Whats wrong with following twist to the grid or and Martingale? | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Related Instruments

GBP/USD

Similar Threads

Martingale, Anti-martingale, and Compounding 40 replies

Martingale, Reverse Martingale, Modified Martingale, Maths 1 reply

How to change this Martingale to Reverse Martingale ? 3 replies

Martingale vs. Non Martingale (Simplified RoR vs Profit) 1 reply

Grid, Burst Grid, Grid Risk and Grid Profit 1 reply

Attachments: Whats wrong with following twist to the grid or and Martingale?
Exit Attachments

Whats wrong with following twist to the grid or and Martingale?

  • #21
  • Jun 18, 2023 7:40am Jun 18, 2023 7:40am
Quoting aalaarajy
...So, it is not any up movement that will mean the pair is in an uptrend or any down movement that will mean that the pair in a downtrend. We have to look at the pair for many days and find out in which direction we are now, and if it have broken back the opposite direction or it is just pulling back up and down in same area.
Yeah, I've done a lot of statistical analysis on two to five bar patterns and could never find any correlation between a certain pattern and positive expectancy (after taking account of swap and spread). I also did analysis of uptrends and downtrends of defined pip differentials as it related to the percentage of "up bars" versus "down bars" and was amused at the number of strong uptrends where the percentage of down bars was approximately equal or greater than the percentage of up bars and vice versa for strong down trends. The unique characteristic of these strong trends was that the pip movement of only a handful of bars made up the bulk of the uptrend or downtrend.
  • #22
  • Jun 18, 2023 11:54am Jun 18, 2023 11:54am
Quoting davelansing
Quoting aalaarajy
{quote} Hi aalaarajy, Attached is a screen shot of the NZDUSD and AUDUSD from the last three weeks. I just opened a NZDUSD buy trade on June 14th due to the AUDUSD rising 100 pips between June 12th (0.67641) and June 14th (0.68809). Within that time frame, two NZDUSD trades were closed (one trade opened on June 1st due to a 100 pip rise in the AUDUSD between May 30th and June 1st, and the 100 pip TP was hit on June 13th, and a second NZDUSD buy trade was opened on June 8th due to a 90 pip rise in the AUDUSD between June
...
Aha, see. There were a little misunderstanding from me to your exact approach. I thought that you wait for the NZDUSD to go up more than 100 pips in two days before you open a buy for NZDUSD. And the sell is to wait for it to close in two days by more than 100 pips.
And this for the AUDUSD also.
But out of your last message I now understood that you open a buy AUDUSD if the NZDUSD closed in two days more than 100 pips. So your NZDUSD will be depending on the AUDUSD and not for its own movement. And this is also for the AUDUSD. It have to depend on the NZDUSD movement in order for it to trigger a trade.

Now, this is more interesting to work with. So now I will open a demo account and look for the last two days for the NZDUSD movement in order for me to open an AUDUSD trade. And also look to the AUDUSD in order for opening an NZDUSD trade.
Also I will look at other corelating instruments and make their conditions according to their movements with same approach.
  • #23
  • Jun 18, 2023 12:01pm Jun 18, 2023 12:01pm
Quoting davelansing
{quote} Yeah, I've done a lot of statistical analysis on two to five bar patterns and could never find any correlation between a certain pattern and positive expectancy (after taking account of swap and spread). I also did analysis of uptrends and downtrends of defined pip differentials as it related to the percentage of "up bars" versus "down bars" and was amused at the number of strong uptrends where the percentage of down bars was approximately equal or greater than the percentage of up bars and vice versa for strong down trends. The unique characteristic...
Ok, bear over with me. What do you mean with that? That it will never be worth to look for several bars and trying to follow its movements?
Ok, what if we only considered the bars that have movements in other areas than the previous one?
I mean what if you look at 6 or 7 bars in a daily time frame and find out that one or two of them have moved 300-400 pips down, but the others is up trends but only in the same range and not above it. So you still consider this is a down trend even if more than three bars show you positive movements, but because it is lower than the highest point in the down movement you just consider it a pullback and not a real uptrend?
  • #24
  • Edited Jun 19, 2023 8:39am Jun 18, 2023 1:21pm | Edited Jun 19, 2023 8:39am
Quoting aalaarajy
{quote}{quote} Aha, see. There were a little misunderstanding from me to your exact approach. I thought that you wait for the NZDUSD to go up more than 100 pips in two days before you open a buy for NZDUSD. And the sell is to wait for it to close in two days by more than 100 pips. And this for the AUDUSD also. But out of your last message I now understood that you open a buy AUDUSD if the NZDUSD closed in two days more than 100 pips. So your NZDUSD will be depending on the AUDUSD and not for its own movement. And this is also for the AUDUSD. It...
Hi aalaarajy,

As I described in posts#13 and #17, to open an AUDUSD sell trade, EITHER the AUDUSD OR NZDUSD has to move down by 100 pips over the two days. Likewise, to open a NZDUSD buy trade, EITHER the AUDUSD OR NZDUSD has to move up by 100 pips over the two days. I'm always using the movement of the FASTER pair to determine whether to open a new trade. I trade this way to avoid opening a bunch of opposing trades against a strong trend (i.e., not selling the AUDUSD during a strong AUDUSD and/or NZDUSD uptrend or buying NZDUSD during a strong AUDUSD and/or NZDUSD downtrend). Also, because I'm trying to minimize swap costs knowing that my trades will be opened for days if not weeks, I only buy NZDUSD and sell AUDUSD. Any two highly positively correlated pairs can be used, but like I said, I had to consider the swap (and margin costs) when selecting the pairs to trade.

