Anyone have success with Pyramiding

Discussion in 'Professional Trading' started by lasner, Feb 18, 2010.

  1. lasner

    lasner

    Anyone ever use pyramiding with any luck. Any books that could be recommended would be appreciated.

    Thanks,
     
  2. There are a few books on it - Ryan Jones comes to mind, though I wouldn't recommend it.

    The hard part is to find backtesting software that enables pyramiding (i.e multiple contracts at different prices)

    A software called trading blox does allows this.

    http://www.tradingblox.com/

    Choosing the right software to test this for your strategy is important. the same thing does with rolling into and out of options where the price may go in or out of your favor. Some people say never buy when the trade goes against you, it depends on whether you have a specified level to be buying at, and if the level you desire gets hit, you'll be buying into it.

    Alternatively, some people only scale into (other word for pyramiding) when the profit of the trade increases.

    The main thing is that at any given point, you don't ever exceed the threshold for your risk.

    If you read the turtle rules (search turtlerules.pdf) online, you'll find that one of the main tennants of trend following is to systematically increase position size when the trade is moving in your favor, in specific increments... up to say 3% of capital is at risk in the trade. There will be a series of stops at each incremental level that you will trade around, adjusting the position size in levels (the trend followers call these units).

    In general you'd bet a specific number of units in the beginning, on entry, and then pyramid a specific level into and out of the trade.

    The trend followers risked a max of say 3% for each correlated asset, and a max number of units at any given time.

    You can take these ideas and apply it to a system of your own design through testing and measuring the risk taken on through pyramiding.

    I haven't seen a situation where pyramiding without a profitable strategy would work, though people like Ryan Jones like to suggest otherwise (hes been called the final frontier in snakeoil on here)

    If there is any edge in pyramiding, it is thru the application of risk control on multiple uncorrelated trading strategies. People such as Ray Dalio at Bridgewater Associates have been interviewed and said that trading different strategies through regime changes with risk control(which may involve pyramiding) give them a big edge over their competitors.



    Good luck
     
  3. It "works" best when you're trading in a strongly trending market. :cool:
     
  4. The whole idea is that when the market isn't trending (chop) you'll take a series of relatively small losses. When the market starts to trend, you up the position size at intervals in the trend. It's sound risk management since no none can really predict when a trend will start or stop. All you can do is anticipate by probing the market.
     
  5. Most ppl build an inverse pyramid. They add the same number of contracts as the market moves in their directions. If you do it correctly every additional segment should be smaller. Why because the market pulls back and you risk of loss is greater .
     
  6. 1) Ideally, you don't get caught up in choppiness if you just happen to be in the market that's making new price extremes day after day after day.......
    2) You have to be able to "buy high" and "short-sell low". :cool:
     
  7. Ed Seykota has written a good bit about pyramiding and risk management in general. He's stated (I'm paraphrasing here) that the pyramid imagery is quite literal: As the favorable trend unfolds each add-on is progressively smaller. He's also said that you pyramid up to your optimal position size, not on top of it.

    Here are a couple of links closely related to the thread.

    http://www.seykota.com/tribe/risk/index.htm
    http://www.turtletrader.com/pyramids.pdf

    Keep in mind that there's no right answer about how big the add-ons should be or at what interval to add. This a framework only.
     
  8. lasner

    lasner

    Does anyone have any good boos to recommend
     
  9. Michael Covel's "Trend Following" talks about this some.