NT8 Performance Analysis Thoughts - futures io
futures io



NT8 Performance Analysis Thoughts


Discussion in NinjaTrader

Updated
      Top Posters
    1. looks_one jflaggs with 9 posts (0 thanks)
    2. looks_two ShatteredX with 5 posts (1 thanks)
    3. looks_3 Neo1 with 3 posts (0 thanks)
    4. looks_4 xiaosi with 1 posts (0 thanks)
    1. trending_up 2,732 views
    2. thumb_up 8 thanks given
    3. group 6 followers
    1. forum 20 posts
    2. attach_file 3 attachments




Welcome to futures io: the largest futures trading community on the planet, with well over 125,000 members
  • Genuine reviews from real traders, not fake reviews from stealth vendors
  • Quality education from leading professional traders
  • We are a friendly, helpful, and positive community
  • We do not tolerate rude behavior, trolling, or vendors advertising in posts
  • We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community.  It's free and simple.

-- Big Mike, Site Administrator

(If you already have an account, login at the top of the page)

 
Search this Thread
 

NT8 Performance Analysis Thoughts

(login for full post details)
  #1 (permalink)
jflaggs
Los Angeles + California/United States
 
 
Posts: 17 since Aug 2017
Thanks: 7 given, 0 received

Hello,

We are developing a strategy that, so far, has shown promise. However, we are having trouble getting around spread commissions. We're curious what type of performance others are looking for in a "good" strategy so that we can start to understand how this strategy compares.

At first glance, we were happy with the performance analysis...until I realized that after commissions we are essentially at break even regardless of the decent performance results. The image shown is for 4 months of market replay on USDJPY and I am calculating commissions based on a constant spread of 1.4p. All trades are sized at 1 lot.

So my questions stands, how good does a strategy need to be in order to beat commissions evilness? Placing just 1.85 trades per day on average has earned me total spread commissions nearly equal to my net profit for 4 months. Your thoughts are appreciated.

Attached Thumbnails
Click image for larger version

Name:	pips.PNG
Views:	145
Size:	26.6 KB
ID:	239933   Click image for larger version

Name:	usd.PNG
Views:	107
Size:	24.1 KB
ID:	239934  
Reply With Quote

Journal Challenge April 2021 results (now extended!):
Competing for $1800 in prizes from Jigsaw
looks_oneMaking a Living with the Microsby sstheo
(107 thanks from 17 posts)
looks_twoSalao's Journalby Salao
(33 thanks from 8 posts)
looks_3Deetee’s DAX Trading Journal (time based)by Deetee
(28 thanks from 11 posts)
looks_4Learning to Profit - A journey in algorithms and optionsby Syntax
(14 thanks from 9 posts)
looks_5Maybe a little bit different journalby Malykubo
(9 thanks from 8 posts)
 
Best Threads (Most Thanked)
in the last 7 days on futures io
Would You Sell Your System?
73 thanks
The Crude Dude Oil Trading System
44 thanks
Big Mike in Ecuador
42 thanks
The New Micro Contract - MICRO BITCOIN coming May 2021
25 thanks
futures io site changelog and issues/problem reporting
24 thanks
 
(login for full post details)
  #3 (permalink)
 choke35 
Germany
 
Experience: Intermediate
Platform: Other
Trading: ES, YM, 6E
 
Posts: 2,668 since Feb 2013
Thanks: 5,100 given, 6,555 received


If you didn't (over)optimize the system before, the results are not bad.

One of your next steps could be a deeper look into the MAE / MFE profile of your trades.
Average losses are (very) high compared with the average wins.

If you can cut some relatively large losing trades without losing too many of your winners,
that could solve much of your commission problem.

Reply With Quote
 
(login for full post details)
  #4 (permalink)
 xiaosi 
Brisbane, Queensland, Australia
 
Experience: Advanced
Platform: JIGSAW/SIERRA CHART
Broker: MacQuarie Futures/AMP Clearing/CQG
Trading: HHI, HSI, FDAX
 
xiaosi's Avatar
 
Posts: 502 since Feb 2012
Thanks: 448 given, 533 received

I heard one time a good rule of thumb was to have a minimum average trade of greater than $50 usd after slippage and commission. This was for intraday systems. Makes sense to me as a hurdle.

