treydog999 

I tested several different kinds of trade management techniques using 6E over a 10 year period. My samples were 39157 as the lowest sample size as large at 130785 trades. So this is definitely approaching the law of large numbers. Guess what?? Trade management DOES NOT MATTER. All trade management techniques converge to 0, without slippage and commissions. I posted the full results here Coin Toss Experiment (Strategy) I have decided to compare on the same spreadsheet my experiments. This time using a 1 minute chart on 6E. the standard 15 ticks remains the same. I tested the following. 1:1 2:1 1:1 and 2:1 scaling out 1:1 and 2:1 scaling out with 15 tick trailing … since it made more sense there and continues with the work there. Basically. Trade management only affects 2 things: Average winner/Loser ratio and % win rate. But they trade off in such a way that the expectancy always converges to 0. I guess you could actually do random trade management and come out the same way. The entry system is going to be the key component to the edge. I am starting to think that people who say that they scale out to control risk, or scale in to build big winners. its all a bunch of hooey since the other statistics in turn shift to create a net 0 change in expectancy. The only benefit is psychology.

And/Or the exit system, as your exits are as random as your entries. (Instead of If CoinToss = +1 than BuyLong you are using e.g. If CoinToss = +1 than Sell at CurrentPrice + 15 Ticks OR Buy at CurrentPrice - 15 Ticks, whatever comes first. There is no difference in randomness if the market moves randomly from the point on you are in the market.)

Hi just noticed your thread, I didn't go through all the replies in details; but I thought I would share my opinion nonetheless.

I personally believe that one shall exit from position only if receiving a signal that the pattern or strategy at use is failing... Otherwise, you shall not exit and should keep waiting until you either reach your target or get that failure signal at a later stage.

Your target can also change along the way of course, there is no harm in this and you might exit earlier than initial target, but that's ok too as long as you have a methodology and strategy that gives you this signal.
Let's say you are targeting the 20SMA for instance, and with time that 20SMA obviously moves down little by little and thus you update your profit taking order as you progress...

I don't know what strategies you use and how you generate your signals to take a trade, but this is not so important in itself, because if some parameters and setups on the charts would convince you to take a trade; there would be for sure an anti-setup as per your same strategies that would convince you to avoid the trade or take the opposite side if you were completely out of the market.

These are the signals that I wait for in order to invalidate a trade. In a sense that if I am long, I don't exit the position unless there is a setup that would convince me to take the short if I were completely out of the market.

In terms of money management, I do not scale in nor scale out on perhaps 90% of my trades. I only do it on my equities which I trade very long term and add each month to my positions, but this is more investing than trading and I guess is out of topic here.

This approach to go all in all out is also I believe described and recommended by Linda Raschke in her book "Street Smarts"; but many other traders and statisticians recommend it too of course.

What I do is have an initial stop and initial target when I first open a position mainly for security reasons if a connection fails, PC fails, no power, or any mishap of this type... I will know that at least the exchange has my order and it will get executed. Other than this, I just watch the charts and price action, and if I see a failure signal I just hit close position.

That would obviously happen at any price between my original stop and my original profit target; and it will be perfectly OK for my risk management plans since if I am losing I would lose up to the initial stop which is max 1% of total account, and if I am in a winning situation then any gain is welcome small or big doesn't really matter.

But I repeat, one shall exit position if a setup failure signal is generated, on not on the first pullback or something...
This complicates things for automatic strategies of course, or perhaps makes it impossible to implement, I don't know.

Either case, I often recommend against automatic trailing stops, they are dumb stops that do not read price action, do not understand charts nor indicators, market sentiment, news, volatility, you name it.
I prefer to trail my stop manually, placing it under support for a long position (or above resistance for a short position) and at a price that I am sure it won't hit unless the pattern is failing - and if that's the case I will be more than happy to step out, or if in front of the PC at that time, more often than not, I would have already hit the close position button anyway.

Finally, to conclude, and if you absolutely want to use those automatic trailing stops, what I used to do in the past when I tried them for a while, was to trail the stop keeping in mind the average true range of the time frame I am trading, and then when I reach a certain level of gains, usually equal to the initial amount i was risking, widen the stop and update frequency to double the original setting. This can be easily programmed in ninjatrader as an ATM strategy; but I am sure many platforms do too.

So basically, if my original stop was 15 ticks and that's ok and not too tight as per the ATR, I would trail the stop until i am 15 tick in the money, and at that stage the stop widens automatically to 30 ticks, and would allow room for natural market breathing. This logic was coming from the fact that after the initial move, price has to range sideways and consolidates and thus needs more room to breath. At the same time, I would in worst case scenarios lock in some profit nonetheless equal to my initial risk I was willing to take, so that was fair enough.

I do not use these trailing stops anymore noways.

From the perspective of a scalper, I manage my trades based on whether I think my edge still exists within the market. This is very similar to what is more commonly referred to as "when price action invalidates your set up".

When I define my edge, I look at momentum and time as the two most important variables. Thus, I will be willing to scratch a trade when these two factors goes against my initial reason for entering the trade.

A scratch trade in my definition is a "loss". It might be above my entry, or below. However, (as a scalper) I make the effort to take the scratch whenever I feel that my edge is lost.

My stop loss is fixed below my entry. It can be more correctly termed as a "catastrophic loss", when sudden market volatility moves price adversely from my entry. I make the effort to protect my stop loss from getting hit. For example, I actually view it as a "bad trade" if I allow my stop loss to get hit in a moderate to slow moving market, as I should have exited earlier (by scratching with a Limit Order).

In short, in my opinion, I think there is no real right answer to trailing a stop or having a fixed stop. What is more important is how you DEFINE YOUR EDGE. It can be objective (with the use of indicators, crossovers etc.) or discretionary (reading order flow etc.). The moment you find your edge waning and probabilities do not stack in your favour, it is time to pull up the stops or exit the position.

I know this may seem very scalpish, but I believe it applies to swing trading as well. If you define your edge as a swing trader by price action trending, when the trend stops (objectively or subjectively defined), it is time to pull up your stops. It might seem a little harder if you hold 4 hour positions and have to constantly monitor them, but perhaps that is why I choose to trade in minutes .

Just my 2 cents (and my first post in this wonderful forum!)