A Trading Riddle re: TP and SL Targets

[Jemook]

Ok, here's my spreadsheet. I've filled it in with AUDCHF data (yellow boxes). Very easy to see what the best TP and SL settings are - just look for the biggest number of WINS and then look on the same row for the SL used. In this example 40 TP/SL would have given us 20 wins if used for the past data.

Gary, can you think of any super duper calculations to do with the supplied data?

Jem

Count Pair TP SL Aim.xls

[dietcoke]

I recorded (for the last 10 days) the maximum pips price moved BELOW the daily pivot. Here they are:

0
32
76
2
32
43
10
24
3
5

Now based on these past results I want to find a number to use as my SL today. I want to make sure it's a number that gives me the best chance of my SL not getting hit. If I average these numbers i get '23' which would get hit 5/10 times if i had used that for the past 10 days. So that is definitely not the way to go about it. Here's what I've come up with:

Jem,

An interesting exercise. However, given the information you provided, there can only one correct answer.

"place the stop below the most adverse movement"

This will result in 100% profit. However, as we all know, it isn't that simple.

The question of what stop depends hugely on what you can expect in terms of profit:
eg.

Code: Select all

MAE	          MPE1	       MPE2	          MPE3
0	            80	              41	        0
32	            0	              41	       82
76	            0	              41	        0
2	            80	              41	       82
32	            10	              41	         0
43	            0	              41	        82
10	            80	              41	        0
24	            0	              41	        82
3	            80	              41	        0
5	            80	              41	        82
Totals
227	            410	      410	        410

I've shown three columns alongside your maximum adverse excursions showing potential maximum profitable excursions.

You can see almost immediately that the best stop based on the observed PA under those three scenarios would be very different even though they all have the same profit potential.

MPE1 would perform best with a small 11pt stop
MPE2 would perform best with a large 77pt stop
MPE3 would perform best with a medium 44pt stop

Naturally, reality is more complex than the examples I've posted but they serve to show that you cannot judge a SL without an overall context for the full tading strategy.

At a minimum, you need to know how far trades move into profit and what happens when neither target is hit

[dietcoke]

Just seen your spreadsheet

[Jemook]

Please see the attached spreadsheet in my last post DC, let me know if you still have the same concerns. I'm entering at the daily pivot and recording how far we move above the pivot in a day and how far we move below the pivot in a day. In this case the move above the pivot is profit and the move below is loss. I want to make sure my TP gets hit as often as possible and my SL get hit as least as possible.

I want the SL to be the lowest number possible while getting hit as least as possible.
I want the TP to be the highest number possible while still getting hit as often as possible

Based on the data in the spreadsheet that I recorded for 32 days I can see the optimum place to place my SL and TP is 40 pips. This way I'd get stopped 4 times but take profit 24 times for a total of 20 wins.

I'm not saying that I'm right at all! Just showing how I'm looking at this at the moment.

[Jemook]

LOL cool that you saw the spreadsheet - what are your thoughts now? I find this fascinating, can't wait to get to the bottom of it!

[dietcoke]

Do you just hold all trades until they hit one of the TP or SL?

[Jemook]

I sure do! I'd like to get this sorted first and then look at skewing the SL/TP for better risk:reward. Unless you already know something I don't (and you probably do!) hehe

[garyfritz]

(I'm in a training this morning so can't look closely at it yet, but here's a quick response:)

The spreadsheet is a good first approximation. A couple of comments:

* Your spreadsheet currently combines PT & SL figures to give a total combined P&L. But it's combining **equal** PT & SL, i.e. R:R = 1:1. It may be that e.g. a PT of 50 and a SL of 20 is your best answer.

* You can't just add together the SL losses and PT gains. Remember that you're sizing your positions to a fixed % risk. That means smaller SLs might produce more losses, but with small SLs you'll have larger position sizes, meaning your PT wins will be larger. We might be able to model that in the spreadsheet but the current one can't do it.

* I'm not sure this simple spreadsheet can accurately determine the actual best results, especially with smaller SL/PT values that are BOTH hit. It may require a full backtest to really get an accurate answer.

[Jemook]

Gary! Good to hear from you. My purposes at the moment are to create an indicator for us to use for 10.2/a or for anyone entering at the pivots. Eventually I'd like to add basket functionality too. It will use the TMA Slope Values to determine how many days to look back for analysis so it only uses trending daily bars. ATM the spreadsheet results look like the best way to determine future TP/SL levels for the indicator.

Right now I'm looking at 1:1 R:R @ 2% risk. I understand that there may be a better solution for this which is why I've put it out there to you guys. For now I don't mind 1:1 or in this example 40 pips TP/SL because my success rate is through the roof. At 40 pips TP/SL I'm getting stopped out 4 times and winning 24 times. That's pretty good for me so far. If you can show me another way of analysing the numbers supplied on the spreadsheet to win even more often than 6 wins to 1 loss then I'm all ears mate.

Again, thanks for helping me see the light. If you can help me see more light (is that even possible?) then it would be a great help. I'm all ears for you Gary! Would love to see you modify my spreadsheet or if anyone else wants a crack feel free.

Show me what is a better TP and SL than 40 pips for the data supplied, and why! That's essentially what I'm after

Cheers,
Jem

[Jemook]

Also, please ignore the profit/loss/combined profit & loss values. I've only got them in there to help me understand the results. What I'm interested in is making sure my TP gets hit the most and my SL gets hit the least (or the value that gives me the most wins!)