Is this tradeable and how to optimise???

After back testing a particular strategy, i got results like this

SL is constant 2%

When i have Profit as 1%(RR=2:1), then i get
Winners=56
SL hits=17

Profit 2% (RR=1:1)
Winners=39
SL hits=34

Profit 3% (RR=1:1.5)
Winners=35
SL hits=38

Profit 4% (RR=1:2)
Winners=31
SL hits=42

Profit 5% (RR=1:2.5)
Winners=26
SL hits=47

Profit 6% (RR=1:3)
Winners=24
SL hits=49


The results are obtained as if we trade each of those separately.

Seniors, plz advise whether this strat is tradeable. If yes what wud be the % of qty to be booked in each of these pre-selected return %s such as 1% 2% etc.?

for eg: book 25% of qty in 1% level, 20% in 2% level, 20% in 3% level....and once T1 is achieved then move TSL to this level etc

dont want to confuse myself by reading those high funda books as suggested in some other threads. I tried couple of them earlier...for my edu background, those seems way over my head

Wud appreaciate a straight forward simple answer

Hope some good minded people helps me

Some how Profit 3% and 6% looks better than others.

would certainly need other values of result of backtesting before deciding if this is tradable.

I would have certainly recommended you 1 or 2 books (not very high funda), but you have ruled out the option.

So my simple Straight fwd answers (i am committing a crime by judging system only by those parameters, but only since you asked for).

No I would not trade this strategy.

(have seen lot better strategies and most of the systems I am trading have better RRR)

If you use amibroker, see its tutorial on how to optimize.

May be if not books, I request you to read some of my posts at this thread
http://www.traderji.com/tradestation/51967-strategy-performance-report-should-system-traded.html
(don't read the books recommended there, but atleast understand that there are other so many factors to be taken into account).

sorry the simple straight fwd ans took so many lines.
I hope I have not discouraged you (only trying to prevent you from a massacre), if so my apologies.

happy trading

Thank u for replying

I had seen that thread earlier and had a very good insight into it. Understand that there are so many parameters to look into before judging a strategy. There are so many things which i am unable to get through.

Another problem is that i cud not be able to code those in my way. I lack competancy since i dont code in Ami but in VB. And also a non-mathematics/non-programming background makes it more difficult.

Coming back to the problem, i had tested it again using a simple filter and the results are given below...when u find time, request u to chip in with ur readings...thanks

Profit 2% (RR=1:1)
Winners=27
SL hits=17

Profit 3% (RR=1:1.5)
Winners=25
SL hits=19

Profit 4% (RR=1:2)
Winners=24
SL hits=20

Profit 5% (RR=1:2.5)
Winners=20
SL hits=24

Profit 6% (RR=1:3)
Winners=19
SL hits=25

So, what is your real, not hided question ?

probably there are so many hidden questions in this forum which might have prompted u to ask me this question

Dan, Pickup the right question

See the only problem here is you have by default taken 44 trades in all categories. Now I am sure you know that for the System 1 (R:R::1:1), it will trade more frequently than System 5 (R:R::1:3). So you have to take into account the compounding effect of system 1 will be higher over system 2.

Without the actual compounding time it is anyones guess, but if you grant me the leverage to say that "Time taken to close the trade is proportional to the Reward" (in Practice it will never be linear relationship, very complex forumalation which is worthless to figure it out), then in that case
System 1 will is very very profitable compared to any other system.

alternatively if you only penalise the amount of RR for Win% then system 3 and 4 perform very poorly while rest are around same.

This takes into account that you have factored the slippage and trading cost ? (if not then system 1 may turn out to be performing not better than system 6).
Also note that with smaller holding period there is a trade-off, it will have shorter drawdown thus better UPI and Sharpe ratio but lower CAR and cost of trading increases. Aim should be to earn as much as possible with as less trades as possible. (this directly doesn't mean higher TF are better, they come with their own problem like large DD and risk of any unforeseen events (like major news)).

