Take 10 pips or hope for breakout?

[NVP]

I have a strategy that I can get about 5-20 pips or so pretty easy. Sometimes I take profit there and it goes on to do 50-100-200+ pips and that annoys me or I'll move stop up to break even when it's ~5 pips profit and it stops out even after hitting 10+ pips or so.

The problem or conundrum I'm having is whether to just take the 5-20 pips when it's there or just move my stop to break even and hope for 100+ pips? I know that you know nothing about the setup I use, it's sort of a combo of a trend following and a band trading strategy.

Just mentioning two trades(I made almost 10 trades so far this week, all break even and one stopped out at a loss), I went long USDCAD and took profit at about 12 pips, but then it went on for another 50. I also went long EURUSD, I could have taken profit at about 10 pips, but then just moved my stop to break even and got stopped out. Just judging from these two trades, it seems move the stop up and hope for more is a better bet, but I just wanted to hear from someone more experienced.

split the trade into 2 (or even 3)......apply different Exit strategies to each and net the best of all worlds ...:smart:.

N

 

[the hare]

I have a strat that makes 10 pips on 75% of the entries, but it loses 100 pips on the other 25% - can you make me rich ?

It would be very rare indeed to find a system with such a large negative edge ! If you find one, let me know

 

[neil324]

Just set it up mechanically. Take 2/3 off leave 1/3 to run.

Job done.

 

[bbmac]

So a 'sample size' of 10 trades is sufficient for you to determine whether a long term viable edge exists is it ?

Lets say it is, and that the 75% strike rate stands up (ie 75% winning trades as a % of total trades,) and I win 10 pips on each of the winning trades but lose 40 pips on each of the remaining losing trades...so I win 75pips (7.5 x 10) on the winners and lose 100pips on the losers (2.5 x 40)....to make this an overal winning strategy you would need to increase stake size with each losing trade (dutching/martingale)...what size bank in units should I consider using if I am starting with 1 unit as the 1st trade of the sequence ?



G/L

I may be able to. But I will have to see the characteristics of the strat. Call out 10 trades will suffice.

 

[BeginnerJoe]

So a 'sample size' of 10 trades is sufficient for you to determine whether a long term viable edge exists is it ?

Lets say it is, and that the 75% strike rate stands up (ie 75% winning trades as a % of total trades,) and I win 10 pips on each of the winning trades but lose 40 pips on each of the remaining losing trades...so I win 75pips (7.5 x 10) on the winners and lose 100pips on the losers (2.5 x 40)....to make this an overal winning strategy you would need to increase stake size with each losing trade (dutching/martingale)...what size bank in units should I consider using if I am starting with 1 unit as the 1st trade of the sequence ?



G/L

As you already know my belief about the market, 10 trades is plenty, assuming you don't get kicked in the b*lls much sooner than that.

Who said anything about martingale ? We can get into the details once you have demonstrated you can produce what you claim. As for the sample size, I have my own ways of assessing it and make allowances where reasonable.

If you succeed with the test call outs, then I will trade 10 more call outs using micro size to ensure the market is given an opportunity to react to real trades, and see how you do then. If you still succeed, then we are in a position to talk about bank sizes.

 

[bbmac]

The initial question was rhetorical to demonstrate that it's not all about entries,...I have no such edge - it was a ficticious example as others clearly understood judging by the rep given to the post. Doh !

As you already know my belief about the market, 10 trades is plenty, assuming you don't get kicked in the b*lls much sooner than that.

Who said anything about martingale ? We can get into the details once you have demonstrated you can produce what you claim. As for the sample size, I have my own ways of assessing it and make allowances where reasonable.

If you succeed with the test call outs, then I will trade 10 more call outs using micro size to ensure the market is given an opportunity to react to real trades, and see how you do then. If you still succeed, then we are in a position to talk about bank sizes.

 

[cablemonster]

oh darn it.

 

[BeginnerJoe]

The initial question was rhetorical to demonstrate that it's not all about entries,...I have no such edge - it was a ficticious example as others clearly understood judging by the rep given to the post. Doh !

It's unsurprising to me you have no such edge. You'd be rich if you did. But I always give people a chance to prove their claim. Now I hope split can see the utility of my asking for statements. Otherwise all the newbs will start grasping at any straw going.

 

[bbmac]

No I am not surprised either - any edge with a negative expectancy which this one as I have described it - has - is no edge I am interested in - but you clearly were and that too is unsurprising to me lol!

It's unsurprising to me you have no such edge. You'd be rich if you did. But I always give people a chance to prove their claim. Now I hope split can see the utility of me asking for statements. Otherwise all the newbs will start grasping at any straw going.

 

[BeginnerJoe]

No I am not surprised either - any edge with a negative expectancy which this one as I have described it - has - is no edge I am interested in - but you clearly were and that too is unsurprising to me lol!

It's only negative because you can't see what I can see. I think you are cross eyed from staring too much at all the lines in your chart to see things clearly, to see there are other possibilities.

 

[the hare]

The initial question was rhetorical to demonstrate that it's not all about entries,...I have no such edge - it was a ficticious example as others clearly understood judging by the rep given to the post. Doh !

Pearls before swine

On a positive note, Im reliably informed the beacon is shining brighter day by day

 

[cablemonster]

this pic is for the hare.

 

[Thermal Ebb]

split the trade into 2 (or even 3)......apply different Exit strategies to each and net the best of all worlds ...:smart:.

