NexusFi: Find Your Edge


Home Menu

 





Van Tharp's Max Expectancy


Discussion in Psychology and Money Management

Updated
      Top Posters
    1. looks_one rickt with 7 posts (4 thanks)
    2. looks_two Big Mike with 4 posts (4 thanks)
    3. looks_3 caprica with 3 posts (16 thanks)
    4. looks_4 wuming79 with 3 posts (0 thanks)
      Best Posters
    1. looks_one gordo with 5.5 thanks per post
    2. looks_two caprica with 5.3 thanks per post
    3. looks_3 Anagoge with 4 thanks per post
    4. looks_4 rickt with 0.6 thanks per post
    1. trending_up 26,527 views
    2. thumb_up 47 thanks given
    3. group 24 followers
    1. forum 31 posts
    2. attach_file 1 attachments




 
Search this Thread

Van Tharp's Max Expectancy

  #11 (permalink)
rickt
ma
 
Posts: 9 since Oct 2009
Thanks Given: 0
Thanks Received: 4

If you posted the whole formula that is used, it is missing the opportunity portion which takes time into account.

I wonder what would happen if the absolute value of the losing trades was replaced by the standard deviation of all the trades? Would that have any merit??

Reply With Quote

Can you help answer these questions
from other members on NexusFi?
Build trailing stop for micro index(s)
Psychology and Money Management
Better Renko Gaps
The Elite Circle
My NT8 Volume Profile Split by Asian/Euro/Open
NinjaTrader
Are there any eval firms that allow you to sink to your …
Traders Hideout
NexusFi Journal Challenge - April 2024
Feedback and Announcements
 
Best Threads (Most Thanked)
in the last 7 days on NexusFi
Get funded firms 2023/2024 - Any recommendations or word …
59 thanks
Funded Trader platforms
36 thanks
NexusFi site changelog and issues/problem reporting
22 thanks
The Program
20 thanks
GFIs1 1 DAX trade per day journal
19 thanks
  #12 (permalink)
 verge 
Cape Town, South Africa
 
Experience: Beginner
Platform: NinjaTrader
Broker: Zen-fire
Trading: EUR/USD
Posts: 41 since Aug 2009
Thanks Given: 22
Thanks Received: 12

Here is the complete formula:

percentWin = (double) systemPerformance.AllTrades.WinningTrades.Count / systemPerformance.AllTrades.Count;
percentLose = (double) systemPerformance.AllTrades.LosingTrades.Count / systemPerformance.AllTrades.Count;
aveWin = systemPerformance.AllTrades.WinningTrades.TradesPerformance.Currency.AvgProfit;
aveLose = systemPerformance.AllTrades.LosingTrades.TradesPerformance.Currency.AvgProfit;
expectancyTemp = (aveWin * percentWin + aveLose * percentLose)/ Math.Abs(aveLose);

btw - I opened it with Sharpdevelop (at www.sharpdevelop.net) which is a free C# IDE.

Give it a try and make all the changes you want. Personally I need to try the optimization formulas out there first before I go and change things.

Reply With Quote
  #13 (permalink)
rickt
ma
 
Posts: 9 since Oct 2009
Thanks Given: 0
Thanks Received: 4


Thanks.

I examined the code and it is missing the part that differentiates expectancy from expectancy score. What the presented code gives is a modified expectancy. If we take expectancyTemp and multiply it by opportunity as presented by the article I linked to in a previous post in this thread, we would then have expectancy score.

I would love to give it a try and make changes, but I really, at this time, have no idea about how to use the testing platform thyat you use. I will get thereb though.

Reply With Quote
  #14 (permalink)
rickt
ma
 
Posts: 9 since Oct 2009
Thanks Given: 0
Thanks Received: 4

I have just run a bunch of tests on several strategies that I am playing with and have calculated SQN, expectancy score, and a few other metrics for those tests. I am in a quandry.

Theoretically, when looking at several strategies, there is one that is better than all the others when run over the same time frame for the same set of securities. But better in what regard? That is the question. I am afraid that the answer is in the eye of the beholder.

In this thread we have discussed expectancy, sqn, and expectancy score. Looking at these 3 metrics against my tested strategies, I have not found any correlation reflecting the goodness of any strategy. In other words, what is shown to be best by one indicator is not shown to be best by any other indicator.

So again, the question comes down to exactly what are we looking for to determine goodness?

