When to take your EA to a live account? | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

demo account and live account differ in slippage? FXCM 15 replies

Dukascopy Live Account - Don't Post if you haven't traded live with them! 117 replies

My Live Trades on my Live Account 588 replies

3 week demo account leading up to live account. 20 replies

Demo account versus live account. 19 replies

When to take your EA to a live account?

  • Poll

  • How many trades until you can switch from demo to live account?
  •  
  •  
  •  
  •  
  •  
  •  
  • #1
  • First Post: May 6, 2020 1:29pm May 6, 2020 1:29pm
How long should you test a demo account before going live with an EA? Should it be 50 trades, 100 trades, ect?
  • #2
  • May 6, 2020 2:02pm May 6, 2020 2:02pm
demo acc is usually bad not only because you don't lose hard earned money but because you don't get slipped and there are no swaps. I know only one broker that give you real market conditions and of course ppl complain about that lol
  • #3
  • May 6, 2020 2:10pm May 6, 2020 2:10pm
Quoting HeyYou
demo acc is usually bad not only because you don't lose hard earned money but because you don't get slipped and there are no swaps. I know only one broker that give you real market conditions and of course ppl complain about that lol
Oh I get it. So I should have clarified abit. My trades are never open at the end of the day so no swaps as I don't carry overnight. I also am using an EA so I can take the emotion out of the trading. The goal of my mindset and EA has been to gain pips, with that I can not think about the 'hard-earned money' and focus on winning trades. I want to forward test, however I have blown up many accounts in the past and am trying to different path. I have back tested 2 years worth of data on 6 pairs, all showing similar results to how the EA is performing in a demo account. Would this not translate to live account?
  • #4
  • May 6, 2020 2:13pm May 6, 2020 2:13pm
Is this thread about going live with an EA after demo trading? If a demo trader wants to live trade, I doubt if he or she should start with an EA. Doing that may be a recipe for disaster.

KP

Quoting JoeN017
How long should you test a demo account before going live with an EA? Should it be 50 trades, 100 trades, ect?
Do your homework, follow the footprints of smart money
  • #5
  • May 6, 2020 2:20pm May 6, 2020 2:20pm
Quoting KeenPips
Is this thread about going live with an EA after demo trading? If a demo trader wants to live trade, I doubt if he or she should start with an EA. Doing that may be a recipe for disaster. KP {quote}
Yes, sort of. I am asking if I am using my EA on a demo account, when would it be safe to try on a live account? I know there are differences between a demo vs live account however due to my work schedule it is more efficient to open a trade with set parameters that I would be adhering to as if I was live trading. I close every open trade before the end of the day so I do not have to account for swaps as I am not holding overnight.
  • #6
  • May 6, 2020 2:27pm May 6, 2020 2:27pm
That means that you are not a newbie trader. You are using the demo to test your EA? No newbie will be thinking of using an EA to learn Forex trading; for one thing, he or she will need to have some level of skill or knowledge to be able to decide which EA to use. At any rate, I hope you get the answer you want.

KP

Quoting JoeN017
{quote} Yes, sort of. I am asking if I am using my EA on a demo account, when would it be safe to try on a live account? I know there are differences between a demo vs live account however due to my work schedule it is more efficient to open a trade with set parameters that I would be adhering to as if I was live trading. I close every open trade before the end of the day so I do not have to account for swaps as I am not holding overnight.
Do your homework, follow the footprints of smart money
  • #7
  • May 6, 2020 2:35pm May 6, 2020 2:35pm
Quoting KeenPips
That means that you are not a newbie trader. You are using the demo to test your EA? No newbie will be thinking of using an EA to learn Forex trading; for one thing, he or she will need to have some level of skill or knowledge to be able to decide which EA to use. At any rate, I hope you get the answer you want. KP {quote}
No I have been trading off and on for 10+ years, learned to make EAs about 3 years ago. I feel as if I have finally made an EA that can do exactly what I want it to do and have been running on a demo. I appreciate the input and am thinking after 100 trades, if the EA has not made any mistakes and is profitable then I will run it on a live account with the expectation that it will be successful based on my backtesting and demotesting.
  • #8
  • Edited 2:58pm May 6, 2020 2:39pm | Edited 2:58pm
Maybe you should edit your thread title, perhaps using the first part of the first sentence in your post #1 - specifically mentioning EA.

Trade safe.

