Anyone use an average up/down strategy like the following with success? For example say you enter e-mini future (ES) short at 1000 when your indicators/guess said to go short and the price ticks up to say 1002. Instead of closing out the position with a stop loss at 1002, average up by going short 2 additional contracts. Math is 1000 plus 1002 x 2 = 3004...and not considering commissions 3004 divided by 3 = 1001.33. Then say it went against you an additional 2 points at which point you go short an additional 4 contracts. Math is 1000 plus 1002 x 2 plus 1004 x 4 = 7020..... and 7020 divided by 7 = 1002.86. If you then ran out of money to short more you close out the 7 contracts at say 1006 and your loss is = to 1006 less 1002.86 = 3.14 x 7 x $50 = $1099. If you are convinced your original decision to go short is valid the advantage of averaging is you don't get stopped out early, only until it goes 6 points against your 1000 entry price. And when/if it turns before reaching 1006 you are never that far away from a profit or breakeven.
Larryb, read my reply at your other post. Anyone that uses this system is headed for disaster.............
there are many successful traders who will scale into positions...does that mean they add to losers, yes...if you can perfectly time every entry, congratulations to you, but I doubt you can do it...And this whole notion of taking small losses all the time...fine, it works sometimes depending on the volatility, other times you have to bite the bullet and give it some room... BUt I am not here to criticize what you use...nor should you be so quick to judge a strategy which you really know very little about...remember we are talking scaling in, not sitting on a trade for three weeks hoping, praying that it comes back...this is a big difference between scaling in and averaging down...
Vulture, I'd love to meet someone who "can perfectly time every entry".. Don't think I ever will. My opinion is directed to every trader out there who reads these posts trying to pick up techniques to help them in there trading. If the message gets across that this is not for beginners, only for very experienced, great. If you use this technique Vulture good luck to you. Personally I think there are easier ways. You say small losses are ok sometimes unless there is volatility. If there market suddenly got volatile due to an announcement I'd be glad to be stopped out. That's the last type of market I would like to trade
In all the backtesting I have done it doesn't make sense to scale into positions. If your strategy has a positive expectation and you get an entry signal, the best strategy is to put your entire position on. If more often than not the market is going to go in your favor (because your strategy has a positive expectation, right?), then, averaged over many trades, the profit from having your entire position on immediately will outweigh the losses avoided by a strategy of scaling into a position as the market goes against you. The same goes for exits. I don't understand taking half the position off after x points of profit and letting the rest of the position run. Backtest which exit is more profitable and take the entire position off at that point. Scaling into and out of positions does make sense, however, if you are trading so large that the liquidity just isn't there. But this isn't the preferred strategy.
I agree with Aaron. That too has been my experience. However, I think that there is nothing wrong with adding more leverage to what is historically a profitable entry, especially if you are getting an even better entry price. However, if your historically profitable system tells you to flip sides after you have loaded up the boat, you need to be prepared to take that loss right away at that time. Like Aaron said, if the liquidity is there, why not take the full position from the get go. My guess is that taking this extra leverage would improve the system performance, and might even help imrove the equity curve monotonicity, but would not improve your returns as much as doubling up on the initial entry signal. That has been my experience.
Aaron, I have done a considerable amount of backtesting on the index futures and I concur with your statement that putting on the full position and exiting the full position are the optimal strategies, at least if your system is any good. I know some systems gurus and self-appointed experts advocate scaling out of winners, and while the technique has surface appeal, I could never quantify an advantage from doing so. This may confuse some people as I also have advocated doubling up on losing trades for those who know how to trade. I think there is a simple explanation. For me, doubling up is purely a discretionary technique, designed to rescue a losing position under conditions that for whatever reason, I feel are favorable. I really don't consider it "adding to a losing trade", as I view the entire position as the trade. The risk:reward is favorable, as I am not going to let the position run much if any against me, but I can recoup my entire loss if it goes my way by only half the adverse move.
uhh....averaging into a position is one way to look at your system; market making is another way. what you are proposing to do with the e-mini is the same thing that naz MM's do with stocks. buying and selling .05 up or down.
on a larger scale this is why "commericials" end up getting longer and longer into market declines (see the COT reports) not because they are stupid or have inherently better information. during a cursory glance of short-interest stats for stocks i noticed that the short-interest % rose as the stock rose, implying the market makers in these stocks are satisfying demand by going short. averaging into positions like this is a valid strategy: you just have put on way less size at each entry point than you would normally. as for scaling out - i too found that in system testing scaling out of positions did not improve results. however it remains "valid" in that you could look on it as taking a shorter-term and longer-term trades at the same time each with their own exit points - thus you get "time diversification" of sorts...