Paul Tudor Jones V. Dr. Alexander Elder

[jackb 10]

Just got Elder's latest email and I was left a bit slack jawed after reading it. Here's some pertinent excerpts:

 

Today I would like to look into one of the techniques popular among experienced day-traders – doubling down on a losing position. This brings down the average cost and allows you to get out of a losing trade at a profit. This technique is absolutely deadly in inexperienced hands – all it will do is double up your losses. It is only recommended for experienced traders with excellent discipline. First point – doubling down is only recommended for relatively small-size losing trades. If you are in a large enough trade that the loss is stressing you – then you absolutely should not add to that trade and increase your stress level. Doubling down is acceptable only when you feel relaxed and unperturbed, even while your trade is under water. This is a key factor that almost everybody overlooks – doubling down is not just a technical trick; it is acceptable only for relatively small, unstressful trades. Next, the time to double up is when your losing long trade is hanging just above a bottom at which you are determined to draw a firm line. You must say: my hard stop goes here, and if my trade slides down to this level, I am out, automatically.

 

To repeat my rules for doubling up:

• It is acceptable only for relatively small positions. If the size of your open loss has you feeling stressed, then definitely no doubling.

• Double up only after you see good technical signals

• Place a hard stop on both halves of the trade as soon as you add the second position – a must!

• Handle both positions as a single trade, monitor its average price, and get out as soon as you cut your loss.

 

Contrast this with Paul Tudor Jones' statement: "Only losers add to losers.:

 

Let's see how TL members feel about this topic...

[Kiwi 16]

There are two parts to it:

 

- It is something that can cause people trouble if it is an emotional response to a currently losing trade rather than part of the plan.

 

- To add it to a plan you need to do the stats. Entering low will have a poorer win rate than entering higher (you miss all the winners that take off from a higher point) but in return your loss is smaller and your win is bigger. So, as with each potential entry do the expectancies. Also ask yourself: if entering initially has a win rate of 65% then given a low entry will have a win rate of 40% or 30% (say), would I be doing this trade by itself if I hadn't already invested in a (now losing) position.

 

If the entry low down isn't something you'd do normally then you shouldn't double up (lower win rate, fewer trades per day, but higher win/loss). Something to test is "what happens if I buy the stop loss + a couple of ticks" on every trade I might have entered?

[BlowFish 10]

Pretty sure there are a couple of occasions in the famous PTJ video where he is buying into a market that is moving against him. Of course I could be completely mistaken. Wasn't there a segment where he is buying a falling Deutschmark all day? Finally stops out with a fairly substantial loss. Dunno it's a while since I watched it.

 

I guess it depends on your definition of 'loser' if price is still in your 'buy zone' (even if lower) then perhaps it is not a loser.

[Jeff65 10]

There is a difference between adding to a loser and scaling into a trade as it moves against you. Adding to a loser is often unplanned and violates your risk-per-trade rules. However if your plan is to risk say 3 percent of your capital on a given trade and your plan is to scale in at 1 percent intervals then this is fine.

[MightyMouse 11]

There is a difference between adding to a loser and scaling into a trade as it moves against you. Adding to a loser is often unplanned and violates your risk-per-trade rules. However if your plan is to risk say 3 percent of your capital on a given trade and your plan is to scale in at 1 percent intervals then this is fine.

 

In which case, you have to accept that there will be trades that go in your favor that have only 1/3 ( simpled version of your example) of your position on. On the other hand, each stop out will be with your max position.

 

So, your winner is going to have to go 3 times as far as your loser to break even on a day where you have a single scale winner and a max loser. That doesn't sound like love.

[Kiwi 16]

That doesn't sound like love.

 

 

Thats why you've got to both do the sums and make sure you can stand the win ratios on those third trades.

[BlowFish 10]

There is a difference between adding to a loser and scaling into a trade as it moves against you. Adding to a loser is often unplanned and violates your risk-per-trade rules. However if your plan is to risk say 3 percent of your capital on a given trade and your plan is to scale in at 1 percent intervals then this is fine.

