Let's talk about adding to winners

I think everyone understands that adding to winners should be done with equal or smaller size, since adding bigger size will raise your average cost by too much and a small pullback will turn it into a loser.

I know that Anek said adding to winners is the most powerful money management strategy around, however he said this in the AHG1.0 doc and as I understand it they don't trade that way anymore. They may still average in, though. They said that even if the last add or two end up being losers, overall the trade will still be a big winner.

That makes sense to me if you manage to catch a big trend that has many HHs and HLs and you add at each pullback (assuming you can tell when the pullback is over and the trend is about to resume).

Does adding to winners only work if you can tell when trends are coming? Seems like if price doesn't make a big trend, adding to winners is a losing strategy.

I never have any idea when a trend is coming or how long it's going to go. I have a big post on my blog about why I never add to winners, but I'd like to talk about this strategy and see if I can understand what I'm supposedly missing. Or is it one of those things that sounds good on paper but isn't really helpful in real life (like trailing stops)?

Question: when you add to winners, how do you know when to close the trade?

A violation of HH/HL doesn't seem to work. Price seems to go wherever it wants and doesn't care about making a LL in the middle of an uptrend.

I've also often noticed that a LL after an up trend, even if it is the start of a downtrend, ends up being so far away from the high that your winners turned into losers if you wait for it.

So unless you happen to get a textbook trend, how is adding to winners profitable?

 

http://www.youtube.com/user/SimpleAndHard?feature=watch

The link is to vids of a style where they add to winners.

Gives you an idea of a strategy used using 3-2-1 adds

 

Adding to winners is profitable when you get good vol moves in your favor with little retrace along the way.

Different strats do well in different conditions. No one size fits all .

 

It's interesting that you mentioned this.

When I was studying jjrvat's method with HH, HL, LH, and LL and only taking trades in the direction of the trend, one of the things I thought of trying was adding one contract on each pullback and not closing any of them out until the trend changed, but it didn't seem profitable unless like you said you happened to get textbook trends.

My question would be, how do you know when a trend is ending vs. just pulling back?

 

And the thread got moved to a forum with low activity. Oh well

 

I don't think this conversation will prove fruitful unless everyone specifies the time frame they are speaking of.

 

well in the one instance adding provided me with the best results (flash crash 2010),I only had a few minutes .

 

You'd only know it when you are already in the position.The key is to catch the top/bottom as near as possible to it.You then just out or scratch if its a pullback or hold if its a new trend.

And why would one ever want to add to winners anyway?

 

I've looked at that before, too, and come to similar conclusions.

 

Can you give specifics? How do you know only if you're already in the position? How do you catch the top/bottom as near as possible to it? How do you know if it's a new trend in order to hold?

In certain circumstances it can be profitable.