CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. 64 % of retail investor accounts lose money when trading CFDs with this provider. You should consider whether you understand how CFDs work and whether you can afford to take the high risk of losing your money.

Experience Metric Suggestion

I’ve been having a think about the Experience metric and thought I’d float some ideas I’ve had. Just a disclosure. I massively dislike this metric, as I’ll explain, but I have been an active trader for 7 years and an active automated trader for almost 4 years yet when I start a new Darwin I have experience of 0. This I feel unfairly penalises people who do the right thing and start a new account for a new strategy rather than just re-use another account to get an instantly high experience score. This is what’s made me think of the suggestions below.

Firstly a rename, I don’t think ‘Experience’ is a good representation of what it really means. It is a good metric for showing a strategy has been working over time but I don’t think ‘experience’ really describes it well. I was going to suggest ‘activity’ instead as I feel that is a more accurate representation of what it is really showing, market activity over a certain time period.

Secondly, make it a two-way metric. Currently experience can only go up, as a result anyone can migrate any old account they’ve been using for a few months and get an instantly high score even if it wasn’t the same strategy, the results will be different but it will improve their overall scores. My suggestion ties in with ‘activity’ where if the account is inactive for say 1 month, every month after that it looses 1 experience score to show it is inactive. I would measure inactivity as no existing positions and no new positions in a month. It’s rare but some accounts can be inactive for a period of time but still keep high Experience scores.

Hope I’ve explained that and I’m looking foreword to see what everyone would think, even if everyone disagrees with me :laughing:

Cheers

Tom

2 Likes

You are right but it would be very difficult to implement.
The right name would be significance.

3 Likes

@TraderTom Brilliant suggestion and spot on with the name change for Activity - [quote=“TraderTom, post:1, topic:1863”]
market activity over a certain time period
[/quote]
Off the bat I would assume that “Experience” to the majority of viewers would speak to the length of the track record so automatically a 5 yr track record would get higher rating than a 6 month track record but unto the “two way metric” it could be something like using the average daily trade frequency of the entire track record as a measure and computing daily trade activity which would add/minus points to the “Experience” score daily.
Eg Trader A has a track record of 12 months with an “average daily trade frequency of 10 trades”…days that less than 10 trades are taken should minus points and days when 10 or more trades are taken adds points to the “Activity” metrics.

Just my thoughts…

2 Likes

I like this too, then it kind of works a bit more like the ‘activity’ section on Darwina where the activity is calculated, this from the description of activity in Darwina:

"Traders are penalized when their monthly market exposition is lower than their average market exposition in the previous 3 months.

Market exposition for Activity gets calculated in the same way it gets calculated for the Experience investable attribute."

1 Like

To add to your suggestion of the revision of the “Experience” there should also be some type of “pie chart” to decipher the type of activity revolving around “Scalp”,“Intra-session”,“Intra-day”,“Intra-week”, “Swing”, “Position” and I am guessing there will be varying views as to what consist of each trading style but I think a general concensus from the community would suffice and concluded quite quickly.

Now I was reading earlier on a forum how the algo don’t really favour different trading styles (will try to find the old link). I don’t think should be the case unless if the risk management is inconsistent.
I think it was a case of having a trade open with 50 pip vs having a trade open with 10 pips with the same risk and being penalized.

My argument is a versatile trader who can manage a portfolio with dynamic trading styles adds further to the “diversification” of that portfolio in itself and should be added value instead of being penalized on condition that the risk management is consistent

1 Like

What you are writing in the last post is more a problem of consistency algos.

This is a great idea - and additional it could be shown when the last trade was opened and the last trade was closed.
To make it perfect the traer should als be able to leave there when he is on vacation and does not trade because of his absence until dd/mm/yyyy. :slight_smile:

2 Likes

Who are you referring to @CavaliereVerde?

If you trade more markets and timeframes you dont’get penalyzed in Experience but in Consistency (R+ R- Dc)
Consistency should state if someone continuously change behaviour but unfortunatelly it isn’t so, it just penalyzes people using trailingstops or breakevens.
BTW if you are well difversified it is an advantage for Experience because you will make more trades.

Inactivity of systems should be penalyzed but it doesn’t reduce the statistical significance of the trackrecord.
There are many topics here on how to penalyze inactivity but it seems that Darwinex don’t want to do it.
https://community.darwinex.com/tags/communication

Another subject is the abuse of migration.
IMO it’s better not to mix too many subjects.
Also over-ruling can be poisoning, simple systems work even if not perfect. :wink:

I personally think everything you’ve said is kind of off-topic and unrelated? The issue I have, and the suggestion I’ve made, is entirely about the experience metric and nothing to do with consistency.

Let me give an example. I have been trading for 7 years and automated trading for 4 years, so I would say I meet the criteria of an ‘experienced trader’ at least in my view. I have opened a new account recently for a strategy I’ve been building for a while, I thought it was right to open a new account for it rather than migrate or use an existing account due to the fact I’ve made changes to better fit Darwinex vs the previous broker I used. Because of this I’m going to start with an Experience of 0 and as a result be penalised on scores because I’m ‘inexperienced’ eventhough I’m a trader of 7 years and this system has been active for almost 2 years.

Now I could use my existing TSC darwin, instantly listed without needing to wait a month and existing scores but I don’t think that fits the morals of the program. Equally I could migrate but from testing the rests between brokers are different so I didn’t feel this was right either.

Take all this into account and by me doing what I believe to be the right thing to not falsify anything I’m serving investors as I would like I’m going to be listed as inexperienced which I don’t agree with. I think it could be named differently and definitely shouldn’t be a 1 way metric imo.

1 Like

Yes I remember perfectly your case! :smiley:


If you aim for a “quick score” you can paste a trackreckord done with a different strategy but a smart investor will notice it and it will worsen your credibility.
It depends of your target, being advanteged tomorrow or being consistent in 2 years… :wink:

1 Like

Fully agree, this is exactly why I’ve done what I think is right and started an entirely new account just for the current iteration of this design.

However people will always abuse the system, especially when it is this easy to abuse to start appearing in filters or Darwina etc.

Thats my main issue, its easily abused and open to abuse by anyone. Also the fact this metric is one of the heaviest weighted, it impacts overall score, loads of filters and Darwina very heavily I think it distorts the platform somewhat.

1 Like

Yes but the solution is working on consistecy algos to make that really spotting a consistent ruleset, not just pepole closing always 20 pips…

The other correction to do is to limit the experience coming from a migrated trackrecord.

1 Like

Agreed, if experience is going to stay as it is at least removing experience for migrated accounts would help. But then on the other side would that reduce the number of people migrating accounts from other brokers to Darwinex? Something to consider.

If it was removed for migration and the consistency algo’s penalized changes heavily that might help the overall issue that I think exists. Those 2 combine would help stop anyone re-using accounts with different strategies to get good scores, however the issue with that is strategies that use machine learning/deep learning or similar and constantly adjust themselves (as mine do) might be seen by the consistency algo to be inconsistent because they keep adjusting their own parameters? Might be something else to consider?

2 Likes

We cannot pretend consistency to be compliant with machine learning while it is fooled by a simple trailing stop… :smiley: :smiley: :smile_cat:

Migration is a necessary evil because Darwinex needs traders.

1 Like