This forum is for students of the Bearish Butterfly trading strategy to share information.

Backtesting

17 replies, 5 voices Last updated by Alex Bershadsky 3 years ago
Viewing 17 reply threads
  • Author
    Posts
    • #13097

      Alex Bershadsky
      Participant
      @alexbershadsky@yahoo.com
      Points: 183
      Rank: Newbie

      Anyone have that last excel sheet?

      The links no longer work.

      Thank you

    • #3502

      Michael Grech
      Participant
      @mgrech@shaw.ca
      Points: 207
      Congratulations Hawaii Challenge Goal Accomplished!
      Rank: Newbie

      david, is there an email address i can reach you at?  i have some backtesting questions not related to the BB that i’d liek to discuss.  i am at mgrech at shaw dot ca.  thanks.

    • #3241

      David Tang
      Participant
      @tangpd
      Points: 250
      APM2
      Rank: Newbie

      Hi, I just generated back-test results with a profit target of $15K for both Bearish-Butterfly and Bearish Butterfly with New Delta Guidelines. Compared to the previous 2 (BB & BB with New Delta Guidelines WITHOUT profit target), the best-to-worst performance is

      1. Bearish Butterfly with No Profit Target, New Delta Guidelines at 21 DTE, Max. Loss at $15K – $338K
      2. Bearish Butterfly with $15K Profit Target, New Delta Guidelines at 21 DTE, Max. Loss at $15K – $319K
      3. Bearish Butterfly with $15K Profit Target, Max. Loss at $15K – $278K
      4. Bearish Butterfly with No Profit Target, Max. Loss at $15K – $201K

      So looks like using new delta guidelines when closer to expiration is definitely a winner, and it does pay to quit when you hit the profit target as month-to-month P/L will be much more consistently positive. The difference between not having a profit target and hoping to hit jackpot (item-1) versus having a profit target (item-2) is only $20K and a lot less heartache 🙂

      The graph of the 4 BB back-test results from 2008 to 1025 as follows:

      You can also find the detailed log files in my drop-box:

      https://www.dropbox.com/sh/qxs25rtv21rbmo9/AAA7rDNEvbF9Sa4_P4ZFZWTea?dl=0

      Cheers,David

       

       

    • #2998

      David Tang
      Participant
      @tangpd
      Points: 250
      APM2
      Rank: Newbie

      OK, I have just completed generating the back-test for BB with new delta guidelines, and have uploaded it to my DropBox.The link is as follows:

      https://www.dropbox.com/sh/qxs25rtv21rbmo9/AAA7rDNEvbF9Sa4_P4ZFZWTea?dl=0

      On a quick glance, the P/L for BB with new delta guidelines looks much better than doing a basic BB which is understandable as you have to do more work to constantly balance additional positive and negative Deltas with the right quantity of butterflies, and not just bluntly putting on or taking out a full butterfly. However, there are still periods where there are consecutive months of losses i.e. April to Oct 2008, April to Aug 2013 etc. I will try next for having a position to close upon hitting minimum $15K profit, and see if overall P/L can improve. I think most of us will prefer steady wins than large swings of winnings and losses:o) Please let me know if you have any other ideas and I will try it out on my program.

      Cheers,

      David

    • #2997

      David Tang
      Participant
      @tangpd
      Points: 250
      APM2
      Rank: Newbie

      Hi DG,

      For the test result that I posted yesterday, I did not set any profit target so the program will execute until expiration which is DTE=2 (expiration day is Saturday and option stops trading after Thursday on expiration week). Likewise, I can set a profit-target and my program will then close the position anytime once it hits that target. Anyway, what I am doing is purely academic, as we know no rational traders will trade like that haha! I think a better way is to trade based on BB rules, and then convert into BB with new Delta guidelines when closer to expiration (maybe 21 or 14 DTE) and then trade until expiration (again with no profit target) just like what JL does in his training videos. I’ll post my results on BB with new Delta guidelines later today and we’ll see how the results compare to the first one.

      With regards to the discrepancies with JL’s results, is he using BB rules with no profit target or with profit target, did he convert to BB with new Delta guidelines when closer to expiration? If you have his detailed log results, we can do a line-to-line comparison and see if any differences in the adjustments.