KP

Quoting JoeN017
{quote} No I have been trading off and on for 10+ years, learned to make EAs about 3 years ago. I feel as if I have finally made an EA that can do exactly what I want it to do and have been running on a demo. I appreciate the input and am thinking after 100 trades, if the EA has not made any mistakes and is profitable then I will run it on a live account with the expectation that it will be successful based on my backtesting and demotesting.
Do your homework, follow the footprints of smart money
  • #9
  • May 6, 2020 2:57pm May 6, 2020 2:57pm
Quoting JoeN017
{quote} Oh I get it. So I should have clarified abit. My trades are never open at the end of the day so no swaps as I don't carry overnight. I also am using an EA so I can take the emotion out of the trading. The goal of my mindset and EA has been to gain pips, with that I can not think about the 'hard-earned money' and focus on winning trades. I want to forward test, however I have blown up many accounts in the past and am trying to different path. I have back tested 2 years worth of data on 6 pairs, all showing similar results to how the EA is...


and what about slippage ? don't underestimate it.

it costed me 1 year and a half of forwardtesting.

when you trade low frequency and you lose 100-150% more on a single trade... it's really hard to profit
  • #10
  • May 6, 2020 2:58pm May 6, 2020 2:58pm
An apt title

Stay safe, trade safe.

KP
Do your homework, follow the footprints of smart money
  • #11
  • May 6, 2020 3:02pm May 6, 2020 3:02pm
Looking at your trade explorer, I personally believe it would be okay for you to go live with a small account. If month after month, forward-testing continues to go as you expected, add a bit more capital over time.

The reason why I believe it's okay for you is because your strategy is not a scalping strategy. It aims for a decent amount of pips. Also, seeing that you back-tested 2 years worth of data on several pairs, it seems like you did your research. Back-testing more would be even better to know the potential dangers of the strategy, but I think you're off to a good start. (More forward-testing on a demo account never hurts though.)

Some people believe demo trading/back-testing is useless, but I personally believe it is extremely useful. If you can't even profit consistently with fake money, how are you going to profit consistently with real money. As long as you're risking/aiming for a decent amount of pips, I do not think there will be a large discrepancy in the results between live accounts and demo accounts. Once you start trading huge lot sizes, maybe you will start seeing more discrepancies due to larger slippage, but until then, I would expect you to have similar results with both live and demo.

If you were scalping, back-testing and demo trading may not mean as much. You should not put as much faith into a scalping strategy on demo/back-testing. Spreads/slippage impact your trade much more if you're a scalper. Of course, this does not mean you should test a scalping strategy live right away. I still think you need to do quite a lot of testing in demo/back-testing whatever the strategy may be, but for scalping strategies, you should put more emphasis on live forward-testing.

If you hold trades over the weekend, there may be some discrepancies there too, but you already mentioned it's an intraday strategy.
  • #12
  • May 6, 2020 3:06pm May 6, 2020 3:06pm
Quoting HeyYou
{quote} and what about slippage ? don't underestimate it. it costed me 1 year and a half of forwardtesting. when you trade low frequency and you lose 100-150% more on a single trade... it's really hard to profit
My EA has max slippage set so if it is greater than what I want then the trade won't be executed. Might not get into a trade I want however it could limit what you experienced.
  • #13
  • May 6, 2020 3:15pm May 6, 2020 3:15pm
Quoting autemfx
Looking at your trade explorer, I personally believe it would be okay for you to go live with a small account. If month after month, forward-testing continues to go as you expected, add a bit more capital over time. The reason why I believe it's okay for you is because your strategy is not a scalping strategy. It aims for a decent amount of pips. Also, seeing that you back-tested 2 years worth of data on several pairs, it seems like you did your research. Back-testing more would be even better to know the potential dangers of the strategy, but I...
Thank you for the input autemfx! I agree that more research is needed and while the results in backtest and demo so far are good, there could still be something that I am missing. I completely agree that you need to be able to make money on a regular basis in a demo account before attempting a live account. My thought in going into a live account is to get more data that I can compare to back test and demo account before putting more in a scaling up. My reason for this thread is to get confirmation that I am ready to start this EA on a live account. I really appreciate your insight!
  • #14
  • May 7, 2020 2:52am May 7, 2020 2:52am
Quoting JoeN017
{quote} My EA has max slippage set so if it is greater than what I want then the trade won't be executed. Might not get into a trade I want however it could limit what you experienced.
still, you are not experiencing slippage now.

no offence, I don't understand why forwartesting on demo... it's practically the same of using the strategy tester.

