If you're Martingaling because you believe price always comes back to a certain level to give you a profit, then you don't use stops; you're averaging until price comes back. There is the possibility a trade may take you out of the game. A trader I know once put a position on near the end of a day on an extremely overbought trending stock, looking for a nice pullback the next day. In pre-market the following morning price suddenly moved 50% against him in about 2 or 3 minutes on a news release. He was able to escape with a low six-figure loss. If he'd been Martingaling into the position as price trended strongly throughout that week and had on significantly more size than he had, there was a strong possibility the hard stop would've been his broker automatically liquidating his position and requesting additional margin to cover the rest of the loss his account couldn't cover. Price itself signals reversals and there is no reason to average down unless you're managing such large sums of money that you have to average into and out of positions. Averaging down as trending instrument pulls back to the trend line is a commonly-used strategy. As far as I know, most institutional investors managing huge sums of money are trend followers, not counter-trend faders, though I'm sure there are plenty of hedge funds that employ the risky strategy. Whitney Tilson's fund had built a NFLX short into their largest short position over a period of time as the price just kept trending against them week after week. They covered a couple weeks ago not far from the all-time high (so far). The easy money is made in the direction of a trend. Averaging up makes sense because you're using existing unrealized profits to leverage yourself in the direction of price movement. Previous resistance in an uptrend tends to serve as support. Why? Because traders watching how nicely price broke through the previous resistance level prior to that, would like to buy somewhere near there or slightly higher, finding that to be a "safe" entry. If they buy near previous resistance, they are getting a "bargain" price in the trend and if the trend fails to hold up, they have a clean level at which to take a small loss and re-evaluate the price action for the next move. If the trend resumes at that level and they then add to the winner as price breaks through the previous high, sure they've increased their cost basis, but they now have twice the size on in the direction of the move. If price pulls back down to the next "previous R becomes S" level, they look to add again and if the trend fails to hold there, they can exit for the profit on the original position and a small loss on the remaining positions, with the end result being a profit. If price holds up there, they now have 3x size on as price moves to break the previous high again, which is how strong trends work and why it makes sense to add to winners rather than add to losers if you have a very large account to trade with.
I don't get the point of the topic regarding existence or non-existence of the edge. Gaining edge is not that difficult at all, there are many of them, even available for free as complete trading strategies, and most can be studied quite simply. The fact that most lose money in trading is derived not from the difficulty of gaining an edge, but from difficulty of getting the proper mindset to trade correctly. And abilite to handle small losses like normal part of the business process is one of those key psychological elements. BTW, this thread is a very demonstrative example of how hard it may be for people to accept that simple fact that decent profits without risk are rare occurence, to say the least.
02-25-11 01:34 AM Quote from nLepwa: ... Now if you have a stationary return process and you know its parameters (mean/std) then that's a different story... Would it be accurate to rephrase this as follows? "If your mean reversion edge works, averaging down can make sense." ----------------------------------------------------------------------------------------- Yes above is a very different story. But what Maverick said is correct. If you doubled down to zero, the amount that price needs to get back to and the time it will take will still equate to what you would have earned risk free. Run the numbers to prove it.
Unfortunately, the OP has a very limited and largely wrong understanding of martingales. On top of that this discussion has started from the wrong point; it is essential to establish the nature of the underlying process first and only then one could possibly apply the notions such as "martingales" to it. It is exceptionally rare to see any serious research published here in terms of the suitable conjectures that could be used as acceptable price behavior models. Without such models discussions related to their attributes (such as martingales) are pointless. Cheers, MAESTRO
the mean price in the graph is 1100,with the belief in returning to the mean as illustrated in the chart ,the turning points as it staggers up or down are your decision points,there are a lot of ways to trade it,depending on holding time, dictated by your acct size and risk comfort,apply this to shorter time spans
I agree, So accepting risk actually is placing your bets (gamble) and exiting often...rinse and repeat and you can catch the profit after a few tries netting a profit. More and More...I see trading as gambling as there is not any control of the outcome. But by accepting this above (italics)...it is "not gambling" Change the way you trade...Change the way you think... This OP's qwest to martingale his way to an edge...is just denial. ES
applied to a stock like nflx, the future may look like this and you could look to take profits and reenter on the bounce or just hold and add on the bounce according to the size of your acct and comfort with risk