Another word about Martingale...

Discussion in 'Trading' started by Liberty Market Investment, Sep 26, 2019.

Do you do martingale/averaging?

  1. Yes

    4 vote(s)
    66.7%
  2. No

    2 vote(s)
    33.3%
  1. We all know that it’s best not to average positions against trend. One thing is to hear that another thing is to withstand temptation to actually do it. Why do people rely on Martingale is a last resort?


    It is widely believed that some Martingale-based trading systems are superior to others. The overriding factor behind all Martingale variants is the money management, or specifically the 'death sequence'. No amount of testing can determine (without a prohibitively high amount of statistical error) the likelihood of whether an extremely low probability event will occur sooner, rather than later. From this standpoint, all Martingale systems have an approximately equal probability as to when they will implode, and of course (because of their inherent recovery algo) all of them will remain profitable until that point. Hence all Martingale systems are essentially created equal.


    Every Martingale variant aimed at mitigating risk incorporates a tradeoff that has a commensurate downside:


    • Using a more benign sizing progression (than doubling) cripples the recovery rate, i.e. necessitates >1 winning trade to return to breakeven.

    • Limiting the number of progression steps does likewise (e.g. if we cap the number of steps at 5, then it takes 31 wins @ 1 unit, instead of just 1 win @ 32 units, to recover from 5 consecutive losses).

    • Starting the progression with a tiny position size (to allow room for more doubling steps) reduces return in exactly the same proportion.

    • Withdrawing money as insurance against the death trade reduces the funds needed to accommodate a given number of progression steps (and of course there's no guarantee that the death trade won't occur before the withdrawal). And if position size is based on account equity, then it will lower future returns also.

    Another widely believed myth is that progressive staking systems can somehow improve overall expectancy. The mathematical reality is that they merely create a higher average bet/position size, which (exponentially) increases risk of ruin. The longer you trade using martingale, the greater the risk of encountering the 'death sequence'.


    Let’s look at what would have happened to a typical retail trader’s account had he been averaging his long positions “in continuation” of the previous strong bullish trend.


    [​IMG]

    As you see, even if you’re guided by volume analysis, you could have averaged your position at least 3 times after the original entry! Needless to say you account would have been ruined long before the current downtrend ended. In fact, market did not reach your original entry price even in a month time!

    [​IMG]

    If you have a already have a strategy (entries+exits) that delivers positive expectancy, then martingale will not improve it; and if you don't have such a strategy, then martingale will not provide it. Or to put it simply, it's better to trade profitably and safely, than profitably with unnecessarily high risk.
     
    MACD, ElectricSavant and kellys like this.
  2. MACD

    MACD

    Works for me but I am always hedged.