We are all doing this to make money. So that must be a criteria - how much money is returned. But because we all invest different amounts, the dollars returned must be a percentage of dollars invested. And because we know that 2% returned over 2 days is better than 2% returned over more than 2 days, we can divide the % dollars returned by the time frame (either minutes or days depending on whether we day trade or not). Taking that a step further, because we all invest over different time frames, we can multiply the result by 365 (for non day traders) or 480 (for day traders) in order to better compare the results.

Performing all of this, we have the following equation

($ returned / $ invested) * (365 / days in market) or
($ returned / $ invested) * (480 / minutes in market)

I am trying to decide whether this metric is sufficient or whether we must also account for variability using standard deviation or some such in which case we coulod end up with the following

(($ returned / $ invested) / stdev) * (365 / days in market) or
(($ returned / $ invested) / stdev) * (480 / minutes in market)

Thanks for letting me theorize here. I needed some place to discuss it. I would appreciate any and all input.

Rick

Reply With Quote
Thanked by:
  #15 (permalink)
 samurai 
CO
 
Experience: Intermediate
Platform: NinjaTrader
Posts: 29 since Sep 2009
Thanks Given: 228
Thanks Received: 45

Rick,

I believe it all comes down to one's personal preference.

Trader A may prefer a system that has a lower expectancy but also a lower variability over a system with a very high expectancy but with a larger variability of returns. Whereas Trader B may be looking for much higher returns and is willing to risk greater variability to get them.

Personally, I need to have an idea of the variability of the system I'm trading. That's why a metric which accounts for variability suits me better.



Regards,

Samurai

Reply With Quote
  #16 (permalink)
wuming79
Singapore
 
Posts: 3 since Apr 2010
Thanks Given: 0
Thanks Received: 0

Hi, I hope someone reads this.

May I know how do I classify breakeven trades as? Do I count them as a win or a loss in the expectancy calculation???

Reply With Quote
  #17 (permalink)
 
Big Mike's Avatar
 Big Mike 
Manta, Ecuador
Site Administrator
Developer
Swing Trader
 
Experience: Advanced
Platform: Custom solution
Broker: IBKR
Trading: Stocks & Futures
Frequency: Every few days
Duration: Weeks
Posts: 50,399 since Jun 2009
Thanks Given: 33,173
Thanks Received: 101,539


wuming79 View Post
Hi, I hope someone reads this.

May I know how do I classify breakeven trades as? Do I count them as a win or a loss in the expectancy calculation???

The chance a trade is ever exactly breakeven is extremely slim, if you include commission (which you should).

Your platform can be configured to include commission in the net profit so I advise you do this, and as long as your commission is not exactly the equal to a tick of profit, then there should never be an exactly breakeven trade.

Mike

We're here to help: just ask the community or contact our Help Desk

Quick Links: Change your Username or Register as a Vendor
Searching for trading reviews? Review this list
Lifetime Elite Membership: Sign-up for only $149 USD
Exclusive money saving offers from our Site Sponsors: Browse Offers
Report problems with the site: Using the NexusFi changelog thread
Follow me on Twitter Visit my NexusFi Trade Journal Reply With Quote
  #18 (permalink)
rickt
ma
 
Posts: 9 since Oct 2009
Thanks Given: 0
Thanks Received: 4

My understanding is that you do not count them as a win or a loss, but that you do not reduce the total trades. In other words, if you have 2 wins, 3 losses, a 4 breakevens, you would use 2 as the # of wins, 3 as the # of losses, and 9 as the # of total trades.

Of course, it should be a rare occurence to have a truly breakeven trade unless you are trading commissionless securities which would happen with mutual funds or some of the ETFs available through Fidelity.

Reply With Quote
  #19 (permalink)
wuming79
Singapore
 
Posts: 3 since Apr 2010
Thanks Given: 0
Thanks Received: 0

Oh, I see. yah, I forgot about the commissions. Thanks guys.

Reply With Quote
  #20 (permalink)
wuming79
Singapore
 
Posts: 3 since Apr 2010
Thanks Given: 0
Thanks Received: 0



rickt View Post
My understanding is that you do not count them as a win or a loss, but that you do not reduce the total trades. In other words, if you have 2 wins, 3 losses, a 4 breakevens, you would use 2 as the # of wins, 3 as the # of losses, and 9 as the # of total trades.

Of course, it should be a rare occurence to have a truly breakeven trade unless you are trading commissionless securities which would happen with mutual funds or some of the ETFs available through Fidelity.

Hi, I'm reading Van Tharp's "Trade Your Way to Financial Freedom". I don;t quite understand how the opportunity cost is calculated. Does anyone knows how he calculate it? The book seems to be measuring by number of opportunity per day. How does one calculate when he is doing forward testing?

Reply With Quote




Last Updated on January 22, 2015


© 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