Hi Nina,
Just to continue from my last post. So many people here including me are very happy with the entry and yet so many diverse opinions on exits. I am sure lots of folks here are interested in the exits too. Without a proper exit strategy this system is only "half baked".

I had a cursory glance on the charts over last 2 months and I dont see a definite pattern to exploit as an exit. Sometimes it can trend for 2-3 days for about 200+pips (talking about cable) and most days it just crisscrosses about the 50EMA. Keeping aside hypothetical trades, we need to find a balance between those long trends and range bound days which we will loose the pips.
I simply dont think that over a long run trying for a 15-20pips for this system for a S/L of 35pips is going to work becoz of the R/R ratio.

You cannot take 1 step forward and 2 steps backward. You just arent getting anywhere.

As I suggested trying for a 70pips TP for a 35pip S/L is good. However maybe someone with programming skills can backtest what are the optimum exits.
70 pips is not optimum. Maybe 60 or 80 would work better, I am not sure.

Or we can just compare intraday trades with different exits to see what exits are working best.

This would require active participation from all the traders who trade this system. We just might be able to get this system complete with suitable exits.

Any help from fellow traders would be appreciated. Or any better suggestions.
Nina has done excellent job of bringing togather a good system, but unless the suitable exits is handled this thread too will join other systems which have started in a blaze and ended up in some corner of the fx forums.

Any of you think that a chat for trading this system live would be a good idea?
Thanks again

This R/R thing is a myth, or at least it is poorly understood. The ratio between target and stop loss means nothing by itself. It doesn't show how profitable a system is. Take 2 systems, A and B. System A has a target of 100 pips and SL at 50, but it only wins 3 out of 10 times. System B has a target of 20 pips and SL at 100, but it wins 9 out of 10. Out of 10 trades, you would lose 50 pips using system A (with the "good" RR ratio) and win 80 using system B (with the "poor" RR ratio).

If Nina's entry is followed by a +10 pip move 6 times more often than a -30 pip move, then for every 7 trades you would make +60 - 30 = +30, a VERY good system in spite of the so called bad RR ratio. If, on the other hand, Nina's entry leads to -35 pips twice as often as it leads to +60, such a system would lose 10 pips for every 3 trades (on average of course), and it would be bad even though the RR ratio sounds good. This is just a theoretical example, I am not saying that it is the actual way Nina's system behaves.

The only way to determine the best settings is to test various pairs of RR on historical data, at least 50 trades. Then I suggest going to the simulator at http://hquotes.com/tradehard/simulator.html - please read the explanations there, enter the values for Win/Loss and Win Prob - and see what your expectations should be using the system.

Hi RadicalMoses
No worries, mate. (A Mexican/Italian in the middle of California doesn't get to say that too often. )
Most "traders" lose money because they don't know how to trade.
Emotions has nothing to do with discretion, just as it has nothing to do with a mechanical system. Emotions can affect both methods, but they are something totally separate.

Let me give you an example of Discretion. I'll use the IWCR "system" that has a popular following on the forum right now. IWCR seems to be fundamentally sound, but because the developers wanted to sell the 'system', they had to set hard and fast rules. Rules that don't always make sense when you just take a minute to see what price is actually doing. Let me show you what I mean. We'll assume that we have an Active Wave, that price has already closed inside the Retractment Channel, and it is now making it's way to the 75% line for a Long entry. All you're waiting for is for a full bar to close above the 75% line to enter (The Rule).

1. You get a full bar that closes above the 75% line. It's a doji, hanging man, shooting star, pinocchio bar, or any other strong Reversal bar. Why would you want to enter? It would be better to wait to see if price breaks above the high of the reversal bar before entering.

2. Price closes above the 75% line multiple times with some of them forming multiple tops. These multiple tops form shorterm overhead resistance. You finally get a full bar, a small inside bar, that closes above the 75% line. Why would you want to enter? Place a buy stop above the overhead resistance and enter on the breakout.