Please dont get me wrong if my reply haven't fully answered your doubt yet.
if you can please post a complete report. Will be more than glad to have a look and we can discuss the merits/demerits. this is how we all learn.

cheers and happy trading.

Thanks, i was waiting for ur reply and glad that i got one now. After couple of days, i thot u wudnt reply at all...

To clarify, these are all same system/strategy. Only thing is that i close the position at the particular RR. Meaning that the entry in all the cases are same and the exit (full qty) is according to the RR. So sys1 will not trade more than sys5 etc.

I have taken into account the slippage(both in entry and in SL) and even brokerages.

Since i have hard coded it, i dont have any readymade report generated. I only come out with this output as RR%s

sorry, my bad. Generally I am little busy with lots of strategies development going on. I may miss to check the thread but if you PM me I will surely reply.

May be my point wasn't clear. I hope you understand that success/failure of a strategy is a function of time and money, where time, I believe, plays critical role. If you are trying to compare two or more variations of a system and basing it on absolute return (Net Profit) then you are doing injustice to the profession of making money. I say so because you have fixed the number of trades (44) in each system and then you are comparing the return (risk and reward) of both.

System 1 would have certainly completed the 44 trades much much before system 5 or for that matter before any other system (this going by the probability distribution and it need not be always, but at-least we can be statistically confident, pun not intentional).
So if you fix time, then System 1 will have more trades than system 5
Or if you fix trades then System 1 will have completed in shorter time period.
Now I am sure you are aware of the power of compounding, which makes the difference here.

I am basing my argument with one assumption that you know why system with lower risk/reward will complete trade in shorter time, it is the drunk man scenario[/B where one whose home is nearer will reach sooner (atleast most of the times). Trades with larger Profit target will take time to reach the target level (ofcourse in both the cases the absolute failure trades (i.e trades hitting the SL in all the cases) will take same time but it is other trades which makes the difference).

new-to-markets said:

Since i have hard coded it, i dont have any readymade report generated. I only come out with this output as RR%s

Click to expand...

Which Platform do you use ? May I suggest Amibroker or TradeStation and if doing forex Strategy Trader by FXCM (you can get the s/w if you open a free demo a/c).

yup...got ur point now

btw, in ur opinion, which are the other parameters which are absolutely very very important before deciding upon a strat, apart from NP. I read abt number of continuous loss making trades in one of other posts.

I have confusions interms of:
-If the same strat is used in other stocks, they give very much varied results
-We are not sure that this result will work out in the future (regret for layman-ishness)

i have coded in VB...i find it a bit more difficult to code in Ami bcos of C based sintax, which i never have learnt

Let me know if u require any other parameter, which u feel important, then i will try to code it in

You said apart from NP. Funny that, I don't see NP. Atleast CAR always above NP, and still CAR/MDD is more critical than CAR for me. Apart from CAR/MDD, I see UPI and Recovery Factor. (ofcourse other 20 more parameters, but these are first confirmation metrics.

No one is layman here , or should I say everyone is layman here (if you prefer glass half empty).

On serious note, yes all these parameters give different results on different stocks or portfolio or systems. and this is where the discretionary comes into play. You didnt' think that strategy developement and system trading would be all Logic and formula based. (I know how programmers like to make decision tree for everything, hey I am myself a programmer. Hate making discretionary decisions)
So once you make bad decisions you will have experience, and once you have experience you will make good decisions, regarding which parameters to take (it all comes down to your personality , comfort zone, understanding and last but not the least, expectations from the market)

You are trying to fight an impossible battle, or should I say metaphorically, trying to fit the snake in the can. I am myself a VB junkie, all my programs I had made in VB (infact not even .Net but VB 6.0). But sooner you switch over better for you. It is never too late to learn !!
There are only limited things you will be able to achieve by using VB, ofcourse by no means I am undermining the programming capability of you or the VB language but specialsed languages exist for a reason. By trying to reinvent the wheel I think you are doing unjustice with your time.

Read Howard Bandy's book on Amibroker , very good learning place and starts from scratch and takes you far enough to code all your need.