N

NVP beat me to it. That's precisely what you do. Back testing is largely irrelevant as it does not factor in the real-time psychological element. We're all experts in retrospect and would have done this and would have done that. In reality, faced with making decision in real-time, it's obviously a lot different.

By running your models in real-time with small stakes you get to fully test them and you get to feel how they feel as you try them all on. Keep meticulous records and don't depend upon your memory for which method is optimum. The ones which net you the most pips will stick in your mind, but they may not be the ones you need to be remembering.

The only further suggestion I'll make is to bias your trading to methods which require the fewest decision points from entry to exit. Each trade isn't just a decision for entry and one for exit. It's a decision at every print along the way. Decisions, even those where you end up doing nothing take energy and that can sap you as badly as loss of capital.

If you find that you've got two or more methods that seem to have a roughly equal performance profile then you're a lucky guy as you can choose which and when to run them. That's the fun part and that's what trading should always be. Sounds like you can get to b/e quickly so shouldn't be too much of a push for you to get comfortable with leaving the rest on and just have some fun with it. Good luck.

 

[barjon]

........... you can't see what I can see.............

Yes, I suppose it must be the rarefied atmosphere on that 0.5% cloud (I won't ask for a statement ) that does it. Such a shame that, with that special vision not given to the rest of us, your contributions to the boards are 99.5% (there's that percentage cropping up again ) destructive with the odd "I know something you don't know" thrown in just to keep the pot boiling. All very helpful to many, I'm sure.

 

[cablemonster]

NVP beat me to it. That's precisely what you do. Back testing is largely irrelevant as it does not factor in the real-time psychological element. We're all experts in retrospect and would have done this and would have done that. In reality, faced with making decision in real-time, it's obviously a lot different.

By running your models in real-time with small stakes you get to fully test them and you get to feel how they feel as you try them all on. Keep meticulous records and don't depend upon your memory for which method is optimum. The ones which net you the most pips will stick in your mind, but they may not be the ones you need to be remembering.

The only further suggestion I'll make is to bias your trading to methods which require the fewest decision points from entry to exit. Each trade isn't just a decision for entry and one for exit. It's a decision at every print along the way. Decisions, even those where you end up doing nothing take energy and that can sap you as badly as loss of capital.

If you find that you've got two or more methods that seem to have a roughly equal performance profile then you're a lucky guy as you can choose which and when to run them. That's the fun part and that's what trading should always be. Sounds like you can get to b/e quickly so shouldn't be too much of a push for you to get comfortable with leaving the rest on and just have some fun with it. Good luck.

a thoroughly sensible post sir.

 

[bbmac]

Lol - No It's a negative expectancy because it's a negative expectancy. Period. (7.5 x 10pips/trade = +75) less (2.5 x -40pips/trade = -100) = -25 divided by total trades [= 10] = expectancy of -2.5pips/trade.

Thanks for the concern over my optical health - but I can see things very clearly for what they are, - all things, as I am sure many others can !

Two ways to make an 'edge' with a negative expectancy profitable ?

1. Martingale/dutching - ie increase the next trade stake size after each loser - stop at a winner.

and/or

2. Average the existing trade on the same or correlated (directly or inversely correlated) instrument hoping for a 'mean reversion' By the way how speaking of averaging How are your cable longs getting on a la Cablemonster's thread - ie your methodology to keep buying in a clear downtrend ?

It's only negative because you can't see what I can see. I think you are cross eyed from staring too much at all the lines in your chart to see things clearly, to see there are other possibilities.

 

[jacknapier]

To the OP, you made the above statement. What does this mean?

Well... I suppose it's subjective since it seems to work a lot better than what I was doing before. ie: gambling... It seems to work, since the only real draw down I've had since I started doing it are when I don't follow my rules. You know, gamble type trades or impulsive trades.

I'll have to read the rest of the posts tonight as I'm busy atm.

 

[cablemonster]

By the way how speaking of averaging How are your cable longs getting on a la Cablemonster's thread - ie your methodology to keep buying in a clear downtrend ?[/I]

lol I think he got his ar5e handed to him. I did try to tell him the bottom of the monthly range was on balance of probabilities the most likely outcome. I think he was busy 'manipulating the manipulators'.

 

[the hare]

Yes, I suppose it must be the rarefied atmosphere on that 0.5% cloud (I won't ask for a statement ) that does it. Such a shame that, with that special vision not given to the rest of us, your contributions to the boards are 99.5% (there's that percentage cropping up again ) destructive with the odd "I know something you don't know" thrown in just to keep the pot boiling. All very helpful to many, I'm sure.

That's a bit unfair. As a fellow poster of absolute drivel (and worse, according to your content management team) I have to show a bit of solidarity with Joe. We provide a useful function, allowing others to highlight the error of our ways.

Some of the brightest shining contributions recently at the beacon have been made in direct response to Joes posts.

You might well argue that if it wasn't Joe posting garbage someone else (such as my good self) would. You may be right, but not all members are as resilient in the face of criticism and ridicule as Joe and I.

I presume your criticism of Joe was a personal opinion rather than an official t2w perspective on his contribution ?

 

[wackypete2]

lol I think he got his ar5e handed to him. I did try to tell him the bottom of the monthly range was on balance of probabilities the most likely outcome. I think he was busy 'manipulating the manipulators'.

Peter