NexusFi: Find Your Edge


Home Menu

 





What is the best Risk Reward Ratio? But is that the right question?


Discussion in Psychology and Money Management

Updated
      Top Posters
    1. looks_one WoodyFox with 11 posts (19 thanks)
    2. looks_two SpeculatorSeth with 10 posts (15 thanks)
    3. looks_3 Sandpaddict with 9 posts (2 thanks)
    4. looks_4 Madness with 9 posts (23 thanks)
      Best Posters
    1. looks_one wldman with 6.5 thanks per post
    2. looks_two Madness with 2.6 thanks per post
    3. looks_3 WoodyFox with 1.7 thanks per post
    4. looks_4 SpeculatorSeth with 1.5 thanks per post
    1. trending_up 17,035 views
    2. thumb_up 126 thanks given
    3. group 191 followers
    1. forum 74 posts
    2. attach_file 12 attachments




 
Search this Thread

What is the best Risk Reward Ratio? But is that the right question?

(login for full post details)
  #21 (permalink)
 
trendisyourfriend's Avatar
 trendisyourfriend 
Quebec Canada
Legendary Market Wizard
 
Experience: Intermediate
Platform: NinjaTrader
Broker: AMP/CQG
Trading: ES, NQ, YM
Posts: 4,524 since Oct 2009
Thanks Given: 4,164
Thanks Received: 6,015


kevinkdog View Post
And that is without costs factored in!

If you have 1:1 NET R:R, you have to win about 55% of time just to breakeven (depending on slippage and commissions).

So, yes I agree: Something else must be at play to win. (that's the edge you hopefully have).

Exactly, this makes me think about a roulette player placing a token on the 2nd dozen bet. You bet one token to win two. He thinks, i have a good RR so let's play. slippage and commissions = zero numbers.


Reply With Quote
The following 2 users say Thank You to trendisyourfriend for this post:

Can you help answer these questions
from other members on NexusFi?
The space time continuum and the dynamics of a financial …
Emini and Emicro Index
NT7 Indicator Script Troubleshooting - Camarilla Pivots
NinjaTrader
ZombieSqueeze
Platforms and Indicators
Futures Close-Out Policy
Commodities
Request for MACD with option to use different MAs for fa …
NinjaTrader
 
Best Threads (Most Thanked)
in the last 7 days on NexusFi
Just another trading journal: PA, Wyckoff & Trends
33 thanks
Daytrading ES & NQ
32 thanks
Spoo-nalysis ES e-mini futures S&P 500
28 thanks
Battlestations: Show us your trading desks!
24 thanks
Retail Trading As An Industry
23 thanks

(login for full post details)
  #22 (permalink)
 
WoodyFox's Avatar
 WoodyFox 
Columbus, Ohio
 
Experience: Intermediate
Platform: NinjaTrader
Trading: Futures
Posts: 409 since May 2016
Thanks Given: 195
Thanks Received: 876


SBtrader82 View Post
yes sure!! First let me say that I think that there is no right or wrong in trading, so there is probably not an absolute optimal RR from a math standpoint. However if you do not scalp and you try to take some "reasonable" chunk out of the market, I think that most of the times anything in the range 3.5:1 to 4.5:1 risk reward is the best way to go.

I came to this conclusion first from a practical experience and second by build a very simple simulator.
You can see a thorough description of my experiment here:




In the experiment I was assuming that prices are "normally distributed" with variance equals 1. In this scenario the optimal value of RR depends on the stop that you use relative to the variance.
In practical terms you can think that the variance is the ATR of the day and you must assume that the stop cannot be too small compared to ATR. Thinking that your stop can be 0.05 when the variance is 1 would be unrealistic because it would mean that you can be extremely precise in your entry. Also your stop must be "reasonably big" compared to commissions. If you combine the theoretical results of the experiments with some practical experience you will probably arrive at the same conclusion.
Consider that in real life: you don't know what the daily range will be, you don't know the exact probability distribution of the market etc...