      Cheers,

      David

       

    • #2996

      D G
      Participant
      @thetaeater@gmail.com
      Points: 771
      30 Day Affirmation Challenge
      Rank: Freshman

      Hi David…AWESOME JOB…..the output from your program is very comprehensive, seems like you did put a lot of effort into it…wow!! It looks like BB made almost 200K in 8 years of your test period. I did notice that in the 93 trades that your program backtested, there were LOTS of trades that the program took the trade all the way into 2DTE, I think, many traders do not do that, guidelines generally do not take the trades all the way to 2DTE……..why are you not closing the trades much earlier? I do not see much benefit in taking the trade all the way to near expiration. For e.g. your trade #2, has +$29,975, would you not close it when 30% profit target ($15,000). Also, why do you have trade #26, with a whooping loss of 39K, would you not close it at approx 15K loss?

      Also, your 2012, 2013 and 2014 results are +5.4%, -93.8% and -0.6% whereas John Locke’s projected results for those years are 129%, 0.9% and 146% respectively. Why such large discrepancies?

      .

      Thanks.

    • #2993

      David Tang
      Participant
      @tangpd
      Points: 250
      APM2
      Rank: Newbie

      Hi DG,

      Yes, I created an automated back-testing program, the program uses raw data – underlying prices, option prices, greeks purchased from a third-party data service provider, and using the strategy rules (i.e. Bearish Butterfly) my program will open positions, make adjustments and close the positions based on the rules settings or whether maximum profit/loss is hit. The software can do fully automation, so 8 years of data from 2008 to 2015 can be generated in under 2 hours. The program also allows manual analysis, I can choose to enter a trade in any historical time, go step-by-step each day on the trade, and if there is “Action Required!” alert I can choose to make adjustments based on the recommendation by the program, try new adjustments, not do anything and move forward in time or close the position. Also, my program is linked to TOS using Real-Time Data (RTD) so my ‘live’ trades are captured in the program. So whenever there is a trigger for a adjustment required due to live price movement or the Greeks or both, my program will flash an “Action Required!” alert. I find that is very useful as we are know when we are in the thick of action during trading, its crazy to fumble calculating the Delta-Theta ratio or remember rules like “adjust when Delta exceeds 250 and price is outside expiration B/E point”. Similarly by using the RTD feature, I can use TOS’s historical prices and Greeks (TOS’s OnDemand) to do my back-testing. However, TOS’s OnDemand has not been working for quite a while, they stopped giving Greek data (the Greeks either shows N/A or 0). I have been checking with TOS development team on the status for a while, so far they haven’t given a date as to when the historical Greeks data in OnDemand will be restored. Perhaps all TOS clients like us should collectively give pressure to TOS haha!

      Back to your questions, what I mean by “New Delta Guidelines” is the new guidelines proposed by JL in his BB series (ie. when trading near expiration or when profit target is exceeded and you choose to stay in the position). Using New Delta Guidelines, the butteries will not have full quantity but between 0 and 10 in order to balance the Delta. I will generate my results based on following the “New Delta Guidelines” post it in my DropBox.

      Lastly, I can’t remember which month it was that I had the infinite loop issue, will check and let you know. I appreciate if you can double-check my data results against yours as I do not own any 3rd party options tools (OptionVue, OptionNet etc..). Many Thanks!

      Cheers,

      David

       

       

       

    • #2990

      Sherri Locke
      Keymaster
      @KDAKSj4ux
      Going to The Trading Triangle LIVE 2016Ultimate Income Trader Workshop

      You can now add a picture up to 2mb…

    • #2984

      D G
      Participant
      @thetaeater@gmail.com
      Points: 771
      30 Day Affirmation Challenge
      Rank: Freshman

      Thanks David.

      1. I think there is some issues going on with the community website that past attachments are gone and users can’t add new files to the posts. I believe, John/Sherri are looking into it.
      2. Interesting to know that you “re-generated” the back test results for so many years quickly. Did you create an automated back testing program to back test? There are pros and cons of doing that, but I think everyone could benefit if u can share your approach.
      3. You said, “program gets into an infinite loop i.e. triggers a butterfly removal due to D/T ratio, but once removed, triggers adding back the recently removed butterfly due to current price to reference point.”
        However, in human back-test scenario, I have not run into a situation where I have I run into an infinite loop. You may want to check to make sure that your programming of BB is accurate. Or you can point out a month where that happens and I can “manually” back-test it.
      4. When you say, “New Delta Guidelines”, do you mean guidelines after less than 21DTE?

