back in the Accumulation Zone and ready to be acquired. I am also adapting to algorithm uncertainty in my own way with new short-term based trading programs called TraderCadence being applied to industry groups that I see will continue to grow based on their assets and business plans.

Please read our website disclaimer with regard to this article and all commentary we publish.


Like to give your opinion? Join our Community Discussion