In the same thread I show a picture of the optimal RR for different values of the stop loss, see the results here:

Please don't misunderstand my experiment because it's not a "mathematical theory of everything", it's just a practical experiment that was meant to give me some hindsight. Also when I say a "optimal value" of RR, I mean something that can be right to use 85% of the days. Of course market is always changing and there are days in which you can ask the market a 10 times your risk, because there is a big trend.

@SBtrader82
If you assume continuous market participation and use no filters or directional bias...Using a normally distributed market, your number are going to be pretty close.

In late 2018, I started to understand the value of standard deviation and the importance behind the Black Scholes model.

Using a simple 1 day standard deviation based on implied volatility and using no filters or directional bias, you can see with the calculations below a 2.8/1 WL ratio can be obtained. The calculations are made by not merging quarterly contracts (4 years worth)... Its also assumes continuous market participation and only trades one contract/order. I always include commissions and slippage is minimal at this time frame.



Capture



Although its not the best equity curve, it still produces decent profits. I stopped tracking it this way back in March of 2020, because I learned there are 2 ways you can improve the drawdowns and produce better profits. One is the fact that Volatility has skew. The other I will keep to myself. LOL

At one point, I tested this back nearly 15 years and it never stopped working. So to all those that think strategies stop working...Some things never change. Thanks Karl Pearson and you Black Scholes dudes.

So getting to the point...with a better understanding and using the 2 improvements, you can increase this to about 3.5/1 WL. This is very close to what you are assuming is best. Also knowing how to calculate standard deviation under different time frames, you will start to see this 3.5/1 keeps working at higher a lower time frames. That of course is limited as you near the extremities. There a additional factors that most be considered both ways.

In short, I agree that 3.5/1 is a very solid consistent ratio...But.

What happens when you add filters, directional bias, etc?

I have strats that out perform the one above and have higher and lower W/L ratios.

I might go as far to say its a median ratio looking at many strategies, But I wouldn't limit yourself!

Reply With Quote
The following 7 users say Thank You to WoodyFox for this post:
(login for full post details)
  #23 (permalink)
 SpeculatorSeth   is a Vendor
 
Posts: 780 since Apr 2016
Thanks Given: 22
Thanks Received: 1,018



WoodyFox View Post
In short, I agree that 3.5/1 is a very solid consistent W/L ratio...But.

What happens when you add filters, directional bias, etc?

I have strats that out perform the one above and have higher and lower W/L ratios.

I might go as far to say its a median W/L ratio looking at many strategies, But I wouldn't limit yourself!

In other words, the optimal win loss ratio depends on the signal itself. It's also going to depend on your own goals. This is where other metrics become necessary. What was the max drawdown of the strategy? Can your account handle that? If you are holding for long periods of time what is the risk adjusted returns? It's not very useful when one person is talking about a market making strategy and someone else is showing a CTA strategy. They have completely different goals.

Which to the original point is why it is important to understand how these factors impact the overall system. Traders using low R strategies think they have a predictive signal when the expectancy of the system is largely due to taking on excess risk. That matters when you get into the drawdowns, and that's why it is so incredibly common to see these traders blow out. It's simply not suitable for the size and goal of the majority of retail accounts.

Reply With Quote
The following 2 users say Thank You to SpeculatorSeth for this post:
(login for full post details)
  #24 (permalink)
 
WoodyFox's Avatar
 WoodyFox 
Columbus, Ohio
 
Experience: Intermediate
Platform: NinjaTrader
Trading: Futures
Posts: 409 since May 2016
Thanks Given: 195
Thanks Received: 876


TWDsje View Post
In other words, the optimal win loss ratio depends on the signal itself.

Maybe so.

The discussion is simple though...Do the markets have preferred ratios. We are simple collaborating the ideal that one does exist...That its possible the market is structured around a ratio that exists for some balanced statistical reason. We already know you can make money using such methods, why not make it a fun exercise right?

More often when finding edges, peeps will look to find what the market will give them. Sometimes its nice to set your own terms of what you would like to get and then wrap the market around that. I mentioned this once before on another thread and was questioned by a fellow successful trader (One that many would regard highly here) how the ideal of this was confusing to him. I could tell he was very off put by this and felt his scepticism. But the simple truth is a lot of good edges live around the 3.5 mark, and I agree this is a good goal to live around. I do not in any way limit myself there, but it can be a gauge if you will.

