Questions - Buy New Highs, Sell New Lows Strategy

Discussion in 'Strategy Development' started by elcowen, Oct 12, 2011.

  1. elcowen


    Does anyone trade by buying the new highs and/or selling new lows on a 30 second or 1-minute chart?? I did some back testing and it looks to be fairly profitable scalp play. Just curious to see if anyone has experience with this.

    Example: Wait 15 min after open, buy new highs with a 4-5 tick trail. It only looks to be profitable between 9:45-10:30 am EST.

    I have not found a way to stay out of false breakouts. Any suggestions?

  2. Hey, whats up ?

    Did you meant how many floor traders do not trade this scalping style ??? I dont know ??? Thanx...

    To avoid false breakouts, you must watch the price in the overall context on all chart timeframes....regarding to support/resistance...and probabliy market turnpoints....
    or, you will never understand to pick tops and bottoms correct with a high win rate !!!
    Its not easy, but it can be done !!!:cool:
  3. It works EOD on stocks at all-time highs. See Eric Crittenden's paper "Does Trend Following Work on Stocks?"
  4. this may work. although i haven't backtested it, just looking visually where entries/exits would have been seems to support the idea. basically you use really tight stops, scale in (in thirds) as the trend continues and exit all at once when the trend ends. you can trail stops/move to BE on a portion if you want to reduce risk however you may/will miss out on trades that pull back and continue. i'm interested in this strategy on this time frame so if anyone has any ideas i'm all ears.
  5. elcowen


    Where would your initial stops be? How do you protect from false breakouts is the biggest problem.
  6. on next support/resistance, or if there is none, then on 1 : 1 level from, minimum expected target support/resistance....
  7. the initial stop would be say 1% (of your equity) from entry (it could be lower if you think that's too high) and you would scale in (in thirds).

    you can't protect against false breakouts b/c no one knows beforehand which breakout will fail/turn into trend that pays for all the other failed breakouts.

    for ex. divide your capital into 3 cars

    price is above high of last 20 bars (say 500 tick)
    buy 1 car risking 1% (really 0.33%)
    if stopped out you only lose 0.33
    if it continues another x atr buy another car risking same % and so on for the third contract.

    to reiterate, during chop you will lose money period however you will only lose a little. it's like losing 0.33, then another 0.33, then another 0.33 then you make 4.00.