Join John Person LIVE at Back to the Futures!
Join John Person LIVE at Back to the Futures!
How to Use Algo Trading to Boost Performance
06/22/2017 2:53 am EST
Can you survive trading in today’s algo world? I believe you can, asserts John Person, a 36-year trading veteran who has written several books and is a featured speaker. Third part of a weekly series.
This is the third in a series of articles designed to help traders better understand what “algo” trading is and what the various inputs and factors are so that you have a comprehensive idea how to build these concepts into your trading routine to boost performance.
As promised, this week’s article will cover why we want to incorporate spreads in our trading models, from various market products as well as different timeframes.
Remember, in this advanced trading environment, diversification does not come from just trading different products; trading system models can be within the same product using different timeframes.
For example: one can have a swing trade or overnight trading model using a daily chart or using 60-minute time frames and a different system one can trade on a separate trading account counter trend trade signals based on 60-minute, 15-minute or 5-minute time frame charts.
Examine the chart below the daily chart is in a daily PPS buy signal (green arrow) while Persons Pivots are in a monthly Bullish buy mode as indicated by the higher projected resistance line in red and the higher support target in green than the preceding month's levels. Below we are using Trade Navigator charts from Genesis Software.
Each of the charts above shows the daily time frame on the left, the 60-minute in the center and the 15-minute chart on the right.
I highlighted June 16 as we were in a signal buy mode. Note on June 16 we had generated an algo sell signal on the 60-minute chart at 4 am and as the morning progressed, see the 15-minute time frame also gave us a PPS sell signal.
The two lower timeframe signals were in-synch which resulted in a nice counter-trend day trade against the daily trend.
An algo system that has predetermined criteria to enter trades, place stops and uses a predefined profit target, would be capable of taking this setup.
The next chart is from Trade Station which has the capacity to do just that. Examine the chart below where the entry to sell short was at 2433.75 at the 7:30 am (CT) open based on the sell signal and the predefined profit target was achieved right before the 8:45 bar close. While the market did move slightly lower from the exit at 2423.25, the nice feature was that the market was able to fill the order by trading lower than the exit target level allowing to generate a profit.
Spreads or pairs trading can be made with or against correlated markets as in the case with inverse Exchange Traded Funds. One popular example is the SPDR S&P 500 Trust ETF (SPY) versus the volatility ETN iPath S&P 500 VIX Short Term Fund (VXX) or the 3X leveraged ETN Invesco PowerShares (UVXY).
Moreover, hedge funds and others have long developed their own basket of stocks to outperform a sector or index, for example, the NASDAQ 100.
We have many stocks that have mirrored the index in their daily price moves.
Now, look at the move in Tractor Supply and Starbucks. Both of these names are in the QQQs but taken out the new hybrid basket of select stocks has even greater outperformance of the underlying index.
The next chart has several markets overlaid in the upper quadrant with the two main markets Gold and 30-Year Treasury Bonds in the middle with the On Balance Volume Indicator in the bottom quadrant.
What we wish to do is to sort out visually, the direct and indirect correlations in price direction of these markets and how they tend to interact with each other over time. The sample period is just over 30 months.
What we see is a strong correlation between Gold (candle chart) the Japanese Yen (futures black line) and Treasury Bond prices (magenta). Finally, the Emini S&P 500 is in green.
From approximately February 2016 through October of 2016, all of these markets moved in synch or tandem, especially the stock index SP 500.
However, notice where the disconnect occurred, right after the November elections. These markets have now departed from their respective historic trending characteristic.
Now under closer examination, you will see where Gold and Bonds remain in a tight correlation. This is why I separated them in the middle quadrant, the gold and green lines, represent Gold price and Bond prices and while they are not in-synch, “tick for tick,” they have been and continue to trend quite closely. The takeaway is that end of day traders and day traders can use this correlation as a spread or as a means of diversifying positions on trade signals.
No one really knows which day one will outperform the other; per se, will Bonds move up .75% when Gold only moves .20% in a single day?
The recent average true range in Bonds is 32/32nds or $1,000 notional value where Gold has a recent ATR of $14.00, or $1,400.00 per day. So one could assume Gold’s intraday moves may outperform those of Bonds.
Once again, that is not a guarantee and perhaps the bulk of the moves happened in the Asian or European sessions, which happens quite frequently.
So if they are correlated in price direction but not percentage changes, and maybe not during US open outcry sessions, if one has a systematic trading program that can define entries and stop loss level by taking positions in both markets, one may stand a better chance to increase performance.
In addition, if one market generates an intraday buy signal while the other doesn’t, it could mean there may be a delayed reaction and one could piggyback the signal in the laggard.
On the other hand, if there is a more concrete signal in one market and less risk to where the entry and stop location would be, it may be beneficial to take the trade in the less risky market.
Here is a great example of those exact concerns. Examine the chart below has the Bonds in the upper quadrant and Gold in the lower quadrant. Notice we had my low close Doji (LCD) bearish signal with a PPS sell indication as outlined in the red box on the chart.
Bonds gave a defined trigger to sell short but the signal was not as pronounced in Gold. The short would initiate at 155 20/32nds and the stop above the swing high at 155 26/32. Gold had a $170 risk and Bonds had a $187.50 risk. However, at the end of the trade note that Bonds only generated a profit of $187 while gold outperformed the sell signal based on the trade in Bonds with a $430 profit in the same timeframe and with less risk.
Correlated trades using an algorithm system can be a real game changer for the trader who is able to take trades in all markets. At least, a trader who employs a system that helps identify trend changes and that can identify entry levels, stop loss levels and profit targets.
Moreover, this method gives futures traders an advantage for intraday trading from the perspective of lower margin requirements and smaller equity balances which are required to day trade in equity-held accounts.
I hope you found this week’s commentary helpful in learning more about what goes into trading system models. Next week’s article will uncover a trading system using time counts on price bars to establish trend reversals as well as defined entry signals.