G.R.A.P.E.S.

Generic Real Time Any Position Exit Strategy

By: Joe Duffy

This excerpt is taken from Joe Duffy’s Ultimate Trading Robot

GRAPES is the acronym for Generic Realtime Any Position Exit Strategy. This is a simple framework for an exit strategy that is a combination of many things we have learned over the years testing many different systems and models. It includes profit objective, stop loss, and move the stop functionality. It is completely mechanical and has no user subjectivity. We will examine each part separately along with an explanation of why and how we use it. GRAPES is applicable to any strategy on any time frame.

GRAPES INITIAL STOP

The initial stop is set according to entry testing. When we find an entry “edge” we want to trade, we know from our testing where the stop is that makes it wrong and/ or hits the $ risk limit. That is the initial stop loss for GRAPES.

GRAPES PROFIT OBJECTIVE

The profit objective is 1.5 times the initial stop loss. Note that it is not 3 or 4 times. There is a good reason for this; how many times have you read that if a simple system is “only 50% accu­rate” with a risk reward of, for example, 3:1 the trader will make a boat load? This is a boat load, but unfortunately not a boat load of money. Setting objectives that far away will result in very, very small win percentages, very often so low as to render the overall system a loser.

Having said that, let us qualify. We do have mechanical trading systems where the risk to reward in terms of our best case Profit Objective is 4 to 1 or more. However, there are a couple reasons why we can do that.

First, the initial profit objective may be 4 to 1 or more, but only if that is just one of several exit strategies employed. The reason is very, very few trades will go on to meet a 4 to 1 or even 3 to 1 risk reward objective. This only occurs in those rare very strong moves. So a mechanical trading system must have other exits to deal with the majority of trades that will not meet the objective. As GRAPES is a simpler framework, we have not incorporated those other exits.

Second, we do have trading strategies where final risk reward is 4 to 1 or more, but that is not because the objectives are hit, but rather because risk is cut as the trade moves on. This is a very important factor and takes us to element three in the GRAPES strategy.

GRAPES MOVE THE STOP

Especially on breakout style trades, we see a lot of market action where after the breakout the price will test back into the range. Your strategy rules must allow for this. Hence, in GRAPES we leave the initial stop loss where it is until the market has shown two consecutive profitable closes relative to the entry. The only exception is if the first profitable close shows a substantial open profit via a wide range bar, then we can proceed to move the stop.

After the above conditions are met, our strategy testing gener­ally shows that at this point the “edge” provided by the entry has done its job. The trader hopefully will get more of a trend in the entry direction, but needs to prepare in case that does not occur.

At this point cut the stop loss amount by 50%. For example, if the dollar risk was $1000, then cut that risk to $500. Or in other words, move the stop halfway between the initial stop level and the entry price.

At this point, our risk reward now becomes 3 to 1 for this trade. If you have a verifiable trading edge the opportunity to move the stop should be met a good majority of the time. Remember we are only looking for a couple of profitable closes, not a specific risk reward, so this is quite realistic given a proven entry edge.

GRAPES TRAILING STOP

We do not recommend moving the stop loss again after the ini­tial move until the market has obtained at least 80-90% of the profit objective. When it has obtained at least 80-90% of the profit objective, at that point it is okay to move the stop loss to break even. Even though we put this trailing stop in, it is very rarely a factor and does not make much difference in results. The larger point is do not use breakeven stops until price is at least 80% of the way to the objective. In the very large majority of circumstances we have tested, ignoring this observation will erode your overall results.

OTHER EXITS

When using the GRAPES framework as part of a mechanical strategy the parameters presented here may not be optimum; use GRAPES as a guideline or starting point for your exit strategy.

The trader will also benefit by employing other exit strategies in addition to GRAPES. Following are some additions the trader might consider. These additions are an overview of exits we have used, though not all at the same time or in the same trading system.

We hope the logic will get you started thinking about what to test:

  1. If Bars Since Entry greater then “n” bars then adjust exits.
  2. If Profit in Trade greater or less then “z” then test adjust ex­its. Test both positive and negative integers for Profit in Trade.
  3. If Bars Since Entry is greater then “n” AND Profit or Loss in Trade greater or less then “z” then adjust exits.
  4. If a swing high or low of “n” degree has formed since the entry bar then adjust exits. “Degree” is the number of bars on both sides of the swing bar that must have higher lows or lower highs to label it as a swing high or low. All of the leading techni­cal analysis software has this “swing” function built in.
  5. If RSI (or specified indicator) reaches “x” level and this indicator reached “x” level within “y” bars ago then adjust exits. Similarly, if the difference between current RSI reading and RSI reading “n” bars ago is less then a threshold then adjust exits. These techniques attempt to take into account maturity of the trend.
  6. Stop-on-Close exits.
  7. Volatility Breakout exits.
  8. Reversal Signal exits.
  9. If “n” Consecutive occurrences of any condition, then adjust exit. An example may be RSI over 80 for “x” consecu­tive days. This could measure persistence of trend (overdone) or likelihood of trend persisting (more to come eventually). All technical analysis software has a “consecutive” function that al­lows you to test how many consecutive occurrences you have of any condition.
  10. If “n” period high or low or “x” period swing high or low is more then “y” average ranges or dollars away, then adjust the stop loss exit.
  11. If any specified condition occurs, then adjust exit “n” bars later. For example, if RSI is over 80 for “x” consecutive bars then exit “y” bars later. Or if RSI is over 80 for “x” consecutive bars do not exit at objective until at least “y” bars later; or use these conditions to recalibrate exits.
  12. Analyze the Minimum or Maximum risk or reward of a series of potential exits. For example, Profit Objectives based on Average True Range, Dollar Value, or “n” day break. All soft­ware has a “minimum” and “maximum” function to find these values from a list.