How to measure choppyness?

Discussion in 'Trading' started by 0008, Apr 14, 2003.

  1. 0008

    0008

    Is there any way to measure the choppyness (or trendness) of an instrument?

    Thank for any comment!
     
  2. For trendtraders, choppiness usually correlates well with the shape of your equity curve... if your equity curve is choppy, then what you are trading probably is...

    For scalpers, the shape of the equity curve cant be used to proxy choppiness very well... however the gradient of the equity curve may be a good measure of choppiness, since the profit to commission ratio increases when scalping chop...

    Just my 2 cents...
     
  3. 0008

    0008

    So you mean I need to trade it before I know its choppyness? Is there any way to measure it from its historical data?
     
  4. prox

    prox

    Use some moving averages and watch it braid.

    Look at pivot highs or lows get taken out with no followthrough and then reversing. Therefore, trapping breakout and trend traders but not having any followthrough the other direction. People then just give up in frustration and sit on the sidelines which makes everything even more boring.

    Basically a megaphone expansion pattern or a triangle converging pattern.
     
  5. No, of course not. Prox has the right idea.

    First decide on what time frame you are looking at. Then pick a moving average that roughly corresponds to your trading timeframe.

    Then calculate the number of inflection points (changes of slope from positive to negative and vice versa) over some historical time period. The more inflection points you have, the choppier the instrument in this particular timeframe. If you apply this to multiple instruments you can quantitatively rank them all on historical choppiness (which may not always predict future choppiness, but it's a pretty good guess).

    If you want to get fancy, you can also look at mean distance between local minimums and maximums on the historical chart in the resolution that you are intersted in.

    Cheers,
    bbc
     
  6. No... like the guys above said you can SEE when things are relatively more choppy... but to get an even better picture of things, you should try and get a handle on how well you are trading the chop... $'s are $'s and if you can profitably navigate through relative chop, then why not do so? Hence the need for a retrospective assessment of how well you are navigating the chop... all that matters is that your equity curve (i) does not itself become choppy (ii) maintains a positive slope for a multiple of the time that it endures a negative slope...

    Some people say that you should avoid trading chop... I say that such an assertion is not so clear cut, and is contingent on the trader-specific capability to navigate relatively choppy environments... lets put it another way: imagine 10 fields of grass, with the first field being free of anti-personnel mines, the second field being full of mines, the 3rd being free, the 4th being full etc etc... at the end of each field is a pot of gold... you can only get to a mine-free field by
    a) navigating a mined field
    OR
    b) waiting 10% of the available mission time for the mines to dissolve into the soil

    You can get 5 pots of gold by just playing for b) each time... or you can play for a potential 10 pots of gold a) and b), by navigating the mined fields... you may lose a limb or two on the way, but if you are good you could get 10 pots if you do the a) and b) combo versus the b) alone strategy...
     
  7. You need to decide which time frame you want to trade and work from there.
     
  8. Try the ADX indicator.
     
  9. or ADX/R
     
  10. on a discretionary basis: flat MA means choppy. absence of HL/HH or LL/LH. on a mechanical basis, is suggest something along the lines of blueberry. pick some dumb MA crossover systems like 5 and 20 (depending on your desired trading frequency/timeframe) then merely run it on the markets you wanna explore. the ones with best equity curves are least choppy. there:)
     
    #10     Apr 14, 2003