Minor issues with how to accurately track daily trade performance. Commercial solutions?

[Laissez Faire]

Hi all,

I'm repeating a post which probably got lost in another thread. I day trade the e-mini S&P 500 using Ninjatrader 8 and Ninjatrader Brokerage (Dorman Trading).

I track my trade performance through Ninjatrader's trade performance report, but I'm not entirely satisfied with that one, so I export the data to Excel and finally to my own trading journal/trading records.

Basically, I like to keep track of how many points per contract I'm able to extract from the markets per day. I like to track this on a per contract basis as that should give an objective measure regardless of position size.

However, certain problems present themselves when trading multiple contracts. For someone who's actively scaling in/out (I usually don't), I imagine it becomes even worse.

a. Different trading size throughout the day skewing the points per contract statistic

If you net 10 points on a single trade with 5 contracts, but then take a net loss of 10 points on 1 contract on your next trade, you're basically left with 0 points on a per contract basis, although the end result is

(10 x 5) - (10 x 1) = 40 points total

The alternative is to simply average those 40 points by the average number of contracts traded, i.e., 40 / (5 + 1)/2 = 6,7 points per contract.

But not quite correct either, is it?

Equally, if you lose 50 points on a 1 contract trade, but gain 10 points on 5 contracts on the next trade, you're left with a net loss of 40 points per contract, although you're breakeven on the day.

Here's an example from my older records, actually. As can be seen, I was up 6,50 points per contract on 3 contracts traded. But as I started trading 1 contract, I actually ended the day at 0 points per contract when each contract is summed separately, although the end result was net 13 points total.

It seems like simply averaging the end result by the maximum/average # of contracts may be the only viable solution for this particular problem.

b. A single trade is split into multiple trades:

Not always, but I frequently experience that multiple lots are split in multiple parts.

Yesterday, I netted 19 points per contract on a single trade on 4 contracts, but Ninjatrader splits this into one trade @ 3 contracts and one trade @ 1 contract.

I assume this is because these trades are crossed separately, although it happens in the same time interval.

So, while the result per contract was 19 points, my bookkeeping for the day would say I netted + 38 points per contract.

This one could easily be adressed if Ninja could aggregate all trades which have the same exit/entry price at the same time interval, but it clearly does not.

Does any other platform do this? Is there a trade software/trading journal which addresses this somehow?

It may be that I'm the only guy who keeps track like this, but if not, I'm wondering how you experienced ES traders deal with these issues.

PS: In the past I've tried to aggregate my trades manually, but it's very time consuming to do and I simply can't be bothered to do that. Alternatively, I'd have to make a script in VBA or something which does the aggregation after I've imported this to Excel.

Thanks in advance.

 

[formikatrading]

Left you a PM but you may want to test out Trademetria.

 

[murray t turtle]

Hi all,

I'm repeating a post which probably got lost in another thread. I day trade the e-mini S&P 500 using Ninjatrader 8 and Ninjatrader Brokerage (Dorman Trading).

I track my trade performance through Ninjatrader's trade performance report, but I'm not entirely satisfied with that one, so I export the data to Excel and finally to my own trading journal/trading records.

Basically, I like to keep track of how many points per contract I'm able to extract from the markets per day. I like to track this on a per contract basis as that should give an objective measure regardless of position size.

However, certain problems present themselves when trading multiple contracts. For someone who's actively scaling in/out (I usually don't), I imagine it becomes even worse.

a. Different trading size throughout the day skewing the points per contract statistic

If you net 10 points on a single trade with 5 contracts, but then take a net loss of 10 points on 1 contract on your next trade, you're basically left with 0 points on a per contract basis, although the end result is

(10 x 5) - (10 x 1) = 40 points total

The alternative is to simply average those 40 points by the average number of contracts traded, i.e., 40 / (5 + 1)/2 = 6,7 points per contract.

But not quite correct either, is it?

Equally, if you lose 50 points on a 1 contract trade, but gain 10 points on 5 contracts on the next trade, you're left with a net loss of 40 points per contract, although you're breakeven on the day.

Here's an example from my older records, actually. As can be seen, I was up 6,50 points per contract on 3 contracts traded. But as I started trading 1 contract, I actually ended the day at 0 points per contract when each contract is summed separately, although the end result was net 13 points total.

It seems like simply averaging the end result by the maximum/average # of contracts may be the only viable solution for this particular problem.

b. A single trade is split into multiple trades:

Not always, but I frequently experience that multiple lots are split in multiple parts.

Yesterday, I netted 19 points per contract on a single trade on 4 contracts, but Ninjatrader splits this into one trade @ 3 contracts and one trade @ 1 contract.

