First-Strike & One-Night-Stand users, show yourself and your mods, if you dare!!! | Page 9 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

How Do You Compound Your Account While Paying Yourself? 23 replies

I dare you show your statement 13 replies

Twoblink tweaks and mods 61 replies

Vegas 4 hour Tunnel mods 4 replies

Attachments: First-Strike & One-Night-Stand users, show yourself and your mods, if you dare!!!
Exit Attachments

First-Strike & One-Night-Stand users, show yourself and your mods, if you dare!!!

  • #161
  • Jan 20, 2009 5:24am Jan 20, 2009 5:24am
Quoting mikkom
Way of the turtle is a good book btw, let's see if this link works
http://books.google.com/books?id=C3O...um=3&ct=result
Thanks for the link, seems like I have this e-book somewhere on my hard-drive, and will study the position sizing part...
  • #162
  • Jan 20, 2009 5:30am Jan 20, 2009 5:30am
Quoting Jimmy Jones
The million dollar question. IF I had that figured out I would not be futzing around here .
Just reminded me — I saw a nice 4-part video series about exit strategies, here's part 1, you can get to all 3 remaining parts from there... Just in case you're interested.

In fact I _do_ believe in discretionary trading (I guess this is what we're talking about), the only thing that bothers me is that you have to spend countless hours in front of your computer to "backtest" that system that's called "my own brain & guts"
  • #163
  • Jan 20, 2009 5:31am Jan 20, 2009 5:31am
Quoting radiantchild
Just reminded me —ScamI saw a nice 4-part video series about exit strategies, here's part 1, you can get to all 3 remaining parts from there... Just in case you're interested.

In fact I _do_ believe in discretionary trading (I guess this is what we're talking about), the only thing that bothers me is that you have to spend countless hours in front of your computer to "backtest" that system that's called "my own brain & guts"
Anything worth having is worth working for
  • #164
  • Jan 20, 2009 5:35am Jan 20, 2009 5:35am
Quoting findcount
makes sense if he's selling or promoting something..........so is he ?
I believe Joel mentioned his seminars in one of the FSP/ONS emails, where he teaches his another system, London Squeeze (obviously his own take on trading London session breakouts)...
  • #165
  • Jan 20, 2009 5:35am Jan 20, 2009 5:35am
Quoting Jimmy Jones
I don't believe a mechanical system is the way to achieve success. If it were that easy no one could make money in the markets because no one would lose.

This week, on the First Strike trades that are well into profit now, it seems ludicrous to me to stay mechanical in the approach to the system, even when it becomes clear that the trend may be changing and is heading back and taking profit away. Why not take profit? Why sit and watch price go back and stop you out and say "shoot, well I followed the system"?

It also seems ridiculous to enter...

Hi,

Everything depends on the approach - whether live trading or back-testing.
(unless you are trying to backtest a scalping model which most likely will not hold up in live trading)

Otherwise, it is about the known parameters. The only real known parameters you can define is your risk, i.e. your stop on every trade and your position size and your diversification.

You cannot define your profit since you don't know at any point in time whether the market will keep going your way, in fact you don't know where it will go full stop.

One other factor is that the price will move, sometimes a little and sometimes alot. But when? No way of knowing. So what can we do? Be positioned in the market as often as possible for as long as possible with as low risk as possible in order to capitalize when the odd massive move happens in your favour.

To control risk further, you can diversify. Have the same exposure spread over 2, 3, 4 or 5 currency pairs (or other instruments) to increase your chance of being in the right one(s) when it happens, but have no increased risk by doing so.

And the last issue is time. Give the portfolio or whatever you call it time to develop. Big sustained moves don't happen that fast. Big moves can of course happen when volatility is massive like autumn last year but with diversification you smooth out some of that risk associated with it.

And to bypass the temptation to "take profits" specifically in your backtest - which as you said is likely, use time-exits without compromise.

Entries are the least relevant point, except to use maybe early breaks of supp/res. which can be done mechanically as well.

That should do it.

regards
daytrading
Enter Signature
  • #166
  • Jan 20, 2009 5:47am Jan 20, 2009 5:47am
Quoting radiantchild
Thanks for the link, seems like I have this e-book somewhere on my hard-drive, and will study the position sizing part...
I don't know if it's available as ebook, at least I have a hardcover version
http://www.amazon.com/dp/007148664X
  • #167
  • Jan 20, 2009 8:19am Jan 20, 2009 8:19am
Quoting radiantchild
(and I call that book The Bible)
i did not read the replies bellow this post but i will take some popcorn.
  • #168
  • Jan 20, 2009 8:39am Jan 20, 2009 8:39am
Quoting giraia_br
i did not read the replies bellow this post but i will take some popcorn.
Well, no religious wars happened, and probably will not... I consider this book a must read for all traders, it was a real revelation to me at the moment and directed me into mechanical systems and mechanical trading, but actually I do not label all other styles and approaches "heresy"...
  • #169
  • Jan 20, 2009 8:53am Jan 20, 2009 8:53am
Quoting radiantchild
Nice one, thanks... May I ask — what's the hope than? If something that performs good on a backtest doesn't guarantee future performance (and it actually doesn't), why not just take FSP and make some money while it seems to work? Why backtest at all?