I experienced massive slippage most of the time btw, it's just unavoidable when trading Gold. but maybe it's no big deal on major pairs
  • #15
  • May 7, 2020 4:36am May 7, 2020 4:36am
Quoting HeyYou
{quote} still, you are not experiencing slippage now. no offence, I don't understand why forwartesting on demo... it's practically the same of using the strategy tester. I experienced massive slippage most of the time btw, it's just unavoidable when trading Gold. but maybe it's no big deal on major pairs
So I’ve built a lot of EAs, based on a variety of strategies. All the EAs open and close trades properly however the strategies were lacking or missing something. I have had EAs post amazing results in the strategy tester with 99% accuracy and then when I launch it live it does not do anything remotely close to the same, hence the wanting to forward test in a demo. It allows me to see if there is anything I missed in my coding, something I didn’t account for, or just plain to make sure I feel comfortable with the EA before putting it on a live account. My question on this thread is simple, when is it ok to launch in a live account if the EA has shown proper data in a demo account. Besides, what is the harm in forward testing on a demo? It just gives you more data to analyze that you couldn’t get from the backtesting. When you backtest you can do years or decades of testing, when you forward test it can take years to get the same data.

You seem very concerned about slippage and appreciate the warning on it. I will take into consideration and if I ever trade gold I will certainly keep a close eye on it. Thanks for the input.
  • #16
  • May 7, 2020 4:47am May 7, 2020 4:47am
Joe,
dont bother with much demo forward testing.
open a 200$ live acct, say -30% max dd the, and you are out, stopping this experiment.
max risk: 60$
cost of cinema tickets and fast food dinner for 2 in the mall. an evening out.
if you outlast 4 hours on live acct, the experience worth it already.
limited downside, with the upside that if it is working you can gradually add more funds to your live broker account.

seriously, you demo forward ( which of course never ever equal the live conditions) for weeks, maybe months...to never learn the real potential or lack of the EA.
at some point you will realize that your time ( the hours and days invested into this demo forwarding) worth more than say 60$.
better to realize that sooner than later
there is always, always another trade!!
  • #17
  • May 7, 2020 5:12am May 7, 2020 5:12am
Quoting JoeN017
How long should you test a demo account before going live with an EA? Should it be 50 trades, 100 trades, ect?
3 months on demo and just study the draw down figures. On Demo if it runs over 10% in draw down, delete it.
On live charts the EA will run at double the draw down figures from demo charts, if you are lucky.

And don't get me started on the Mt4 Strategy tester, it is totally inaccurate. You will get your hopes up and get crushed on live charts.

Spreadsheets is the answer with live data downloaded onto your own testing platform
Trading thin liquidity at the boundary of the charts
  • #18
  • May 7, 2020 6:09am May 7, 2020 6:09am
Quoting JoeN017
{quote} So I’ve built a lot of EAs, based on a variety of strategies. All the EAs open and close trades properly however the strategies were lacking or missing something. I have had EAs post amazing results in the strategy tester with 99% accuracy and then when I launch it live it does not do anything remotely close to the same, hence the wanting to forward test in a demo. It allows me to see if there is anything I missed in my coding, something I didn’t account for, or just plain to make sure I feel comfortable with the EA before putting it on...
it makes sense
  • #19
  • May 7, 2020 6:29am May 7, 2020 6:29am
about your question.. IDK, maybe 200 trades, since the frequency of trades is high (56 trades in 2 weeks).

200 trades for a low frequency trader like me would mean >2 years of demo just to see if the ea works lol
  • #20
  • May 7, 2020 7:53am May 7, 2020 7:53am
Quoting HeyYou
about your question.. IDK, maybe 200 trades, since the frequency of trades is high (56 trades in 2 weeks). 200 trades for a low frequency trader like me would mean >2 years of demo just to see if the ea works lol
Yes, this may sound odd, the frequency of trades seem high due the fact that I dont hold any positions overnight, so every trade is closed out before the end of the day. Most of the time, if the trend and indicators are still moving in the direction from the previous day than it will open another trade in the same direction from the prior day, allowing the EA to follow the long term trends in intraday movements. So 5 days of trading the same pair could be the same movement equally one large 5 day trade, or if the trend reverses then it checks that each day and won't get back into the following day. I would say 56 trades is actually more like 15-18 longer trades broken up. Everything is based on weekly and daily charts for entry. The trick with an EA is always figuring out when to close out a trade, positive or negative. I think I am going to try the EA on a small live account next week and let the demo keep running for comparison.
0 traders viewing now