Stochastics and price action

Discussion in 'Technical Analysis' started by feb2865, May 8, 2007.

  1. Hey man.

    I am using stochastics as part of a multiple timeframe system.

    It's actually simple and it's my take on the infamous triple screen.

    When the 1hr chart shows a downtrend (ie the EMA26 is declining) and price is fairly valuated (at or in between the 13ema and the 26ema), then i look at the 10 minute chart and wait for the stochastic 9,3,3 to become overbought telling me to sell into the overall downtrend.

    It's not that simple though. You still have to decide the market dynamic on the 1hr. If the MACD histogram is declining also it's a good time to get bigger, if not, play cautiously...
     
    #11     May 9, 2007
  2. don't have indicies, eurusd 1 and 5min charts example
    stoch 13:3:3 with yellow/red 89:3:3 stoch overlaid
    89 is a arbitrary, settings would decrease as the
    timeframe increases
    Fisher Transform is interesting but can't be adjusted
     
    #12     May 9, 2007
  3. The only 2 problems with stochastics (and those made me replace it with a MACDHistogram as an oscillator) are:

    1) It's subjective in what concerns the time window, as it measures the close vs the range of that time window, and with all the smoothing applied to it, it still issues false signals, and when price action is choppy, you are in for a rough ride.

    You would even have the surprise of getting different signals using two consecutive time-windows, say a 9 and a 10 or 11.

    2)It is normalized and does not have a clear quantitative measure of momentum therefore making the rules of buying when it is oversold and selling when it is overbought useless and most of the time, even combined with a higher timeframe notion of trend it will have you selling into a new uptrend and buying into a new downtrend.

    Plus, it's too sensitive and it is measured using the closes. When you use it intraday, what difference does the close on a 10 minute bar make?

    Therefore, as an oscillator i would recommend using a MACD Histogram calculated using the H+L+C/3 formula instead of the Close price and considering the following rules:

    When MACD Histogram is above 0 it's overbought and when it's below 0 it's oversold. NOTE: This can ONLY be used considering a higher timeframe trend.
     
    #13     May 9, 2007
  4. And then there are people who find the opposite. Isn't the diversity of the markets wonderful?
     
    #14     May 9, 2007
  5. #15     May 9, 2007
  6. He's not trading ES.

    Mark
     
    #16     May 9, 2007
  7. On a 1 minute chart, you really will not need too much testing, just be aware of the shape of the day/hour/30M when you take your stoch readings.

    Dont pay too much heed to the stoch consistency but rather concentrate on your ability to use it consistently.

    ie on a 1M, what are you trying to achieve, given that the price wobbles along in channels.
    Are you trying to fade the extreme points or are you trying to breakout of channels in which case your risk will be the channel depth plus a tic or two.

    The stoch is only going to highlight a piece of price action that is already embedded in the chart. The same can be said for any indicator quite frankly.

    I use price/volume on the ES and the only packet indicator I use is stoch as it highlights the position of the price relative to it's channel.
    I do not use it as a signal, merely as a highlight.

    Someone mentioned they prefer MACD and my comments apply directly to that indicator as well.

    Bear in mind that the stoch is described as a momentum indicator, but extend it's look-back sufficiently and it shows trend. In reality trend and momentum are one and the same. It all depends upon your trading philosophy as to how you use them.

    Because I trade in a frame just above that of a Scalper, my motto is
    "one man's noise is another man's Ferrari"

    What I am saying to you, is that you need to take a number of nested decisions first and then your use of an indicator will more or less drop into place within your trading philosophy.
     
    #17     May 9, 2007
  8. feb2865

    feb2865

    fearless

    thank you so much for your response

    and to all of you that response on this thread, thanks

    it proves that still we can have an educated conversation about trading on this site

    I've been trading all my life with two simple S/R lines. I have no complaints. I tried Moving averages once but no success. I start looking at stochastics just this past couple of weeks and the results I got were more than I expected.

    Very interesting what you said about channels, fearless I will look into that

    to give you guys my rules in detail so far here it goes

    1) Stochastic 14,3,3. 25 oversold and 75 overbought

    2) 1 minute candlestick charts ( I still think hammers are powerful)

    3)The candles will highlight every time the price goes above 25 or bellow 75

    4) I am using a risk/reward ratio of 1/1 as I am not very familiar with YM but so far so good

    if there's no announcements at 10:00 am I consider the first signal only if price makes a new high/low after the signal

    If there's anything at 10:00 AM , then I wait up to the market comes from oversold/overbought.

    I tried to put in the mix some other indicators like MACD, etc but as I've been trading price all my life, I 'll stick to what the price says.

    I backtested as far as Feb 20 2007 and the results are simply amazing. But I think fearless is right. You don't need to backtest it as is more on how we used in other words is not the the tool but how you apply yourself using it. My observation so far is that helps me see what the price is dictating in a more pictorial way. As I said before, I think is completely wrong to trade indicators, you should trade price. Indicators should be used as the word implies, they indicate, that's it.

    By the way, I am testing it on YM as is more gentle on the swings that mini Russell (I trade ER2 consistently) but I guess you could try it on any of the indices. Nothing wrong with YM on the contrary I think trends very good

    If anybody has some input on this subject please don't hesitate.

    Again to all of you, thank you so much for a well-educated conversation. Definitely sets an example on this site.
     
    #18     May 9, 2007
  9. They only work on swings when the market is not trending hard. Take only sto trades where it has done a full cycle. I.E. if you are buying oversold make sure the last trade was selling overbought and the sto was overbought. This indicates a meandering market.

    If you are buying oversold but on the last swing the sto never got overbought then it is generally a continutation.

    Best way to use to Sto's in trending markets supported by news is to use them to get in with the trend on a pullback. So if market is up big let the sto get over sold and then use a trendline to get back in with the market.


    John
     
    #19     May 9, 2007
  10. feb2865

    feb2865

    well is a nice thread about the subject

    I have a neutral bias about mr. Jack Hershey. I have to agree with what JEM said on that thread, he uses so many esoteric words that in my particular case my eyes get crossed. But he's right on what he said in that particular post, of course the part that was written in English. Definitely worth reading
     
    #20     May 9, 2007