No sarcasm, just would be grateful to hear you expand your point...
I think what tdion means is, we've already proven the long term results from FSP are fudged. They've only been working for a very short time. This does not mean that the strategy works. This means it could very well be luck(it also doesn't mean it MUST be due to luck, just that we don't have enough data to discount that possibility)

When something performs well in a backtest, there's no guarantee that they will continue working. Think of it as driving a car( a cliche i know). You've been driving for 10 years without crashing. That doesn't mean you won't tommorow. Road conditions change, you don't interact with the same cars everyday, etc. But not taking the lessons and practicing before charging out on the freeway is just silly. That's what trading a system without backtest is like.


Quoting Jimmy Jones
I don't believe a mechanical system is the way to achieve success. If it were that easy no one could make money in the markets because no one would lose.
It really is that easy. But first you must have a robust system. Also, if you're wondering where the money would come from if the hypothesis that the mechanical system absolutely works were true, you need only stand in front of the mirror.


Quote
This week, on the First Strike trades that are well into profit now, it seems ludicrous to me to stay mechanical in the approach to the system, even when it becomes clear that the trend may be changing and is heading back and taking profit away. Why not take profit? Why sit and watch price go back and stop you out and say "shoot, well I followed the system"?

It also seems ridiculous to enter a trade when a 1hr system trigger has happened (IE first strike closing over the trip line). Why not take the signal as an opportunity and look for a higher...
You could take the signal to exit. A mechanical system doesn't say you can't do it(as long as it proves profitable). But following a mechanical system DOES mean that you do it EVERYTIME the situation presents itself.

Quote
Discipline to me is the ability to recognize you mistimed an entry or formed the wrong hypothesis, and force yourself to cut the loss short before it hits the stop loss. Discipline to me is not the ability to strictly adhere to a mechanical system even when all the signs are telling you a very nice profit is about to turn into a loss. That seems more like ignorance than discipline (no offense intended to anyone here, it is possible to be extremely intelligent yet be ignorant about something specific).
I think you pretty much described what ISN'T discipline.



Quote
If only it were that easy to code up some ea's, perform back test on 90% model quality data, and then hit buy or sell when the system said so.
That's discipline, and no, it isn't as easy as it sounds.

Quoting findcount
makes sense if he's selling or promoting something..........so is he ?
Yes he is=)

Quoting Jimmy Jones
The million dollar question. IF I had that figured out I would not be futzing around here .

Price behavior around pivots, support and resistance (both horizontal and diagonal), big impact fundamental changes, gut feeling, I don't know. I do know it's a lot harder to go broke by taking profit.
Pet peeve of mine. I hate it when I hear people advocating you can't go broke taking profits. You CAN, and you WILL, if your system relies on high win/loss ratios to offset low success rates to obtain +EV. That doesn't make the system flawed. It just makes it slightly more counter-intuitive.


Quote
But, I do not trade for a living yet, so take what I say with a grain of salt. And also remember, if the majority lose, perhaps it makes sense to not do what the majority are saying to do.

While this makes good sense, it should also be remembered that the majority of traders also don't log on and discuss their strategies and failures/success.
  • #170
  • Jan 20, 2009 8:58am Jan 20, 2009 8:58am
Quoting deathbait
I think what tdion means is, we've already proven the long term results from FSP are fudged. They've only been working for a very short time. This does not mean that the strategy works. This means it could very well be luck(it also doesn't mean it MUST be due to luck, just that we don't have enough data to discount that possibility)

When something performs well in a backtest, there's no guarantee that they will continue working. Think of it as driving a car( a cliche i know). You've been driving for 10 years without crashing. That doesn't mean...
To each his own. Sounds like you have it figured out.
  • #171
  • Jan 20, 2009 9:16am Jan 20, 2009 9:16am
Quote
Yes he is=)

What is he selling??
  • #172
  • Jan 20, 2009 9:27am Jan 20, 2009 9:27am
Quoting BentRod
What is he selling??
He charges $$ for his seminars.
Also, you apparently can't replicate his results(this is from his blog) unless you have his money management system(hint, hint).

He allegedly sold it to a company for 10k.

So yes, he does sell stuff.
  • #173
  • Jan 20, 2009 9:32am Jan 20, 2009 9:32am
Quoting deathbait
He charges $$ for his seminars.
Also, you apparently can't replicate his results(this is from his blog) unless you have his money management system(hint, hint).

He allegedly sold it to a company for 10k.

So yes, he does sell stuff.
I wasn't aware of that.