 

Exactly. PTJ had a right hand man that did the analysis, they would agree they wanted to be long x DM and then PTJ would get busy 'working' the order. If you are a buy side trader working a large order your job is to get filled at the best possible average price.

 

It's pretty obvious. Is it an 'oh shoot' emotional response or is it a planned entry where you are fully aware of the risk parameters throughout the process.

[zdo 20]

It's entirely system dependent. Some systems - Never! Some systems it's ok, even part of the game.

 

For systems where it is ok, fully "Doubling down" is usually not the optimal concept. Adding to / building the position is a more accurate description of the proper maneveur.

 

Also, generally speaking, this maneuver is best implemented where the fills are close by, generally in the same as yet incomplete small and short term swing. It is not recommended when there is already a large disparity btwn first fills and where price is now - ie if the original setup is no longer valid, if it's clearly a loser, de nile is rising, and you should have alreay stopped out. That’s what PTJ was talking about and if you haven’t been fortunate enough to fully learn that lesson yet, let me tell that is a wonderful way to pile a whole heap of sustained pain on yo own head...:helloooo:

 

Find attached a recent example of what I think Elder is talking about. Yesterday, I put on a high leverage very short term hedge of a large percentage of my silver derivative long trades. (Several ‘cycles’ were completing but the tipping point for me was by chance walking by a TV at 5 am and overhearing the network talking heads yapping PM’s up “investors losing trust in yadayada and putting their trust in gold” etc etc. It was ‘intuitive’ - pls blve me, I rarely of rarely of rarely ever use broadcast media for cues) Anyway, the bottom triangle was a mkt order fill at around 29.99. The square above it was a “doubling down” limit order placed shortly after the market fill. The right upper triangle was the fill of the limit at around 30.16 in the next 15 minutes bar. All fills within the same small swing / setup is what I think Elder is talking about generally… don’t think he’s talking about doing the first entry then doubling down if price then proceeded to 31 ( or to generalize, past where should already be stopped out ) . I do similar much shorter time frame type 'doubling down' / position building in the indexes in certain conditions. Happens roughly 3 of the days of any week... may post an example if I have time...

 

Again - It's entirely system dependent.... in addition

the key word in Elder's article is "...experienced... "

Edited December 8, 2010 by zdo

[SIUYA 11]

Personally I try never to double down (the times I have have always been costly if not from the monetary points - from a developing bad habits point of view)

What is interesting is that it is always soooooo easy to double down on the loosers but sooooo hard to buy more when you are winning...... and I think this is the crucial element for me.

The offset for the idea -- "This brings down the average cost and allows you to get out of a losing trade at a profit" - is that it also ensures that if you are wrong your loss will much larger than planned....this to me does the damage, and creates bad habits.

 

If you are looking to enter in 1/3rds then you should be looking to do that for the winners as well, otherwise go all in. In this case you need to ensure the expected return is going to be sufficient to keep entering on winners. This is what most cant do as the need to take a profit is too strong.

Think about it in terms of trying to beat VWAP while building a position- if that becomes part of the plan then it makes it easier to get the head around, as you should look at average prices of entry as suggested when doubling down, so why not look at them when doubling up to have you maximum position size on ALREADY with a profit. (generally not so relevant to strict day traders I guess)

[MidKnight 12]

I have excessively done this when I was scalping the nikkei and HSI. It works outstanding in high volatility conditions (like the start of the subprime debacle). Over the past year though I have altered things do it both ways, and that is to double/tripple down as long as the trade idea is still valid AND aggressively add to my winners. The market is fluid and dynamic, so I need to be as well.

 

With kind regards,

MK

[Jeff65 10]

In which case, you have to accept that there will be trades that go in your favor that have only 1/3 ( simpled version of your example) of your position on. On the other hand, each stop out will be with your max position.

 

So, your winner is going to have to go 3 times as far as your loser to break even on a day where you have a single scale winner and a max loser. That doesn't sound like love.

 

Who says one can't scale in as price climbs as well?