3. Multiples closes above the 75% line, but the bars are shrinking and forming some type of bearish pattern. You finally get a bar above the 75% line. Why would you enter? Wait for an upper breakout of the bearish pattern before entering.

4. Price slows down and starts bouncing around the 61.8% line, forming a small channel. A strong bullish bar breaks out of the channel, closing above the 75% line. The next bar is another bullish bar and the whole bar closes above the 75% line. By all means, jump on that trade. It's a good setup.

In all of those situations, by using discretion, you would have had a good reason for entering each trade, or for staying out of a trade. We used the IWCR Model, to give us good setups, but we used our discretion to make the final decision whether to pull the trigger and enter a trade.
It's just my opinion, but I think using a casino as a model for traders to follow is incorrect. The reason is that casinos make the rules, so they can make sure that every game has a positive expectation. We don't make the rules in trading. I think a better model for us traders to follow is that of the professional poker player. They know all the odds, and before making any decisions, they calculate those odds based on all the available information they have in front of them (whatever cards are visible to them). They know what their base chance of winning is, based on the cards in their hand. But, when it comes time to hold'em or fold'em, they don't just do what the odds tell them to do. They use their experience and their ability to read the other players, to make the final decision. They use their Discretion.

Please don't take any of this as criticism. I'm still very much a nube, myself. This is just based on what I've personally learned. Take a mechanical system that you believe could be successful (That's one possible way of doing things). Treat it as a Model, and not a set-in-stone system. Apply a little discretion based on price action, and then you have a chance to be a successful trader.

Please feel free to disagree. Like I said, this is just my opinion.

Keris

Ok, for a moment I was stumped becoz I thought you are trying to say that R/R is not a criteria.

LOL,, Lazy, BTW system A is not valid in anyway, simply becoz :
It has a "NEGATIVE EXPECTANCY", no matter what MM you use it wont make money.

As to system B, let me say this: it has to win over the long run simply becoz it has a positive expectancy and over the long run the profits are larger than the losses even though the % might be less.

I think you have totally got it mixed up about risk/reward ratio and win/loss percentages.

Ninas system cannot survive if you go for 35pipSL and 10pip wins even if you get 70% win/loss ratio.

7 wins = 70pips
3 losses = 105pips
P/L = -35pips

Correct me if I am wrong. I dont want to go into a debate about this.
Maybe I can suggest reading up on "The Trading Game" by Ryan Jones.
IT is supposed to be the bible on this kinda thing.

With this system there is no kinda backtest to acertain the Stats.
It is dependent on 50MA which is infact a trend indicator and if you do not catch those long trends , I dont know how it will cope with those ranging days when the prices simply shuttle above and below the MA.

BTW,, from what I have heard the best systems over a long time are 60% profitable. There are exceptional systems ofcourse but its unlikely we will get to trade them. You do the math.

Thanks again guys.

Keris,

I think there are lots of topics on which we can spend countless hours debating and never get anywhere.
Yea, I read the ICWR manual and have all the indicators, but dont have the patience to follow it. If you use discretion, you add many more Variables into the equation and make it complicated. How it will affect the systems performance? Maybe favourably or not depending on how you percieve the information and use it. Each time you use discretion, can you tell me how it increases the odds in your favour? If you used discretion and not entered the trade and it was a runaway winner,, how would it affect your next decision to trade the setup? Would you use discretion then?

Anyway, lets just get on with this sytem and tweak it so that we can get maximum outta this system. Lets leave the arguements for another sunny day by the beach when we have made our first million. LOL.

The best thing about these forums is we can agree to disagree and still get along just fine.

If I could make money using discretion , I would, but believe me, I have been more often wrong than right.

No feedback on the chat yet?

I'm afraid you didn't read my post carefully. Trust me, I know what I'm talking about. I said 6 wins out of 7 trades, not 7 out of 10. 6 out of 7 means over 85% winning probability, and Nina's system has that kind of winning probability if your target is 10 pips and your SL is 30.