Follow me on Twitter Reply With Quote
 
(login for full post details)
  #5 (permalink)
jflaggs
Los Angeles + California/United States
 
 
Posts: 17 since Aug 2017
Thanks: 7 given, 0 received

choke35, xiaosi,

Thanks for your comments. There is no back testing optimization done here. I will look into improving the MAE/MFE profiles. I can limit the risk a bit more, and possibly take more profits, but that will decrease the 80%+ accuracy that I enjoy seeing. No free lunch right?

Xiaosi, very good point. In fact, exactly half of these trades are simply breaking even with themselves after commissions by design. I will need to take more profit than that.

Would it make a difference to trade a different instrument or market? I haven't looked into it much, but I know that I could get lower commissions with a similar return if I trade stocks because I can get a flat fee of $10 round trip vs the $30 I am paying for FX. Any insight on how futures commissions would compare? Or another instrument?

Reply With Quote
 
(login for full post details)
  #6 (permalink)
 TWDsje   is a Vendor
 
 
Posts: 632 since Apr 2016
Thanks: 19 given, 672 received

Here's how I think about this in my mind. All strategies can be plotted on a grid with probability on the x axis, and average win/average loss on the y axis. The center of the graph will be 50% for probability, and 1 for where your average win is equal to your average loss. Something like this:



The further away from the center of the grid we get, the less common that trade is. We don't want our strategy to be too far from the center or we won't make as much due to too many trades. We also can't be too close to the center or we lose money on commissions.

Most strategies that work for retail traders fall along the two axises:

The first group follows the Y axis. These are strategies where your average win is twice as big as your average loss or more. Such trades could have a probability between 35-55% win rate. Such strategies tend to be more trend based as in order to make twice as much as your risk you'll have to catch some sort of trending move.

The other group follows the X axis with trades that have a higher probability. These are scalping or mean revision types of strategies. These trades tend to have a probability of between 60-80%. The average profit / average loss should probably be above .5, although I suggest looking for trades where the risk/reward are equal.

There's also the big banks area. These trades are probably unicorns, but certainly exist. If you have a trade like this you probably trade someone else's money.

So you fit more in the scalping type of category, but you're just under the break even line. Your average loss compared to your average win is too high. I suggest trying to get that ratio closer to 1:1.

Reply With Quote
The following 7 users say Thank You to TWDsje for this post:
 
(login for full post details)
  #7 (permalink)
 ShatteredX 
Houston, TX
 
Experience: Intermediate
Platform: Python
Trading: NQ
 
Posts: 97 since Apr 2016
Thanks: 130 given, 93 received

Let's see those performance numbers after you add in commission and slippage :-)

Reply With Quote
 
(login for full post details)
  #8 (permalink)
jflaggs
Los Angeles + California/United States
 
 
Posts: 17 since Aug 2017
Thanks: 7 given, 0 received

TWDsje,

That's an awesome graphical representation of the problem and how to work towards solving it. Thank you very much for the good input, Yes, mathematically you're absolutely correct in that we should be at 1:1 for the accuracy we're seeing in order to make it out of the commissions band in your area plot. I think that this is a good generalization--much better than simply trying to trade an instrument or market with lower commissions.


ShatteredX,

Slippage is set to 2p which is fairly realistic, so it's the commissions that are troubling. The performance numbers will be the same after commissions are added, but the net profit will be much less. This tells me that it has potential, but needs some work in the win:loss ratio as others have pointed out in this thread.

