tar
Legendary member
- Messages
- 10,443
- Likes
- 1,314
NO it doesnt , like everything else nothing works for ever . period .
NO it doesnt , like everything else nothing works for ever . period .
megamuel
Experienced member
- Messages
- 1,098
- Likes
- 164
Interesting thread trendie. I've never heard of the term 'grid trading' before. Although I have played around with the idea, not knowing it had a name. I know the dangers of the martingale betting system, but I must say martingale combined with mean reversion and good money management does really intrigue me. I've always liked the idea of waiting for something to deviate away from the mean by quite a bit, then entering with a martingale system up to 3 times. So for (a simple) example, wait until a indicator is oversold, enter long 0.25% of account, have a predefined stop and target with equal risk:reward, if stop is hit enter again with 0.5% of account, if stop is hit again enter with 1%.... If stopped out again wait for new trade and go back to 0.25% of account. It would rely on mean reversion. Just an idea anyway, it would need some work but it does appeal to me.
I also worked on an idea a few weeks ago using martingale and random entries, relying on not having 3 losing trades in a row often enough to outweigh the small wins. On paper it seemed to work so I had it programmed, but instead of the random entries I used fractals to enter to try and improve it. So I'd look for 3 bars forming a fractals as opposed to 5, then open a trade on the next bar providing it opens in the lower 50% of the previous bar for a short or upper 50% of previous bar for long. Stop placed above previous bars high for short and below previous bars low for long. Equal R:R, same martingale staged entry as previous. Back testing results were pretty dismal to be honest! I don't mind it when things don't work though, at least you rule something out and its all fun and games anyway. I'll add picture of back testing results and some notes I gave to the programmer. Would be cool if anyone would like to work on these ideas with me and help me improve them.
Sam.
I also worked on an idea a few weeks ago using martingale and random entries, relying on not having 3 losing trades in a row often enough to outweigh the small wins. On paper it seemed to work so I had it programmed, but instead of the random entries I used fractals to enter to try and improve it. So I'd look for 3 bars forming a fractals as opposed to 5, then open a trade on the next bar providing it opens in the lower 50% of the previous bar for a short or upper 50% of previous bar for long. Stop placed above previous bars high for short and below previous bars low for long. Equal R:R, same martingale staged entry as previous. Back testing results were pretty dismal to be honest! I don't mind it when things don't work though, at least you rule something out and its all fun and games anyway. I'll add picture of back testing results and some notes I gave to the programmer. Would be cool if anyone would like to work on these ideas with me and help me improve them.
Sam.
Attachments
Leave it alone trendie, you know it won't work. It's a dumb system from days before private traders even existed, and when we couldn't short. In the old days, it was expensive to get in and out of positions and price data was never real-time, so stop losses were miles from entry. If price fell after a share purchase, all you could do to hedge was average down. This persisted into recent times - ask the 90% club from 2001.
Today, if you go long and price falls, what we can do now is exit and reverse, go short. You'll be earning from the word go, not eventually-maybe-one day-when price recovers. A series of grid trade entry points just looks like a road map for shorters.
Today, if you go long and price falls, what we can do now is exit and reverse, go short. You'll be earning from the word go, not eventually-maybe-one day-when price recovers. A series of grid trade entry points just looks like a road map for shorters.
oildaytrader
Senior member
- Messages
- 2,806
- Likes
- 125
Some idiot showed a new trader gridding .The noob blew his account of $100k, the idiot blew up $700 k in 3 days.
Think!
oildaytrader
Senior member
- Messages
- 2,806
- Likes
- 125
I can make gridding work , but I prefer based it around support and resistance levels with multiple pairs , using stops and progressive betting.It will work.It can work around daily s/r, and based around dynamic support and resistance .
Don't ask for my secrets.
Hoggums
Senior member
- Messages
- 2,176
- Likes
- 878
I have tested this sort of system to death and unless you allow for a ridiculous amount of grids then you eventually blow up. Take last year from apr-Dec on the equity indices for example.
Over time you end up with huge losses from positions at the extremes of the market, add in rollover costs which even if you are using long dated - futures you eventually have to mark-to-market those horrendous positions.
Problem with allowing say 50 grids to be triggered in your setup is that the profit from each grid move you cash is so tiny you never make any meaningful profit. Then factor in the running losses and margin requirement and you are onto a loser.
Over time you end up with huge losses from positions at the extremes of the market, add in rollover costs which even if you are using long dated - futures you eventually have to mark-to-market those horrendous positions.
Problem with allowing say 50 grids to be triggered in your setup is that the profit from each grid move you cash is so tiny you never make any meaningful profit. Then factor in the running losses and margin requirement and you are onto a loser.
trendie
Legendary member
- Messages
- 6,875
- Likes
- 1,433
OK. Since I know you are meticulous in your research and analysis, I will bear your views in particular in mind.
But I need to do the sums myself.
oildaytrader
Senior member
- Messages
- 2,806
- Likes
- 125
Beyond the comprehension of most of you.It is a dynamic grid system with stop losses ,stochastic indicator ,counter trend trades etc and you need to be intelligent to test it.
trendie
Legendary member
- Messages
- 6,875
- Likes
- 1,433
:-0 is that the most patronising and insulting reply ever? :-0
oildaytrader
Senior member
- Messages
- 2,806
- Likes
- 125
megamuel
Experienced member
- Messages
- 1,098
- Likes
- 164
Who tested it for you?
trendie
Legendary member
- Messages
- 6,875
- Likes
- 1,433
What a jock, trendie's trends seem to be failing, hence search for a grid.
I am a confirmed, dyed-in-the-wool trend-follower. Trends are appearing on a regular basis, thank you very much.
Does it occur to you that I am simply bored with what I have, and am searching for some intellectual stimulation? A challenge.
You IDIOT!
oildaytrader
Senior member
- Messages
- 2,806
- Likes
- 125
No! because I wouldn't be able to back test it myself , the rules would be so complex and conflicting , there I am no smarter.
The only way to test it is to forward test it is on demo, but who is going to spend months testing on demo and working without pay?
oildaytrader
Senior member
- Messages
- 2,806
- Likes
- 125
It is part of my discretionary trading.
