How Keybot the Quant™ Works

For obvious reasons, the secret sauce behind the program is under lock and key, but hopefully an explanation of some of the nuances of the algorithm will be helpful. This algorithm was in development for over one decade (late 1990's through the 2000's) and went live in 2008 (as fate would have it during the 2008-2009 crash; Keybot the Quant™ performed fine).

The algo employs a unique price-time relationship.  Keybot the Quant™ is a price model in its simplest form.  Algorithmic rules as well as economic data indicators and other parameters dictate the algo's direction. The robot is either positioned long or short the stock market there is no middle ground. Keybot the Quant™ is an oscillator at its core so it could care less if stocks are going up or down; it simply wants to lock-on that direction. The quant also has a quirk factor to it which is purposeful. The program would be difficult for others to duplicate thus it should function for years and decades forward; it has already operated for over a decade producing profits in every year.

The algo prints a number for key economic data points and also every Sunday. The Sunday number can be posted anytime Sunday and the reference to 7 PM EST remains on the printout for all Sunday numbers. The algo also prints numbers for month-end (EOM). These functions are used to match-up the Keybot the Quant™ data to historical S&P 500 data.

The main algo number that is shown in the printout is a positive or negative number generated by the Keystone Oscillator™.  The signal line number is a smoothing mechanism.  When the algo number crosses above the signal line, Keybot the Quant™ is itching to go long and when the algo number drops below the signal line, the quant is itching to go short.

The signal cross is the first step in a change of trend move for Keybot. The algo must also satisfy numerous programming rules before the signal change is accepted and the quant actually flips sides committing to the new position (the secret sauce). This is why the algo and signal line numbers sometime cross telling the algorithm to flip position but the algorithm rules take precedence and may prevent the direction change.  Once the algorithm rules are satisfied along with the signal cross, then Keybot the Quant™ will commit to the new trend.

Keybot the Quant™ cycles in and out of long and short ETF's. The double-leveraged ETF's are favored such as SSO for long positions and SDS for short positions, however, if a whipsaw move occurs where the algo flips sides but reverses within 10 hours of trading time, the algo will drop into a mode of using single ETF's such as SPY (long) and SH (short) for 35 days. If there are no other whipsaw's after 35 days, the algo will then return to the double-leveraged ETF mode.

Concerning the printout appearing in the posts, the ‘Benchmark SPX’ is simply the percentage gain, or loss, of the SPX index year to date. The ‘Keybot algo this trade’ number shows the percentage gain or loss for the algorithm program for the last trading move. The cumulative ‘Keybot algo for (year)’ number is the percentage gain or loss year to date for the algorithm computer program. Therefore, the ‘Benchmark SPX’ is compared to the ‘Keybot algo for (year)’ number to determine how the algorithm computer model is stacking up year to date against the SPX.

The ‘Actual results this trade’ number is the percentage gain or loss based on the actual trading and takes into account commissions and other mechanical price entry and exit costs and inefficiencies. The ‘Actual results for (year)’ number shows the actual cumulative percentage gain or loss year to date using the ETF’s. The ‘Actual results for (year)’ number directly represents the success or failure of the algorithm in real trading terms (this is the actual net gain and loss in real dollars using the algorithm). The algo has printed consistent and consecutive winning years every year since it went live in late 2008 during the stock market crash.


Please Read the Terms of Service.

© 2011-2021. Keybot the Quant™. All Rights Reserved.