Reply With Quote
 
(login for full post details)
  #9 (permalink)
 Neo1 
Christchurch, New Zealand
 
Experience: Beginner
Platform: SC
Broker: IB, BC, Dx Feed
Trading: US Equities
 
Neo1's Avatar
 
Posts: 428 since Jul 2014
Thanks: 471 given, 501 received


jflaggs View Post
choke35, xiaosi,

Thanks for your comments. There is no back testing optimization done here. I will look into improving the MAE/MFE profiles. I can limit the risk a bit more, and possibly take more profits, but that will decrease the 80%+ accuracy that I enjoy seeing. No free lunch right?

Xiaosi, very good point. In fact, exactly half of these trades are simply breaking even with themselves after commissions by design. I will need to take more profit than that.

Would it make a difference to trade a different instrument or market? I haven't looked into it much, but I know that I could get lower commissions with a similar return if I trade stocks because I can get a flat fee of $10 round trip vs the $30 I am paying for FX. Any insight on how futures commissions would compare? Or another instrument?

$10 per round trip? sounds expensive, unless you are trading 1000s on units per trade?

"Free markets work because they allow people to be lucky, thanks to aggressive trial and error, not by giving rewards or incentives for skill. The strategy is, then, to tinker as much as possible and try to collect as many Black Swan opportunities as you can"
Reply With Quote
 
(login for full post details)
  #10 (permalink)
jflaggs
Los Angeles + California/United States
 
 
Posts: 17 since Aug 2017
Thanks: 7 given, 0 received



Neo1 View Post
$10 per round trip? sounds expensive, unless you are trading 1000s on units per trade?

That's the commissions fee for TradeStation. Either .01/share or flat 5 bucks. I wouldn't be interested in trading less than 500 shares for the sub $300/share stocks that I follow, so 5 bucks in and 5 bucks out is the better deal. Which platform allows you to trade 500+ shares for less than that? I haven't looked around much yet.

Reply With Quote
 
(login for full post details)
  #11 (permalink)
 Neo1 
Christchurch, New Zealand
 
Experience: Beginner
Platform: SC
Broker: IB, BC, Dx Feed
Trading: US Equities
 
Neo1's Avatar
 
Posts: 428 since Jul 2014
Thanks: 471 given, 501 received


jflaggs View Post
That's the commissions fee for TradeStation. Either .01/share or flat 5 bucks. I wouldn't be interested in trading less than 500 shares for the sub $300/share stocks that I follow, so 5 bucks in and 5 bucks out is the better deal. Which platform allows you to trade 500+ shares for less than that? I haven't looked around much yet.

How big is your account size?

So you will be taking on positions with a notional value of 100k plus? If so, then the results are going to look nothing like the stats that you have posted.


IB fixed rate comms start from 0.005 per share( min $1).

"Free markets work because they allow people to be lucky, thanks to aggressive trial and error, not by giving rewards or incentives for skill. The strategy is, then, to tinker as much as possible and try to collect as many Black Swan opportunities as you can"
Reply With Quote
 
(login for full post details)
  #12 (permalink)
jflaggs
Los Angeles + California/United States
 
 
Posts: 17 since Aug 2017
Thanks: 7 given, 0 received


Neo1 View Post
If so, then the results are going to look nothing like the stats that you have posted.

How so?

I will be trading between 1 and 10 leverage. No more than that. Will definitely have to check out IB!

Reply With Quote
 
(login for full post details)
  #13 (permalink)
 Neo1 
Christchurch, New Zealand
 
Experience: Beginner
Platform: SC
Broker: IB, BC, Dx Feed
Trading: US Equities
 
Neo1's Avatar
 
Posts: 428 since Jul 2014
Thanks: 471 given, 501 received


jflaggs View Post
How so?

Because your Avg time in a market was 121min. A popular sub $300 stock could move $1-2 within that period. If your min position is 500 units and the stock is moving $1-2 dollars, you could be +/- $500-1k. So your Avg winning/ losing trade is going to be much greater than what you have reported.


jflaggs View Post
I will be trading between 1 and 10 leverage. No more than that. Will definitely have to check out IB!

How do you intend to get 10:1 margin from Trade Station? I assume your account size must be in the low six figures.

