Frosty's auto-trading bot goes live with REAL money

Discussion in 'Automated Trading' started by frostengine, Nov 14, 2006.

  1. Businessman

    Businessman

    Was today's $1350 loss part of the system, or a mistake?
     
    #581     Jan 3, 2007
  2. Today's loss was part of the system.. I watched it all unfold in front of my eyes... My emotions said to sell for a -$100.. I could tell the market direciton had shifted.... I spent the better part of the afternoon trying to find a way so that in the future the bot could identify what I saw... but I just can't put it into code.... That is partly why I decided to pull the plug..... the bot needs some sort of way of knowing these types of changes.. I can spot them.. but don't know how to tell the system to spot it..... I think that is the system's weak spot and it needs to be addressed
     
    #582     Jan 3, 2007
  3. One idea would be to identify the usual news based market moving times and use tighter stops/go flat at that point. You know the market is going to trend. You just don't know what way. Why stand in front of the train at those times.
     
    #583     Jan 3, 2007
  4. TraDaToR

    TraDaToR

    Sad... really sad...


    Random things why I think it failed this time:


    - You were not enough capitalized

    - You didn't know when you would stop the system( I think you still haven't hit your historical DD )

    - You are too focused on every single trade and it sometimes result in interfering with the bot.

    - One of the most important to me : You are making too much trades and you trade intraday only. Nothing new but swing systems are much more reliable. Try an optimisation of your different variables. I think you will not get more than 1/3 of the solutions that are profitable on a 5min time frame.

    - Trading 1 only system is definitely real hard since you put all your hopes on one system. I have almost a constant with my systems that 70 % of my systems will produce gains in the end. If you are trading one only system , you still have 30 % chance of losing money.

    - I don't know but it seems that your system does not have a real edge. You're following the long term trend but i don't know it can be called a real edge.

    i hope you'll recover soon. BYE.
     
    #584     Jan 4, 2007
  5. Sim mode resulted in +$115 for the day
     
    #585     Jan 4, 2007
  6. Hmmm.....the move from 1:45 pm to 3:45 pm was $1000 per contract....straight up with little retracement.
     
    #586     Jan 4, 2007
  7. syswizard,

    I have no way of knowing if the bot would have been on the right or wrong side of that move. My apt had a power outage at around 12:20 or so... So I do not know how the bot would have performed in the afternoon... Luckily at the time the bot was flat.....
     
    #587     Jan 4, 2007
  8. I strongly advise against the use of any trend-fading system for the ER2. The ER2 is known for its powerful, intraday trends which is where I suspect Frost is getting hammered.

    He mentioned one day where he made $350 per car, and on that day I lost $800 per car. That day happened to be very choppy, which is why I suspect he made money. Yesterday, however, I was able to realize a profit of $1400 per contract, while he lost approximately the same amount. Yesterday was as trendy (both long and short) a day as you'll ever see in the ER2. It does NOT pay to fight the trend for this particular instrument.

    RoughTrader
     
    #588     Jan 4, 2007
  9. syswizard,

    With all due respect, I have to disagree with this sort of commentary. You assume that one is able to capture the move with 100% entry and exit efficiency, which is of course impossible. You will also note that the majority of the session (excepting the opening downward move) was substantially choppy. Intraday trend followers I think would have considered today a challenging day to make decent money.

    I definitely think Frost needs to work on his system on two fronts:

    1) he seems to be fighting ER2 trends, which is a no-no in my book.

    2) for as frequently as his bot trades, he has not arrived at the point where his platform is stable enough to realize his backtested statistics in real time.

    Frost, I urge you to earnestly reflect on both of these points. In the future, when you start trading on 1 contract again, I would suggest you use something more mature like NeoTicker if you stick with IB, or TradeStation. Either would be fine for small retail e-mini traders.

    RoughTrader
     
    #589     Jan 4, 2007
  10. roughtrader,

    What stability issues are you refering to?? I had only 3 problems:

    #1 was a code screw up I made the night before which did not cause any problem, just ment the bot did not trade that day

    #2 was one day of lost quotes when A LOT of people were having quote problems form IB..... before and since no issue there.. that is to be expected occasionaly though...

    #3 A freak power outage...

    I don't see what stability issues my trading platform has..... the auto trading bot has been doing everything correctly
     
    #590     Jan 4, 2007