Gamblers Dream 2.0 EA

A Hedging, Martingaling, Gridding kind of EA.

So, I have been playing with the dreaded martingale for several years now. Tons of blown accounts (which is what the martingale is famous for).

Since I look at Forex through the eyes of a gambler and not an investor. This EA was designed around that philosophy. Similar to playing multiple seats at the blackjack table. Trading multiple pairs at the same time instead of hands of cards.

Recently, I came up with the idea to limit the number of iterations that the martingale could use. I settled on 8 or less. This way, I could control loss to a maximum as set by the base lot size. .01, .02, .04, .08, .16, .32, .64 and finally 1.28 if all 8 were used.

That being said, I started playing with different entries and exits that would (hopefully) never exceed that fatal 8th loss. Or if it did, the account balance would be high enough to cover the loss (around 20%) and the multiple pairs would recover it quickly.

3/14/21 - New Settings added. New TE added on my profile. New stats on myfxbook too. http://www.myfxbook.com/members/FXSc...eam-20/8069037

Good luck with it, so far all grid EAs crashed my account, some after few days others after 3-4 months

Thank you! I think the secret is limiting the number and size of the grid to when a sequence loss eventually happens, it will not blow the account. Plus, if you withdraw money on a regular basis, you can actually enjoy it and not go for the HUGE accounts.

Once the EA is online, it will be risking 20% of the base balance for a losing sequence. But, on an average it should make that every few days or so. As long as it does not have more than 1 sequence loss every 3 days, it should make more than it loses. All in theory of course.

Ok please keep us updated on the performance.

Will do. Right now I am just running back tests to get an average of grid size per pair. Right now, all pairs are running the same settings, which is not what it will be in the long run. I am just looking for trading errors with the EA.

Last week ending February 26, I was testing functions of the EA on random pairs. This weekend, I am working on back tests to isolate grid sizing and investment requirements and which pairs the EA works best on. I am keeping to a maximum of 20% risk per losing sequence as I want this to be affordable to all trader sizes.

My next post will show data of a 1 year back test on all currency pairs which will give an idea of pairs to trade and more importantly NOT to trade with the same settings. This will give a breakdown of how many losing sequences occur in a testing year. Caveat that every year is different but here we are looking to establish a baseline for continued testing.

The spreadsheet below shows the different amounts lost based on grid size of maximum 8 Longs and 8 Shorts in a grid style format.

So, after spending the last couple days back testing each pair over the same year with the same settings, the spreadsheet above shows how many times a grid closed in max loss. A max loss is defined as placing a limit on how many grid levels a trade has and if the price moves beyond that, all trades are closed in loss. Each grid level for this test was 40 pips, but 2x8, 2x7, etc shows a 2x multiplier in lot size per grid level x the number of levels. The fewer the grid levels, the more often a sequence loses at a lower amount per loss. The more grid levels a sequence has, the less likely a loss will occur, but will lose more $ when it does.

Based on this information, next week I will take the middle road and trade only the pairs in Green for the 2x7 column. A 2x multiplier, with 7 grid levels (hedged). Trading 11 Pairs at the same time, I want to see what the average daily revenue could be to compare with the possible losses based on this data.

Don't want to rain on your parade ole sport

BUT it looks like rubbish to me.

Could be. Would not be the first EA that I've designed that failed outside of the backtest realm.

The EA should average about 1-3% gains per day. With a potential 20% Risk, it only need run a few days streak to recover. These figures do not account for any kind of compounding which will be part of live trading. .01 per $5000 of account balance.

Just to make sure my back tests did not come out as a fluke. I ran a 5 year back test on each one that did well on the 1 year back test. 2/28/2016 to Present.

5 Year Strategy Tester reports to follow...

I am not as concerned on single charts difference between starting balance and ending balance. But the time in between losses. The greater the distance, the more money that is pocketed across multiple pairs to generate daily income.