"Free markets work because they allow people to be lucky, thanks to aggressive trial and error, not by giving rewards or incentives for skill. The strategy is, then, to tinker as much as possible and try to collect as many Black Swan opportunities as you can"
Reply With Quote
 
(login for full post details)
  #14 (permalink)
jflaggs
Los Angeles + California/United States
 
 
Posts: 17 since Aug 2017
Thanks: 7 given, 0 received


Neo1 View Post
Because your Avg time in a market was 121min. A popular sub $300 stock could move $1-2 within that period. If your min position is 500 units and the stock is moving $1-2 dollars, you could be +/- $500-1k. So your Avg winning/ losing trade is going to be much greater than what you have reported.
How do you intend to get 10:1 margin from Trade Station? I assume your account size must be in the low six figures.

I see your point. 121min is not by any means a fixed number. The range is more like 1min to several hours. I don't plan to use TradeStation, I just used it as a reference to compare my options for commissions. Thanks for your input.

Reply With Quote
 
(login for full post details)
  #15 (permalink)
 ShatteredX 
Houston, TX
 
Experience: Intermediate
Platform: Python
Trading: NQ
 
Posts: 97 since Apr 2016
Thanks: 130 given, 93 received


jflaggs View Post
ShatteredX,

Slippage is set to 2p which is fairly realistic, so it's the commissions that are troubling. The performance numbers will be the same after commissions are added, but the net profit will be much less. This tells me that it has potential, but needs some work in the win:loss ratio as others have pointed out in this thread.

The performance numbers will be worse if you add commissions. Less net profit will mean lower Sharpe, profit factor, avg. trade, etc. Am I wrong here? I am a futures trader so my assumptions might be wrong regarding FX.

Back to your original post, any strategy that actually makes a net profit in live trading could be a good strategy. Personally, I calculate a reward/risk ratio to determine whether a strategy is worth trading live or not.

Reply With Quote
 
(login for full post details)
  #16 (permalink)
jflaggs
Los Angeles + California/United States
 
 
Posts: 17 since Aug 2017
Thanks: 7 given, 0 received


ShatteredX View Post
The performance numbers will be worse if you add commissions. Less net profit will mean lower Sharpe, profit factor, avg. trade, etc. Am I wrong here? I am a futures trader so my assumptions might be wrong regarding FX.

Back to your original post, any strategy that actually makes a net profit in live trading could be a good strategy. Personally, I calculate a reward/risk ratio to determine whether a strategy is worth trading live or not.

Hmm, well even when including commissions in NT8 Strategy Performance, the other parameters are not affected. So I think that it is measuring your strategy's fundamental performance characteristics. Yes, if commission were factored into each trade, I think that it would be much worse. I agree, we have lots of work to do to get my risk/reward ratio up!

Reply With Quote
 
(login for full post details)
  #17 (permalink)
 ShatteredX 
Houston, TX
 
Experience: Intermediate
Platform: Python
Trading: NQ
 
Posts: 97 since Apr 2016
Thanks: 130 given, 93 received


jflaggs View Post
Hmm, well even when including commissions in NT8 Strategy Performance, the other parameters are not affected. So I think that it is measuring your strategy's fundamental performance characteristics. Yes, if commission were factored into each trade, I think that it would be much worse. I agree, we have lots of work to do to get my risk/reward ratio up!

I just tested one of my strategies to see if including commissions changed profit factor and Sharpe Ratio.

Including comissions:
PF: 1.38
Sharpe: 0.70

Not including commissions:
PF: 1.40
Sharpe: 0.72

To get commissions to be included properly, you have to edit your Sim101 account in the "Accounts" tab in NT8 and select a commissions template.

This strategy only trades about once a day so commissions don't affect it much. I use market orders so slippage is a much bigger concern for me.

Reply With Quote
The following user says Thank You to ShatteredX for this post:
 
(login for full post details)
  #18 (permalink)
jflaggs
Los Angeles + California/United States
 
 
Posts: 17 since Aug 2017
Thanks: 7 given, 0 received


ShatteredX View Post
I just tested one of my strategies to see if including commissions changed profit factor and Sharpe Ratio.