I assume this is because these trades are crossed separately, although it happens in the same time interval.

So, while the result per contract was 19 points, my bookkeeping for the day would say I netted + 38 points per contract.

This one could easily be adressed if Ninja could aggregate all trades which have the same exit/entry price at the same time interval, but it clearly does not.

Does any other platform do this? Is there a trade software/trading journal which addresses this somehow?

It may be that I'm the only guy who keeps track like this, but if not, I'm wondering how you experienced ES traders deal with these issues.

PS: In the past I've tried to aggregate my trades manually, but it's very time consuming to do and I simply can't be bothered to do that. Alternatively, I'd have to make a script in VBA or something which does the aggregation after I've imported this to Excel.

Thanks in advance.

%%
Sounds like /looks like from your data\\\\\\\\\ 8 trades is worse than 7.
I noticed something like that in my trading/ investing;
even though position size is fairly close\ the last scale in/higher price goofs my partial profits.
So less trading= better profit/or go ahead + get all out/exit, @ once, in one vehicle.
One exception is if I have a better than average profit/can scale out/fine as long as my last leveraged ETF price is profitable by itself...……………………………………………….…………………………………………………………………………………………………...I dont pay commissions but haver sec fees/slippage/sometimes no dividends.

 

[Bad_Badness]

I think most "journal" software are really based on the roundtrip of when the entry-entry. That is how they do the tax reporting so it is the simplest to report and required. Last month I reviewed them, TradeMetria etc. I found them too heavy weight for what I wanted right now. Also too much non-actionable data. I wanted to stick to my tactical issues, and not get lost in derivative statistics*.

An alternative is to look at the price average that is reported when there are multiple contracts. At that point they are effectively combined. I.e. P/L consolidates at various times during the actual trade. This information can be extracted via the API, but some coding is required to automate.

I gave up on the above API solutions as it seem to be more work and possible maintenance for not that much actionable information. The alternative is simply treate all trades as a P/L result when the position is closed. It is readily available, one P/L and one "trade". This allowed all the analysis of trades I wanted to be excel calculated easily.

You might consider what you actually want from the data and work back as a validation. Scoresheets are not as interesting as specific charts-statistics that help you stay on track in your tactics, imo.

Hope that helps.

* Most journal packages give parametric statistics. If you want non parametric statistics (ranking based on properties e.g.), you have to roll your own. I wanted non parametric data to help my tactical execution.

 

[Fonz]

I use journalytix from jigsaw; it's independent from jigsaw trading platform.
It is accurate and works in reel time, which is terrific: no need to wait the end of the day to import your trades.

I also recently discovered tradersync which looks amazing. I didn't check yet all the features.

 

[Laissez Faire]

I use journalytix from jigsaw; it's independent from jigsaw trading platform.
It is accurate and works in reel time, which is terrific: no need to wait the end of the day to import your trades.

I also recently discovered tradersync which looks amazing. I didn't check yet all the features.

Thanks.

May I ask how it adresses the two main issues mentioned? Does it aggregate trades that are splitted?

Would you mind posting a screenshot of a daily summary? Feel free to PM me also if you don’t want to post in public.

Thanks.

 

[tradethetrade]

You will want to track both per points and per points taking into account quantity.

Here is how Trademetria does it:

Shows total points and total points taking into account the number of contracts.

It also lets you see points per trade

You can also view trades in aggregate mode.

 

[Fonz]

Thanks.

..
Would you mind posting a screenshot of a daily summary? Feel free to PM me also if you don’t want to post in public.

Thanks.

I think you can test Journalytix for one or two weeks.
I attached 3 screenshots.
I really like the reel time trade recording and also other functionalities like economic news remainders.

I will soon try (one week free) Tradersync. It looks like we could have more analysis from our trades which what I was looking for with Journalytix.

Hope this helps.

 

[Fonz]

I just tried Tradersync: Apparently not for Futures which is a no go for me.
Also, it is still a "manual trade or import" function. I will keep the reel time trade recording which is just amazing with Journalytix.
Best!

 

[Laissez Faire]

I just tried Tradersync: Apparently not for Futures which is a no go for me.
Also, it is still a "manual trade or import" function. I will keep the reel time trade recording which is just amazing with Journalytix.
Best!

Thanks, Fonz. And thank you for those pictures.

I'll see what I do about this.

MFE/MAE are interesting statistics, too, but the raw data from Ninja is useless to me. Why?

Because if I have a 5 contract trade and the market goes against me by 5 points - the MFE = 5 x 5 = 25 points.

That's why I need to have this data normalized on a per contract basis. This is what I do myself in Excel, but there's still some issues. Particularly when trades are split up.