 

What if the average trade nets 4 or more times risk of a full stop out? There are profesionals who scale in. There is a million ways to successfully trade and even more ways to lose.

[MightyMouse 11]

Who says one can't scale in as price climbs as well?

 

What if the average trade nets 4 or more times risk of a full stop out? There are profesionals who scale in. There is a million ways to successfully trade and even more ways to lose.

 

Scaling in as price climbs makes more sense to me. I don't like adding when price is falling. Simply a personal preference.

 

Professionals are professionals until they run out of luck.

[electroniclocal 10]

Doubling down is an integral part of my trading when the entries are outside in trades.

 

I buy bottoms and sell tops against a context that tells me clearly when I'm wrong. Sometimes the entry is a little early and I double down and exit the doubled down portion at the original entry price.

 

Unless the trade is clearly wrong, stopping yourself out because of price shaking the tree makes it hard to make money. Those unnecessary stop outs are expensive. The medicine for too early an entry in a good trade is to double down.

[zdo 20]

Again, the efficacy of ‘doubling down‘ into losses is very system dependent. (and btw adding to winners is a little bit off this specific topic) Be careful of any posts, including mine, that are putting out tactics as if they are ubiquitous across all systems without even a hint of the type of system being used.

 

Blunt example of the system dependence – doubling down into retracements is generally were this works. However, double down against trends and see what happens!

 

Different systems, different setups yield different appropriateness of doubling down. Even Elder needed to be more specific about the system beneath his tactic of doubling down when price neared his initial stop loss. In some systems that would work. In others it would not help, and in some it would actually lead to blowup… without giving us system properties it would even be ok to start withdrawing the benefit of the doubt and to start raising questions about his initial entry placements and order types, etc.

[MightyMouse 11]

Doubling down is an integral part of my trading when the entries are outside in trades.

 

I buy bottoms and sell tops against a context that tells me clearly when I'm wrong. Sometimes the entry is a little early and I double down and exit the doubled down portion at the original entry price.

 

Unless the trade is clearly wrong, stopping yourself out because of price shaking the tree makes it hard to make money. Those unnecessary stop outs are expensive. The medicine for too early an entry in a good trade is to double down.

 

A lot of negative equity on a trade is a really good sign that you are wrong. You can always get out quickly and get back in if the trade is still good. I am referring mainly to day trades where time is of the essence.

[TheDude 10]

Elder = author NOT a trader.

 

PTJ = PROVEN trader.

[jackb 10]

Elder is a trader as well. But, it is true that PTJ's results are certainly more public than Elder's and PTJ is managing other people's money which I'm not sure Elder does.

 

There's some real interesting perspectives being posted, as I hoped there would be.

 

One thing that seems to be running through a lot of the comments is "My definition of a losing trade is different than yours"? In particular, if I'm scaling in and haven't put my full position on, then I'm not yet able to really assess if it is a loser. I'm confident this wasn't the context Elder was talking about (in fact, the email I received from him walked through a specific trade...so, I'll go ahead and post that as well). This involves a trade that is clearly a loser.

 

What you'll note is that it comes off as if the double up was a "spur of the moment" decision. And thereafter he had to "draw a firm line" with respect to stopping out of the trade, which implies the trade blew through his original stop point; one that was apparently a "soft line." When I said I was left slack jawed, this is what I was referring to.

 

More importantly, he goes on to say that such a "technique" is popular among experienced daytraders (he didn't comment about whether these experienced daytraders were profitable). And that it's a fine strategy when you have small size on.

 

I think the last point is particularly problematic. If you get used to doubling on small size, that only lays the foundation to double on larger size.

[jackb 10]

Elder:

 

Let me illustrate what I mean by showing a very recent trade – or rather a pair of trades.

 

On Thursday, November 18, I was catching up on my office work using my laptop, while running a few day-trades, one after another, on a big screen. The 25-minute chart of the December Euro (not shown) was in an uptrend, and I bought a pullback below value on the 5-minute chart at 1.3633. Then, before I could grab a quick profit, the Euro reversed and my trade went under water.

