NOTICE: Undisclosed across-the-board price increases will be put into effect Friday, February 15th at 11:59PM Central Time. This includes Dakota end-of-day, Dakota R/T, BioComp Profit, add-ins and Workshops.

Grab what you need now. Order On-Line >

Difficult Times for Optimized Traders
For some traders, down markets are really difficult times. Obviously, if you can only trade long, it is very hard to snag those occasional up-days. But even traders of shortable securities struggle, particularly with "traditional" optimized trading systems. Over-optimized systems, even though they look good, are worse than worthless. They are dangerous. Not only are they brittle, but mislead you into over confidence.

These Markets Cause Optimization to Fail
An optimized system finds *the* set of parameters that work best over some past window of time. When the markets turn, it is difficult to optimize because the amount of down-biased prices is a small window and if you go further back, the last few years have been up. If you go further back, the market behavior may not apply to tomorrow's movements. Optimization rarely succeeds in conditions like these. I know, I used to try.

Adaptation is Superior
Instead of trying to optimize, you need adaptive technologies that automatically adjust to changing conditions. You need a trading technology that automatically spans robust regions of profitable trading system parameters rather than perched on a singular precarious peak. We created Dakota's collective-adaptive swarm technologies to do just this.

Collections of Adaptive Systems
Dakota uses collections of trading systems together as a system, rather than just one, spanning a region of system parameters. This, in itself, is a key feature. By combining their signals you increase robustness, increase equity, reduce drawdowns and get smoothing with zero lag. Zero. Second, these parameters can be adaptive so they change as time steps forward, bar by bar, tracking and adjusting, whether the markets go up or down.

With Dakota and shortable securities, the bear is your friend. Bring it on.

On major US indexes, such as the S&P500 futures, SPY and the Russell 2000, counter-trading works very well. Most of our customers are successfully counter-trading these and securities like them right now.

Counter-trading is selling short some distance into an upward run and buying long some distance into a downward run. I theorize that major market indexes have built-in "reversion to the mean" characteristics, that if the price has been rising, there is an increasing likelihood it will turn down (and vise versa). So, the higher they go, the more likely they will revert and fall. This works very well, if done right.

The "Double Smoothed" family of trading systems including the Stochastic, Entropy and now the successful ZScoreXMA operate this way and have proven themselves to be effective. A majority of people trading with Dakota are using these. Dakota users can download these trading systems ("bots") from our forums.

Research Discovery
I just discovered while doing research in our labs that counter-trading the counter-traders is even more effective. Let me explain...

If you have a collection of good quality counter-trading systems, and you measure their equity performance over the last 50 bars, if you take the signals only from the unprofitable ones you will make more money than if you bet on only the profitable ones. Interesting. You will make about the same money as if you took the average of all their signals, but you will be out of the market more, reducing your exposure, for the same gains.

My theory is a good quality counter-trading system also has a mean reversion characteristic on equity performance, that if it is a good system and in a drawdown, it will out-perform the others as it climbs to the mean equity performance again. The concept is like buying the stock of a good company that is presently "beat up" in the markets. It will revert back.

The Dakota multi-run picture below shows a "Monte Carlo" test of 30 equity curves of a "swarm" of 25 Double-Smoothed Entropy trading "bots". They are trading the Russell futures for the last 5 years. By taking my signals from the momentarily unprofitable bots I'm out of the market more and yet obtain an exponential equity curve even without compounding. I think this is because I'm actually compounding the mean-reversion characteristics.

If you have no idea what I'm talking about, just do this:

1. Get Dakota version 2.1.18a or higher
2. Download the Double Smoothed Entropy Bot
3. Select the Russell futures
4. Set Dakota to build signals using "Below Criterion" with a criterion of zero (0)
5. Push Go.

I just did. This is what I got:

About 98% straight equity curve, about 2.5x buy and hold in a bull run. Out of the market about 30% of the time. 13% of perfect trading. A reasonable equity curve that continues making money equally in both up and down markets with little drawdown on trend reversals. Hypothetical of course, but not optimized... adapted.

And like I said in the 2nd Section above, notice the equity curve continues UP even during the market decline. Priceless.

Easy. This one is a keeper.

Thanks!

Dakota is successful. Presently there are some millions of $ being managed using signals generated by Dakota End-of-Day. Soon some number of billions of dollars may be under advisory based on Dakota signals. We will be making some changes to accomodate high volume production operations and we are considering creating a commercial license.

We would like to open this idea for comment. Send your thoughts to sales@biocompsystems.com, particularly if you manage funds using Dakota.

Amsterdam... Learn from the creator of BioComp Dakota how to code bots, stops, equity engines.

Limited seats remain...

Learn More >>
Reserve Today >>