Expectancy is Next Key
In the last entry I suggested that you express all your trading results as R-multiples. So let's say you risk $1000 on each of ten trades and you have the following resuls:
-400=-0.4R
-2000 =-2.0R
-600 = -0.6R
-800 = -0.8R
+10000 = 10R
-400 = -0.4R
-700 - -0.7R
-1500 = -1.5R
-500 = -0.5R
+9000 = 9R
So are these good results? You have eight losers and only two winners? Would you want to trade the two system?
Well, one of the keys to evaluating the system is its expectancy which is the mean R-multiple of the results. So if we add them up you'll find that the eight losers sum to -6.9R while the two winners produce 19R. Thus, our total results are plus 12.1R. And if you divide that by ten trades, you get an expectancy of 1.21R. That means that this system, on average, looks like it will produce 1.21R per trades over many trades. And 1.21R is the expectancy of the system.
More on what you can do with that later.
Van
Comments
Dear Mr. Tharp,
Your July 4 entry is instructive, but does not hold together. You mention risking $1000 on each trade, yet you have two examples of losses of $2,000 and $1,500. Per the "rules" of your example this could not happen, you would have been stopped out at $1000 loss.
Posted by: alfred olbrycht | July 26, 2006 06:16 PM
Hi there
Just wanted to reply to your post. Losses of $2000 and $1500 are perfectly possible. They can happen due to trading errors or slippage or both. Unless you have stops are in the market but still slippage is possible. Hope this helps.
Posted by: Igor | July 27, 2006 04:08 AM
Thats Dr. Tharpe to you
Posted by: Justin L | August 1, 2006 07:30 PM