So back to your quote "In other words, the optimal win loss ratio depends on the signal itself".

Optimal to your signal or the Market itself? Can you tell the difference?

Reply With Quote
The following 3 users say Thank You to WoodyFox for this post:
(login for full post details)
  #25 (permalink)
 SpeculatorSeth   is a Vendor
 
Posts: 780 since Apr 2016
Thanks Given: 22
Thanks Received: 1,018


WoodyFox View Post
Optimal to your signal or the Market itself? Can you tell the difference?

That depends on your definition of "optimal". Optimal total net profit? Optimal max drawdown? Optimal Sharpe ratio?

Which just proves my point. Optimal all depends on what you're trying to accomplish with your trading strategy. If your strategy maximizes total net profit but has a drawdown equal to that first, is it really optimal? Well, I can only tell you that I'm certainly not going to trade that strategy.

Reply With Quote
The following user says Thank You to SpeculatorSeth for this post:
(login for full post details)
  #26 (permalink)
 
WoodyFox's Avatar
 WoodyFox 
Columbus, Ohio
 
Experience: Intermediate
Platform: NinjaTrader
Trading: Futures
Posts: 409 since May 2016
Thanks Given: 195
Thanks Received: 876


TWDsje View Post
That depends on your definition of "optimal". Optimal total net profit? Optimal max drawdown? Optimal Sharpe ratio?

Which just proves my point. Optimal all depends on what you're trying to accomplish with your trading strategy. If your strategy maximizes total net profit but has a drawdown equal to that first, is it really optimal? Well, I can only tell you that I'm certainly not going to trade that strategy.

With all due respect, you have not disproved our exercise.

We are not debating you can get a strategy/edge to produce numbers that fit your capital or any other parameters needed to fit your scenario. This point any profitable trader already knows.
We are simply asking if there is a preferred ratio the market is structured around. One that is optimal for statistical reasons based off a normally distributed (or not) market. A balance if you will.
3.5 has been proposed here...Can you disprove this. You do seem all against the notion of it.

Remember a theory is treated with respect until disproven JMHO

Either way, I respect your opinion. I'm also not sure there is an optimal ratio either, but I do like the discussion.

Reply With Quote
The following 2 users say Thank You to WoodyFox for this post:
(login for full post details)
  #27 (permalink)
 SpeculatorSeth   is a Vendor
 
Posts: 780 since Apr 2016
Thanks Given: 22
Thanks Received: 1,018


WoodyFox View Post
With all due respect, you have not disproved our exercise.

We are not debating you can get a strategy/edge to produce numbers that fit your capital or any other parameters needed to fit your scenario. This point any profitable trader already knows.
We are simply asking if there is a preferred WL ratio the market is structured around. One that is optimal for statistical reasons based off a normally distributed (or not) market. A balance if you will.
3.5 has been proposed here...Can you disprove this. You do seem all against the notion of it.

Remember a theory is treated with respect until disproven JMHO

Either way, I respect your opinion. I'm also not sure there is an optimal ratio either, but I do like the discussion.

Right so let me try this again. You have to define optimal. Without a strict definition of what qualifies as optimal there's nothing to prove or disprove because you don't have a full theory. There's no function for us to optimize.

Reply With Quote
(login for full post details)
  #28 (permalink)
 
WoodyFox's Avatar
 WoodyFox 
Columbus, Ohio
 
Experience: Intermediate
Platform: NinjaTrader
Trading: Futures
Posts: 409 since May 2016
Thanks Given: 195
Thanks Received: 876


TWDsje View Post
Right so let me try this again. You have to define optimal. Without a strict definition of what qualifies as optimal there's nothing to prove or disprove because you don't have a full theory. There's no function for us to optimize.

I'm sorry, have you read the whole thread. Did I or did I not start in on the conversation by quoting one of SBtrader85 studies?


And Look at post #11 where he specifically proposes a 3.5 Ratio...