Including comissions:
PF: 1.38
Sharpe: 0.70

Not including commissions:
PF: 1.40
Sharpe: 0.72

To get commissions to be included properly, you have to edit your Sim101 account in the "Accounts" tab in NT8 and select a commissions template.

This strategy only trades about once a day so commissions don't affect it much. I use market orders so slippage is a much bigger concern for me.

Thanks. I see. I definitely went back and created a commissions template after creating this post. I guess it didn't affect my performance parameters very much, so you're right about that (sharpe ratio, profit ratio, etc are the same).

What's interesting is that it didn't include my commissions in my net gain. The performance results calculated commissions, gross gain, and gross loss separately and used netGain = (grossGain - grossLoss). So I had to subtract commissions manually. Does it do that for you as well?

Reply With Quote
 
(login for full post details)
  #19 (permalink)
 ShatteredX 
Houston, TX
 
Experience: Intermediate
Platform: Python
Trading: NQ
 
Posts: 97 since Apr 2016
Thanks: 130 given, 93 received


jflaggs View Post
Thanks. I see. I definitely went back and created a commissions template after creating this post. I guess it didn't affect my performance parameters very much, so you're right about that (sharpe ratio, profit ratio, etc are the same).

What's interesting is that it didn't include my commissions in my net gain. The performance results calculated commissions, gross gain, and gross loss separately and used netGain = (grossGain - grossLoss). So I had to subtract commissions manually. Does it do that for you as well?

No, including commissions reduces Total net profit for me. It also reduces Gross profit and increases Gross loss.

Maybe it's because you're using Replay? There might be a bug with it. I'm using Strategy Analyzer.

Including commissions:


No commissions:


I'm not totally familiar with NT8 yet, still mainly running NT7.

Reply With Quote
 
(login for full post details)
  #20 (permalink)
jflaggs
Los Angeles + California/United States
 
 
Posts: 17 since Aug 2017
Thanks: 7 given, 0 received


ShatteredX View Post
No, including commissions reduces Total net profit for me. It also reduces Gross profit and increases Gross loss.

Maybe it's because you're using Replay? There might be a bug with it. I'm using Strategy Analyzer.

I'm not totally familiar with NT8 yet, still mainly running NT7.

Hmm. I'll have to ask NT support about that then. Yeah, I need to use replay because my strategy uses tick data. Nice results (: Your commissions are super low. I'm assuming you're frequency is less than 2 trades/day. Very nice PT as well. I don't know if I could survive that Max draw down though haha.

Reply With Quote
 
(login for full post details)
  #21 (permalink)
 ShatteredX 
Houston, TX
 
Experience: Intermediate
Platform: Python
Trading: NQ
 
Posts: 97 since Apr 2016
Thanks: 130 given, 93 received

Thanks. It's been optimized to historical data though so the profit doesn't really mean much, of course.

Yeah, I use NT Brokerage and lease NT, so CME futures are $4-5 per round trip per contract for commissions.

I guess I really have no idea how commissions work with Forex so I'm not sure how that compares with futures when you factor in leverage, etc.

Reply With Quote


futures io Trading Community Platforms and Indicators NinjaTrader > NT8 Performance Analysis Thoughts


Last Updated on August 22, 2017


Upcoming Webinars and Events
 

NinjaTrader Indicator Challenge!

Ongoing
 

Journal Challenge w/$1,800 in prizes!

April

Seven Trading Mistakes Solved With Smart Trading Tools w/Brannigan Barrett

Elite only
     



Copyright © 2021 by futures io, s.a., Av Ricardo J. Alfaro, Century Tower, Panama, +507 833-9432, info@futures.io
All information is for educational use only and is not investment advice.
There is a substantial risk of loss in trading commodity futures, stocks, options and foreign exchange products. Past performance is not indicative of future results.
no new posts