Example Trading System

Discussion in 'Strategy Building' started by sidinuk, Jan 17, 2005.

  1. mogul

    mogul

    ok ok let me get this straight

    some hedge fund is going to try to nudge around the ym to "maybe" trigger some entry stops from some marginally profitable trading strategy on ET?

    some trades go a couple of ticks in your favor just to reverse because your mind points them out like that...they are just random trade outcomes in intraday noise
     
    #211     May 13, 2005
  2. jrkob

    jrkob



    This system is a very simple ORB system, and it's been public for years, if not decades, NOT since this thread has been opened at ET. I've got a book at home (Trading Systems and Methods, Kauffman 1998) that presents a whole set of breakout systems (including this one), and I'm sure you can find older books with the same systems. If I follow your logic the hedge of these systems should have disappeared years ago then, which doesn't seem to be the case (of course we could argue about how significant is the hedge, but that's not the point here).

    I think you largely overestimate the influence of this forum Jonny.

    Not only that, but Sidinuk gives clues as to how to avoid the problem of everybody trading this system at the same time.

    Question for you now: have you tested this system on the HSI ? If yes, what are your findings ?

    Thank you.


    * Mogul's above post is dead on.
     
    #212     May 14, 2005
  3. mogul and jrkob, i respect your opinions and have nothing against you two. but i have to disagree with the points you are both making, because stops are gunned, and systems have statisically performed worse once made public(i have been told by respected traders). i think once you get past a certain size level(not saying im there yet), its not about indicators/systems, but more about feeling for whose holding what and when they are most likely to liquidate and when are they most likely to enter and it becomes just like poker instead of buy when the 4barEMA crosses the 21barEMA, imo. i am speaking strictly of short term trading. i think this system has been working well(historically) because YM is a relatively new table to play at, there are no old hands that have a instinctive feel for the players. just like in poker; you have to know within 10 minutes of sitting at the table who is going to give you their money, weak hands will always provide for the strong hands. i urge you all to test this same strategy on other markets and let you draw your own conclusions. if you both disagree strongly with my opinions i welcome you both to share your strategy with all of us to learn from. peace.


    p.s. respected past ET member 'Acrary', said that only 1 edge ever died a quick death on him and that he attributed that to it being made public........food for thought........

    p.s.s. jrkob, traders down here in aussie are fading the ORB on HSI, but im sure the 'ole' envelope the morning session strat still works under the right conditions.
     
    #213     May 14, 2005
  4. sidinuk

    sidinuk

    OK, as pointed out the example trading system had a truly awful week last week, losing $4575 across the 5 contracts that I monitor. Historically, though, the worst losing week was w/e May 14, 2004, where we would have lost $5542.50. Individually it was the worst week for NQ and YM but not for ES, EMD or ER2.

    NQ and YM are also at worse than historic drawdowns but ES, EMD and ER2 are not.

    It is to be expected that future drawdowns in any system will exceed historic ones, after all any system, to some extent, is optimized on that past data. That is why for money management you should allow at least 2 times the historic drawdown. If we exceed twice the historic drawdown then we might need to consider that the market dynamic has altered - but I see no reason to panic at the moment.
     
    #214     May 16, 2005
  5. jrkob

    jrkob

    Sidinuk, I have a question to ask you.

    Your system has a "no trade on Thursday" filter. This is based on the fact that the expected proffit per trade on Thursdays between January 2004 and June 2004 was negative. Fair enough.

    If I remember correctly you started to forward test this system since 1 Jan 05 is that correct ?

    Below is the breakdown of total proffit (after 2pts slippage and commission) per weekday (based on the Excel file downloaded on your website) and for the YM contract:

    Mondays: -225 pts
    Tuesdays: -13 pts
    Wednesdays: +188 pts
    Thursdays: +224 pts
    Fridays: +125pts

    Do you agree with my above calculation ?

    If you agree with my above calculation, it seems that Thursdays have shifted from worst performers during the backtest period to best performers during the forward test period. Since 1 Jan 05, we have had only 14 traded Thursdays, so I don't think this is statistically significant.

    But doesn't that mean that we should keep an eye on the weekday breakdown to potentially invalidate the No thursday trade filter ?

    Thank you, and keep up the good work.
     
    #215     May 19, 2005
  6. sidinuk

    sidinuk

    After all the doom and gloom of the previous 2 weeks the example trading system got back to winning ways last week - I hope all those 'professionals' who are apparently fading this system didn't get too burned!

    Out of the 5 contracts we follow we had 16 winning trades out of 16 for a gross profit of $3,857.50. The Thursday filter worked in out favour, for once, and saved us from giving back $1,212.50.
     
    #216     May 24, 2005
  7. sidinuk

    sidinuk

    Last week was a period of small daily trading ranges. To be expected after a period of larger ranges the previous week.

    EMD and ER2 made a slight profit last week but NQ, YM and ES made losses which put us 607.50 down for the week overall.
     
    #217     May 31, 2005
  8. New to this thread ...

    Is it available in Amibroker's AFL code?

    Thanks,

    Brooks
     
    #218     Jun 3, 2005
  9. Is anyone trading this live?
     
    #219     Jun 6, 2005
  10. I trade a slightly modified version on ER2.
    Does reasonably well.

    DS
     
    #220     Jun 7, 2005