Capture




Are you sure you are on the right thread? Because your confusing me dude...LOL

I also just asked you to disprove 3.5 just 2 post ago?




Probably just confusion... So no harm, lets move on.

Reply With Quote
(login for full post details)
  #29 (permalink)
 
SBtrader82's Avatar
 SBtrader82   is a Vendor
 
Posts: 587 since Feb 2018
Thanks Given: 222
Thanks Received: 1,333


WoodyFox View Post
@SBtrader82
If you assume continuous market participation and use no filters or directional bias...Using a normally distributed market, your number are going to be pretty close.

In late 2018, I started to understand the value of standard deviation and the importance behind the Black Scholes model.

Using a simple 1 day standard deviation based on implied volatility and using no filters or directional bias, you can see with the calculations below a 2.8/1 WL ratio can be obtained. The calculations are made by not merging quarterly contracts (4 years worth)... Its also assumes continuous market participation and only trades one contract/order. I always include commissions and slippage is minimal at this time frame.



Capture



Although its not the best equity curve, it still produces decent profits. I stopped tracking it this way back in March of 2020, because I learned there are 2 ways you can improve the drawdowns and produce better profits. One is the fact that Volatility has skew. The other I will keep to myself. LOL

At one point, I tested this back nearly 15 years and it never stopped working. So to all those that think strategies stop working...Some things never change. Thanks Karl Pearson and you Black Scholes dudes.

So getting to the point...with a better understanding and using the 2 improvements, you can increase this to about 3.5/1 WL. This is very close to what you are assuming is best. Also knowing how to calculate standard deviation under different time frames, you will start to see this 3.5/1 keeps working at higher a lower time frames. That of course is limited as you near the extremities. There a additional factors that most be considered both ways.

In short, I agree that 3.5/1 is a very solid consistent ratio...But.

What happens when you add filters, directional bias, etc?

I have strats that out perform the one above and have higher and lower W/L ratios.

I might go as far to say its a median ratio looking at many strategies, But I wouldn't limit yourself!

@WoodyFox that's really interesting... in the thread "the war of two world" someone made a similar experiment in another market, but the results were not so positive. I am glad to discover that my intuition about the optimal risk reward can be backtested in the market.

Can you explain how you take the trades in your experiment? you mentioned that there is no directional bias, do you mean that you take random trades that can be either long or short?

Follow me on Twitter Visit my NexusFi Trade Journal Reply With Quote
(login for full post details)
  #30 (permalink)
 
SBtrader82's Avatar
 SBtrader82   is a Vendor
 
Posts: 587 since Feb 2018
Thanks Given: 222
Thanks Received: 1,333



WoodyFox View Post
Maybe so.

The discussion is simple though...Do the markets have preferred ratios. We are simple collaborating the ideal that one does exist...That its possible the market is structured around a ratio that exists for some balanced statistical reason. We already know you can make money using such methods, why not make it a fun exercise right?

More often when finding edges, peeps will look to find what the market will give them. Sometimes its nice to set your own terms of what you would like to get and then wrap the market around that. I mentioned this once before on another thread and was questioned by a fellow successful trader (One that many would regard highly here) how the ideal of this was confusing to him. I could tell he was very off put by this and felt his scepticism. But the simple truth is a lot of good edges live around the 3.5 mark, and I agree this is a good goal to live around. I do not in any way limit myself there, but it can be a gauge if you will.

So back to your quote "In other words, the optimal win loss ratio depends on the signal itself".

Optimal to your signal or the Market itself? Can you tell the difference?

@WoodyFox I like this sentence a lot: "But the simple truth is a lot of good edges live around the 3.5 mark ", I couldn't express it any better!!

Follow me on Twitter Visit my NexusFi Trade Journal Reply With Quote





Last Updated on March 23, 2023


© 2024 NexusFi™, s.a., All Rights Reserved.
Av Ricardo J. Alfaro, Century Tower, Panama City, Panama, Ph: +507 833-9432 (Panama and Intl), +1 888-312-3001 (USA and Canada)
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.
About Us - Contact Us - Site Rules, Acceptable Use, and Terms and Conditions - Privacy Policy - Downloads - Top
no new posts