One of our advisors wrote back to see if they understood that approach.
The odds of flipping a coin and getting heads 25 times in a row is roughly 1-in-33 million. So if we have 33 million flippers and 100 get 25 heads in a row, statistically that is very improbable. We can deduce that group of 100 is a combination of some lucky flippers, but also that some have a "flipping edge." We may not be able to say which is which, but as a group our 100 will still consistently provide an edge in future flip-offs.
Well, that is correct. In fact, if we were developing coin-flipping agents that would be as far as we would be able to go. However, we are in luck because our problem has an extra dimension, which makes it possible to filter-out some of the "lucky" Bots from our trading systems.
Determining Which are the Best Systems.
There are several ways to determine whether a trading system has a persistent edge. For example, we can look at the market returns during the trading period and compare and contrast that with our trading results.
This is significant because many systems have either a long or short bias. That means even if a system does not have an edge, it would be more likely to turn a profit when its bias is in alignment with the market.
We try to correct that bias using some math and statistical magic, in order to determine whether the system has a predictive edge.
It Is a Lot Simpler Than It Sounds.
Imagine a system that picks trades based on a roulette spin. Instead of numbers or colors, the wheel is filled with "Go Long" and "Go Short" selections. As long as the choices are balanced, the system is random. But what if the roulette wheel had more opportunities for "long" selections than "short" selections?
This random system would appear to be "in-phase" whenever the market is in an uptrend. But does it have an edge?
One Way To Calculate Whether You Have An Edge.
Let's say that you test a particular trading system on hourly bars of the S&P 500 Index from January 2000 until today.
The first thing you need is the total net profit of the system for all its trades.
The second thing you need to calculate is the percent of time it spent long and short during the test period.
Third, you need to generate a reasonably large population of completely random entries and exits with the same percentage of long/short time as your back-tested results (this step can be done many times to create a range of results).
Fourth, use statistical inference to calculate the average profit of these purely random entry tests for that same test period.
Finally, subtract that amount from the total back-tested net profit from the first step.
According to the law of large numbers, in the case of the "roulette" system illustrated above, correcting for bias this way, the P&L of random systems would end up close to zero … while systems with real predictive power would be left with significant residual profits after the bias correction.
While, the math isn't difficult … the process is still challenging because it takes significant resources to crunch that many numbers for hundreds of thousands of Bots.
The good thing about RAM, CPU cycles and disk space is that they keep getting cheaper and more powerful.
Some dogs run fast; other dogs do tricks … my 100 pound lab, Duke's best talent is his ability to remain immobile in almost any situation (some call it "laziness" … I prefer "calm" or "even-tempered").
At the other end of the spectrum is Boo, our 20 pound mix between a Beagle and a Boston terrier. Let's just say that he has higher metabolism than Duke.
When Boo hears a noise, he lets out an involuntary, barely audible, sound. His body becomes rigid and his head cocks. He becomes alert and carefully tunes his ears to their most sensitive setting, seeking any information that will help identify the "danger." And now he's waiting for the next intrusion.
Usually there are no other disturbances to follow, and the noise is cataloged and soon forgotten. His alertness level slowly subsides over the next 10-15 minutes or so. He'll go back to napping, a little more fitfully this time … and just a little bit on edge.
Things get a little more interesting when another noise surfaces shortly after the first one. What could once be dismissed now must be treated as a threat – and just to be safe, a threat of the highest order. Now the appropriate response is a series of barks, nervous glances, pacing, rushing off in the direction of the noise to investigate, and a barrage of barks meant to sound more menacing than the source of the noise. Who or what is it? How much harm can they cause? How grave is the threat?
Duke is a different story. Nothing upsets Duke more than Boo. The same noise that prompted Boo to become alert does nothing to Duke. However the first time that Boo makes noise, that makes Duke alert. And the second time Boo makes noise, Duke starts vocalizing, and the third time, well now both of them start yapping and when they both start yapping, then other dogs in the neighborhood start yapping.
I realized that the same thing happens in the market with people.
In the market, it is the second noise – and subsequent noises – that creates the equivalent of the Homeland Security "Red Alert." Once an elevated level of alertness has been established, it takes a long period of relative calm for it to subside. And when on "Red Alert" … any additional noises (for example, the responses of the other market participants) – big or small – will highlight, magnify and further validate the issue.
But, like in the story "The Boy Who Cried Wolf," even legitimate threats are ignored after too many false alarms (or prolonged periods of constant alert). So, bad news about the economy isn't as likely to get people excited after the past few weeks.
While Hillary Clinton deftly followed the wisdom of Napoleon: “Never interfere with the enemy when he is in the process of destroying himself.” It didn't work.
I believe Trump’s success is a direct result of a broader collective dissatisfaction and negative mood trend in America.
As unlikely as the outcome was, socioeconomics explains it well.
Last year, if you told me 2016 would be the year that: the Cubs would win the world series, a reality TV show host would be president, and Britain would leave the EU, I'd have assumed you were on drugs.
Even three months ago I would have been surprised.
Last week, MogIA, an AI, predicted that Donald Trump would win, making it 4 for 4. If you had been tracking the polls, that would have felt ridiculous, but computers can be more objective than you or me.
Trump's campaign was a polarizing affair, with massive backlash from liberals, and a mainstream media which almost completely missed the momentum Trump had.
There're massive parallels between the way the media reported this campaign, and what happened with Brexit. The media echo chamber was pro-Clinton and ended up ignoring the feelings of a large group of voters. In Britain it was the elderly … in America, it was working-class white people.
Trump won swing states like Pennsylvania and Michigan in large part due to those voters feeling ignored and attacked by Clinton. Coal and Gas are their livelihoods, and efforts towards clean renewable energy could leave them jobless.
Think of all the moving pieces that had to align for Trump to get this far. Look at the candidates in the Republican primary, look at what happened with the Democratic primary, and look at voter turnout:
Leading into the final days of the election, you can see a steep drop in markets. Uncertainty often has that effect on a market, but since Trump's election, there's been a rebound.
As well, Putin has expressed interest in restoring ties with the U.S. It will be interesting to see the effects of this on the geopolitical environment.
If there's anything to learn from this experience, it's that trying to time the market is dangerous, and that ignoring dissidents makes you out of touch and vulnerable.
On a lighter note, here's the President-elect, Donald Trump, wrestling at Wrestlemania 23. He's a regular Ronald Reagan.