My point is very simple: the expectancy is given by a combination of 2 factors - the RR ratio AND the % of winners. Merely having a better RR ratio doesn't make a system profitable. You have to actually test various pairs of RR on your charts and pick the combination that leads to the best expectancy. There are systems that work very well with a 5:1 RR, and others with a 1:5 RR, because the % of winners is so different.

I tested the 10 pip target, 30 pip stop loss combination on Nina's entry and the % of winners was over 80%. Please visit the link I posted in my previous message and enter 0.33 as Win/Loss and 0.8 as Win Prob, and see what a nice curve you get. Please test your suggestion of 70 pip target, 35 SL and see what you get. You will find that it has an inferior expectancy.

And the system CAN be backtested, very easily. The indicators used like EMA don't back adjust their position. You can simply go back in time on your charts and manually record all entries and exits using a preset of RR and see how it would have performed.

I think I read your post quite clearly.

Let me get to how the system will perform with 80% Win/loss ratio.
8 wins = 80pips
2losses = 70pips
P/L = 10pips

Averaging about 2 trade a day, that is 40 trades approximately in a month, lets see what we have : 80% profitable
i.e 32 wins = 320pips
8 losses 30pips = 240
p/l = + 80pips

Sometimes in a runaway trend you might not even get a signal for 2 days.

Ok, now how many wanna trade this system?? 80 pips with 80% accuracy?
I think its time to move on.

I agree that both R/R and win/loss are important. We know one part of the equation that its win/loss ratio is good, but what we need is to extract as much as we can.

Please dont get me wrong,, curve looks beautiful. But I doubt I would be trading this way.

There are plenty of guys who are willing to settle for 10pips a day and go back to sleep. I like to make more. Its just my nature.
That being said 10pips a day can make you enuff money to outperform a fund manager.

Anyway lets not juggle numbers around in endless circles. I am sure there is some way to extract more pips.
I will keep working.
My 70pip was just a suggestion, not optimum exit.

However, you just might have suggested something very important. IF we were to trade 3 pairs with same accuracy, it would just do pretty nicely and give very profitable returns every month. Looking forward to reading more on a followup on this kinda exit. I am open minded with a skeptics mind..lol.

Thanks for the link.

Hi,

Updated version of Fisher_org is here http://www.forex-tsd.com/7415-post16.html

Igor

__________________
Let's improve trade skills together
http://finance.groups.yahoo.com/group/TrendLaboratory

OK, radicalmoses, I understand that most people just hate the idea of getting only 10 pips. But don't forget that you can use a lot size that makes those 10 pip count. It could be 10 dollars, or it could be 1000 or more.

Anyway I am not suggesting that RR ratio because I like it. I hate it too, but what if it just turns out to be the best we can get? I would also like to take hundreds of pips per trade if possible. I am only saying that from my personal backtesting the 10/30 performs best using Nina's signals. I would be very happy to have a different RR pair that yields more, e.g. target = 100 and SL = 50, but so far I couldn't find one with a good expectancy (I tested 4 different pairs). If you or anyone else can come up with a better exit for this system please share it. Building an expert advisor would make this sort of backtesting much faster, and it can be optimized to determine the best values for SL and target. My programming skills are not good enough to do it, but maybe someone else can write it.

Thanks, atlast we are on the same page now. 10pips a days is a traders paradise, but it sucks to get 10pips and watch the market go up by 100pips more...lol. No matter how hard we try and stay neutral,, our emotions just dont deal with it. I will be happy if I can find a middle ground, say 30-40pips a day?? My programming skills are not that great either and I guess we will have to wait for a blessed soul to do it for us. Meanwhile, some one suggested CCI for confirming the entry,, however I think it can be used for exit and help us reach our target.
Try plotting 14 period CCI on the chart and use it as O/B and O/S levels to exit. IT looks great on the chart to pin point exits. See how the peaks and valleys coincide with the tops and bottoms of the moves. Here is a chart if anyone wants. Just an alternative till I find something better.

I know keris, I said I wouldnt use discretion,,lol.Sorry, I will put my foot in my mouth.

BTW,, no signals today yet. Its gonna be a long wait.

Attached Images

exits.gif (19.6 KB, 510 views)