IO – A Remedy For Overtrading

In developing a trading system, quantitative data analysis is key, but sometimes a picture is worth a thousand words. The TVO System equity curve (pictured below) came out ahead in 2015 with a 32% return at the end of the year. As you can see, though, the ride was a bit rough. The graph represents the two systems used at the time, TVO (Total Volume Oscillator) and HG (Heat Gauge).


Essentially what we see here is the effect that overtrading can have on a system. The takeaway is that while HG did indeed enhance the original TVO strategy just like in the backtests, the account drawdown was not so easy to swallow. Even though we made a complete recovery, it got me thinking that perhaps HG was sometimes just too much of a good thing.

During periods of high volatility, HG is known to generate a lot of signals close together. The signals are so close that if you take them all, often you will end up averaging into your initial position to the point that too much risk gets put on the table. Added risk is not always a bad thing as it can also lead to greater gains, but having too much based on the same methodology (in this case volume) can magnify the negative moves and lead to an uncomfortable seesaw effect, even if the end result is positive.

A new perspective was needed to put things in check and smooth out the curve. Working with volume and ignoring price data was the out-of-box thinking that originally led to TVO and then HG. Being tied only to volume, however, in many ways formed its own restrictive box. I began to examine other types of data with the same approach I took when crunching numbers for TVO… learn all you can about what everyone else is doing, then throw all that out the window and try to come up with your own method.

After combing through what was available, I honed in on the daily numbers of rising and falling individual stocks. Advancing and declining issues data, as it’s known, is widely studied and there are already many indicators based on it, but after applying variations of the original TVO algorithm, new and interesting patterns started to emerge. The result, I’d say, was an “I-Opener.”

The initial backtests of Issues Oscillator (IO) on the SPY ETF produced a win rate of 78% and an average annual return of 8% over a 15 year period. Here’s the historical equity curve compared to the S&P 500 (SPY).


In 2016, IO was added to the TVO System. The time frame is medium term (somewhere between TVO and HG), and gives us almost as many trades as HG. Because of this we were able to size down the frequency of HG signals and smooth out the curve. The three systems work together in a checks and balances sort of way. If TVO and HG are the Congress and the Executive Branch, then IO is like the Supreme Court, keeping the other two from getting too much out of hand.


One could argue that 2015 was simply a turbulent year for the market and difficult for any system to navigate through. On the other hand, 2016, with the Fed’s ongoing rate hike saga as well as brexit, certainly was no picnic as far as turbulence. Despite all that, the TVO System consistently performs well (The return for ’16 is currently about 12%). With IO now in the mix, we not only have a remedy for overtrading, but also a recipe to slowly and steadily build capital in the long run. -MD

To view past positions check out our Trade History.

To learn more about market volume, The TVO System, or How We Trade, find us on Twitter (@TradingLicks) and StockTwits!

Performance results on this website dated prior to September 2014 for TVO (prior to May 2015 for HG, and prior to May 2016 for IO), including backtesting and trade history, are simulated. Please read our full disclaimer.

What do you think? Leave a Reply