Keeping It Simple

Discussion in 'Index Futures' started by dbphoenix, Nov 27, 2002.

  1. dbphoenix

    dbphoenix

    My figures are slightly different, but the gist is the same:

    The average gap is just under 1.2% (on the Naz). These fill 71% of the time.

    Gaps that are double that - just over 2% - fill only 36% of the time.

    Gaps of 3% (6% of all gaps) fill only 23% of the time.

    Gaps of 4% or more (2.4% of all gaps) fill only 5% of the time.

    [My note: since the larger gaps happen so rarely, it's difficult to say how much upside is left after such a large opening gap, unless it's a Trap Gap; someone with a much deeper archive of charts than I have would have to look at this and assess the probabilities for trading in the direction of the gap, i.e., rather than fading it]

    --Db
     
    #311     Jan 10, 2003
  2. mojo59

    mojo59

    I lied. I got 29.5 too. In at 1064 and out at 93.50 using 5 min bar to trail instead of 3 min.
     
    #312     Jan 10, 2003
  3. jstormbo

    jstormbo

    I also took the fade on ES, about the same time. 17.25 then added to full position on a breakout at 24.5. I think Monday was a bit better than today but still no complaints. My stop on the initial entry was simply a break of the opening low. If it looked like the 15's were going away I'd of got out. As it stands I took my final exit at 30 around 11:15. Probably could of taken a reversal at ~27.5 but decided to go run the engines on my boat.
     
    #313     Jan 10, 2003
  4. dbphoenix

    dbphoenix

    I'd be interested in knowing how scalpers did today. Unfortunately, nobody's keeping journals anymore.

    --Db
     
    #314     Jan 10, 2003
  5. dbphoenix

    dbphoenix

    This is sort of off-topic, but I can't help wondering if much of the action today is due to traders being caught short after hoping that the H&S would play out. After all, it was one of the prettiest I've seen in a long time, though I would have expected most shorts to occur at the neckline, not the right shoulder.

    --Db
     
    #315     Jan 10, 2003
  6. jstormbo

    jstormbo

    I know some guys that tried to sell resistance a couple times and got their lunch money taken away from them this morning. I'm learning I have to be a real hermit in regards to what others are doing or I start thinking too much. I did some scalp trading this week tuesday-thursday during the chop but my approach is to only take a few well placed scalps to try and recoup whatever I lost on the stops from my trend trades.

    Do any of you guys do any automated backtesting? If so how. I'd like to do more of that but I use esignal which has a backtesting function but I haven't really figured out how to write the scripting language module to do what I want. Most of the conclusions I come up with are based on my journals, I have one devoted to gaps. I scribble alot of notes and print out chart formations as I encounter them and make my conclusions from that. I also manually go back thru the charts but it's hit and miss.
     
    #316     Jan 10, 2003
  7. After taking a directional trade on this morning's big gap-fill rally, I've switched into scalper mode and I am batting 4 for 4 so far on my scalps. Today's market has been great for day traders...big price swings for the trend-followers, and enough follow-through on the micro-moves for the scalpers to pick off lots of one- and two-pointers. Day's like today are one of my favorite types of trading days.
     
    #317     Jan 10, 2003
  8. dbphoenix

    dbphoenix

    The strategy I follow can't be backtested other than manually, so I'm afraid I can't be of any help.

    --Db
     
    #318     Jan 10, 2003
  9. jstormbo

    jstormbo

    I think it's the same for me, my trading has a great deal of discretionary component to it. Besides I guess manually scrolling thru charts gives you lots of practical experience you wouldn't get by running a script. I remember when I first started trading and had a subscription to daily charts. Got this big bundle of charts every week, anything more current than that was done with graph paper and a pencil.
     
    #319     Jan 10, 2003
  10. dbphoenix

    dbphoenix

    Though it needn't take as much time as one might think. If I'm trying out a new idea, and I've gone over a month's (20) charts, and I find nothing exciting, I just quit. Only if a preliminary check yields an acceptable probability do I dig deeper and do a more thorough backtest (sometimes I only have to check two weeks' worth before abandoning the effort).

    --Db
     
    #320     Jan 10, 2003