Where are the links to the Seminars and his MM algo?
  • #174
  • Jan 20, 2009 9:44am Jan 20, 2009 9:44am
Quoting BentRod
I wasn't aware of that.

Where are the links to the Seminars and his MM algo?

you need to be on his mailing list to get info for his meetups

also, i seem to have misquoted his price for his algo
http://infiniteyieldforex.blogspot.c...y%20Management

he apparently charged the company 17k, not 10k. Didn't double check before I post. My bad.
  • #175
  • Jan 20, 2009 9:56am Jan 20, 2009 9:56am
Quote
you need to be on his mailing list to get info for his meetups

Fair enough, I'm not on the mailing list, I just check out his page every now and then.

Quote
also, i seem to have misquoted his price for his algo
http://infiniteyieldforex.blogspot.c...y%20Management

I can't see what you are looking at, I'll take your word for it.
  • #176
  • Jan 20, 2009 10:23am Jan 20, 2009 10:23am
Quoting BentRod
I can't see what you are looking at, I'll take your word for it.
please don't. Just scroll down to the bottom of the page and read the paragraphs in red.
  • #177
  • Jan 20, 2009 1:23pm Jan 20, 2009 1:23pm
Based on this quote

Quote
My position size is based directly proportional to the probabilities of the specific trade I am taking. As I am aware of the precise edge of my systems, I can be much more precise than someone who has the rules for the trade, but not the exact figures for its edge(s).

his MM method sounds a lot like kelly (which is basically optimal if you know the possibilities).

One thing that is unclear for me is how can he know the probabilities for each trade. He can have historical probabilities but they mean nothing. Or he can have probabilities based on volatility which is much more likely. Probably it has something to do with calculating optimal position size through kelly like algorithm with probabilities weighted by volatility but that's just my guess based on the above quote.

btw, I have experimented with kelly but for my taste it's too volatile. Just a final note, interestingly using kelly on backtested results with known probabilities the results will be curvefitted.

Note that I tested the method with simple percentage risk method and the results were still positive so it's not just about his MM algo.
  • #178
  • Jan 20, 2009 1:35pm Jan 20, 2009 1:35pm
Quoting mikkom
Based on this quote



his MM method sounds a lot like kelly (which is basically optimal if you know the possibilities).

One thing that is unclear for me is how can he know the probabilities for each trade. He can have historical probabilities but they mean nothing. Or he can have probabilities based on volatility which is much more likely. Probably it has something to do with calculating optimal position size through kelly like algorithm with probabilities weighted by volatility but that's just my guess based on the above quote.

btw, I have experimented...

He can't be using kelly. Because he (claims to) follow his system perfectly, every one of his trades have a pre-defined EV. The kelly system being fed this EV(which is a constant) will return a constant trading size. Thus, if he's using kelly as his MM, his equity curve would look the same as anyone who's trading a fixed size per trade strategy.

As hanover pointed out before, MMs don't help you increase your profits. They can't...it's not what they're meant to do. MMs simply aim to reduce your risk of ruin. Since reduced risk means reduced returns(diversification aside), a good MM system will actually be reducing your returns in return for long term survival.
  • #179
  • Jan 20, 2009 1:38pm Jan 20, 2009 1:38pm
Quoting mikkom
Note that I tested the method with simple percentage risk method and the results were still positive so it's not just about his MM algo.
I can't wait to get home from work to try to replicate your positive results... I have another idea to try, which sounds quite simplistic but why not - to enter bigger position sizes in the direction of a trend, using simple 33SMA as trend indicator. I'm pretty sure this is not what Joel is doing, but somehow I believe in that old "trend is your friend" thing and lets pretend we can catch at least 50% of the trend with that MA...
  • #180
  • Edited 2:09pm Jan 20, 2009 1:56pm | Edited 2:09pm
Quoting deathbait
He can't be using kelly. Because he (claims to) follow his system perfectly, every one of his trades have a pre-defined EV. The kelly system being fed this EV(which is a constant) will return a constant trading size. Thus, if he's using kelly as his MM, his equity curve would look the same as anyone who's trading a fixed size per trade strategy.

As hanover pointed out before, MMs don't help you increase your profits. They can't...it's not what they're meant to do. MMs simply aim to reduce your risk of ruin. Since reduced risk means reduced returns(diversification...
Kelly returns optimal ratio in percentage to risk with each trade, not fixed position size.

The real problem with kelly is that the probability with trading is not actually predictable. Kelly is also very aggressive and kelly leads to very volatile, unoptimal profit curve. I prefer very much lot lower risk, somewhere between 2-5%.

I don't want to get into fundamental argument if MM helps or not but just as an example you can't use kelly if you don't use percentage based position sizing.

The point of MM is to restrict losses and to make your risk level quantifiable, not make gains bigger.

Or maybe I misunderstood you, what does EV mean?
0 traders viewing now