      Thanks for uploading the file thru dropbox.

    • #2979

      David Tang
      Participant
      @tangpd
      Points: 250
      APM2
      Rank: Newbie

      Hi DG,

      Strange to know that my uploaded files are deleted.. I need to check with the IT administrator what happened.

      Anyway, I have re-generated my back-testing results again using RUT prices from 2008 to 2015. Instead of uploading it here, I have placed it in my DropBox. Please see the link below to download the file. This back-test results uses purely Bearish Butterfly rules, there is no use of “New Delta Guidelines”. The rules for roll-ups, roll-downs, adding new butterfly and butterfly removal follows exactly JL’s BB guidelines, but I also make some minor additional rules when sometimes the program gets into an infinite loop i.e. triggers a butterfly removal due to D/T ratio, but once removed, triggers adding back the recently removed butterfly due to current price to reference point.

      The trade will be stopped either upon reaching expiration or when it hits the 15% maximum loss. Appreciate if you can compare the back-test results with yours (using OptionVue or any other back-test tools) and let me know if there are differences.

      https://www.dropbox.com/sh/qxs25rtv21rbmo9/AAA7rDNEvbF9Sa4_P4ZFZWTea?dl=0

      If you like to have back-test results using both Bearish Butterfly rules and New Delta Guidelines, I’ll re-generate the results and post it again.

      Cheers,

      David

       

    • #2936

      D G
      Participant
      @thetaeater@gmail.com
      Points: 771
      30 Day Affirmation Challenge
      Rank: Freshman

      Hi David-

      Where did u upload the backtest files? If on this thread, they seem to have disappeared.

    • #2536

      Michael Grech
      Participant
      @mgrech@shaw.ca
      Points: 207
      Congratulations Hawaii Challenge Goal Accomplished!
      Rank: Newbie

      hi david, yes that helps. thanks.

    • #2525

      David Tang
      Participant
      @tangpd
      Points: 250
      APM2
      Rank: Newbie

      Hi Michael,

      What I mean is upon DTE at 21 days or less, the BB position will be adjusted based on “New Delta Guidelines” (John’s BB 2nd PPT slide 31). However, as my back-test historical data only have 1 set of data per day (3:30pm EST), so I can’t adjust intra-day. Instead I adjust whenever Delta exceeds 250 to bring my position down to less than 250 Delta. The reason I chose 21 DTE is because John’s recommendation to close the original BB position if the reduced profit is hit on 21 DTE (John’s BB 1st PPT slide 14) so I like to see if by converting into new delta guidelines, I can increase my profits. Hope this explanation helps!

      Cheers,
      David

    • #2517

      Michael Grech
      Participant
      @mgrech@shaw.ca
      Points: 207
      Congratulations Hawaii Challenge Goal Accomplished!
      Rank: Newbie

      hi david, thanks for sharing. what do you mean by ” convert into new guidelines on Delta (BX) ” ? thanks.

    • #2462

      David Tang
      Participant
      @tangpd
      Points: 250
      APM2
      Rank: Newbie

      Second lot of last 4 files.

    • #2457

      David Tang
      Participant
      @tangpd
      Points: 250
      APM2
      Rank: Newbie

      Sorry the files didn’t get through as they were too large, I have to break them into smaller sizes and re-sending it again.

    • #2456

      David Tang
      Participant
      @tangpd
      Points: 250
      APM2
      Rank: Newbie

      Hi Michael,

      I’ve done some back-testing using Bearish Butterfly strategy on the RUT from 2008 to 2015. I enter the trade 56 days to expiration using Bearish Butterfly (BB) rules, and I will then convert into new guidelines on Delta (BX) when DTE reaches 21 days, and trade until expiration or when max loss is reached. All adjustments are made only once a day at 3:30pm EST. All the trades are based on $100K planned capital and max loss is $15K.

      Attached Excel spreadsheets has the summary of the P/L as well as the log details.

      Cheers,
      David

    • #1978

      Michael Grech
      Participant
      @mgrech@shaw.ca
      Points: 207
      Congratulations Hawaii Challenge Goal Accomplished!
      Rank: Newbie

      Has anybody done any backtesting on the Bearish Butterfly between 2005 and 2011 and if yes, would you mind sharing? Thanks.

Viewing 17 reply threads
  • You must be logged in to reply to this topic.