Best,

Dave
  • #25
  • Edited Jun 19, 2023 8:53am Jun 18, 2023 1:53pm | Edited Jun 19, 2023 8:53am
Quoting aalaarajy
{quote} Ok, bear over with me. What do you mean with that? That it will never be worth to look for several bars and trying to follow its movements?
Yeah, but its ONLY a personal preference based on MY analysis. I could never develop a positive expectancy trading approach using bar patterns, specifically as it relates to identifying trends. Trend analysis, to me anyways, always looks great after-the-fact. But there are profitable trend traders out there, as there are profitable reversion-to-the-mean traders. Some great trend-related threads are from 60minuteman (Systematic Trend Trading | Forex Factory) and Copernicus (Diversified Trend Trading – Excel Projects | Forex Factory). Basically, you pick any method to get into a trade, and then use a (very) tight stop loss and very large TP to manage the trade; the goal is to hit the low percentage of the fat right-tailed distribution trend trades that yield the big winners (10-20% of your trades), but to get those, you have to suffer through the numerous small loses from the whipsaw trades (70-90% of your trades). Flotsom is another FF member whose posts are illuminating with regards to trend trading. I personally can't handle the constant losing.

My goal was/is to develop a simple and profitable non-directional trading approach to random price movement that takes no time to ascertain whether I'm buying, selling, or do nothing at all. It's not that I'm lazy, I still do a lot of back testing of ideas, but to me the more analysis that goes into making a trading decision, the stronger is your belief that the narrative your analysis is giving you is "true", while forgetting that market movement is mostly random. Price movements, either as a single bar, or a group of bars, that you see when you back test and which forms the basis for many pattern analyses - you have no clue what's behind a move. Was a major price move due to an unexpected geopolitical event? An unexpected (or expected) economic news? Central bank news? A flippant comment/tweet made by a politician, Treasury Minister, or Central Banker? A weather-related event? Unless you have all this historical information annotated on your charts, how can you accurately back test bar patterns and tag the pattern to the news event? How can you ascertain the impact of an unexpected news event on price beforehand?

But don't base your research on my biases. You could easily develop a profitable bar pattern/trend approach that works for you and your risk tolerance.

Best,

Dave
  • #26
  • Jul 14, 2023 8:17am Jul 14, 2023 8:17am
Quoting davelansing
{quote} as long as their average daily correlation is above 70....
Hi Dave,

What do you do if the correlational were to drop significantly below 70% whilst you're in a basket? Nice strategy, will take a closer look.

Regards
Simon
  • #27
  • Edited Jul 15, 2023 10:22am Jul 14, 2023 12:12pm | Edited Jul 15, 2023 10:22am
Quoting swd
{quote} Hi Dave, What do you do if the correlational were to drop significantly below 70% whilst you're in a basket? Nice strategy, will take a closer look. Regards Simon
Hi Simon,

When that happens my trade size was small enough to "wait out" the uncoupling (for AUDUSD/NZDUSD the uncoupling can last a few months using daily data as input and a lookback of 100 days) and still open trades based on the "fastest" moving pair- which means opening NZDUSD buy trades only on up moves and AUDUSD sell trades on down movements. But a few weeks ago, I did a manual back test of 2021 (my original back test was from 2016 through 2020 - a period of strong trends with periods of consolidation in order to stress-test the system), and the results were NOT very encouraging. There was a 6-month period where there was essentially no positive PL movement in the basket. The good news was that the DD was <20%, but 6 months without a TP was rather disappointing (daily swap costs and spreads were taken into account (spread for AUDUSD = 1.5 pips and NZDUSD = 2.5 pips, based on limited Oanda historical spreads, although I had to estimate the swap costs and I errored by using current swap rates). There's probably a way to overcome this lack of positive PL movement, and it may simply involve dropping the pip differential to 25 pips (or some low percentage of the daily ATR) so that more trades are opened within a shorter time frame and that the objective is to hop on the very short-term intraday trends. Alternatively, the old "tried-n-true" of opening NZDUSD buy trades only on higher swing highs and AUDUSD sell trades only on lower swing lows, as long as a minimal pip differential is met, might work as well. Another option is to hedge the hedge, lol. Since I am only buying NZDUSD and selling AUDUSD (to minimize o/n swap cost with Oanda), I'm essentially selling AUDNZD. Therefore, buying AUDNZD at an "appropriate" price point (whatever "appropriate" means to you, i.e., whatever you can identify an uptrend in the AUDNZD) would hedge the hedge. Will need to manually back test this. As I am currently on vacation, I'll have plenty of time to back test!


Best,

Dave
  • #28
  • Last Post: Jul 15, 2023 12:54pm Jul 15, 2023 12:54pm
Quoting davelansing
{quote} Hi Simon, When that happens my trade size was small enough to "wait out" the uncoupling (for AUDUSD/NZDUSD the uncoupling can last a few months using daily data as input and a lookback of 100 days) and still open trades based on the "fastest" moving pair- which means opening NZDUSD buy trades only on up moves and AUDUSD sell trades on down movements.
edit: "...and still open trades based on the "fastest" moving pair- which means opening NZDUSD buy trades only on AUDUSD and/or NZDUSD up moves and opening AUDUSD sell trades on AUDUSD and/or NZDUSD down moves as long as the minimum pip differential is met.
0 traders viewing now