 

I monitored my growing loss, expecting a bounce. Finally I observed a bullish divergence and I decided to double up and put a hard stop at 1.3582. I bought my second contract at 1.3584, meaning my risk on the second trade was only 2 ticks.

 

Prices rallied from that bullish divergence and I sold both contracts at 1.3617. My loss on the first contract was reduced to -$200, my gain on the second became $412.50, for a total gain of $212.50 ($201 after commissions and fees). Had I just banged out of the first contract at any time during that slide, I’d have taken a loss. Doubling up at a good spot turned a losing trade into a winner.

[zdo 20]

"...part of the plan..." or not

Kiwi's first response to OP is the best. If 'doubling down' is part of the plan, the conditions triggering it, the number of times you would do it in a trade, etc are all pretty firmed up...

 

re "before I could grab a quick profit" and "observed a bullish divergence and I decided to double up " and "put a hard stop " Yep, jack, it looks like Elder is shooting from the hip a little bit... but fx trading can be a little bit more amenable to that kind of 'looseness' (ie discretionary) for some. I know I trade ndx's way differently than I do fx's - systems, setups, triggers, w/l ratios, win loss sizes, how I 'define a loser', timeframes, and 'doubling down' / adding to position tactics, number of times I will add to a loser etc etc (7 total entries is my limit in ndx's, have total size limits but no max limit to number of entries in fx, etc.)

 

Kiwi provided the keywords - "plan" and "stats". then with those as base, accumulate "experience".

 

...and, re 'emotional' if it is part of the plan for real, 'doubling down' / adding to the position 10 times and still getting stopped out on all of them shouldn't bother you in the least bit emotionally

 

... Also, I get suspicious of any 'doubling down' tactics where the object / intention is to 'get back to break even', etc. - although a few systems are amenable to it net net...

[jackb 10]

zdo, when you're scaling in and the market is below the intial leg, do your subsequent entries always mirror the same size as the original position or do you ever play with the size of of subsequent legs (e.g., 1, 2, 4, etc.)? Reiterating, when the market is below the opening leg.

[MightyMouse 11]

"...part of the plan..." or not

Kiwi's first response to OP is the best. If 'doubling down' is part of the plan, the conditions triggering it, the number of times you would do it in a trade, etc are all pretty firmed up...

 

re "before I could grab a quick profit" and "observed a bullish divergence and I decided to double up " and "put a hard stop " Yep, jack, it looks like Elder is shooting from the hip a little bit... but fx trading can be a little bit more amenable to that kind of 'looseness' (ie discretionary) for some. I know I trade ndx's way differently than I do fx's - systems, setups, triggers, w/l ratios, win loss sizes, how I 'define a loser', timeframes, and 'doubling down' / adding to position tactics, number of times I will add to a loser etc etc (7 total entries is my limit in ndx's, have total size limits but no max limit to number of entries in fx, etc.)

 

Kiwi provided the keywords - "plan" and "stats". then with those as base, accumulate "experience".

 

...and, re 'emotional' if it is part of the plan for real, 'doubling down' / adding to the position 10 times and still getting stopped out on all of them shouldn't bother you in the least bit emotionally

 

... Also, I get suspicious of any 'doubling down' tactics where the object / intention is to 'get back to break even', etc. - although a few systems are amenable to it net net...

 

Whenever I have doubled up on a trade that went against me, it was impossible for me to resist wanting to get out at break even. At the time I did not have the nerve to stay in the trade with the larger trade size and has happy to take the trade off at BE. This behavior is account destructive. When the trade would go against me and i would add and take it off BE, then the best i could do on this trade when it happened was zero profit - commissions or lose what ever my sell stop was with the higher contract number. If you do that trade 1000 times, you are a guaranteed loser since the best you can do is zero profit and the worst you can do is what ever your stop out is on the higher contracts. Removing this behavior turned me into a break even trader.

 

I still don't do it because i never learned how to allow the larger position to continue to the objective, but I do see and know that there are times when this is a perfectly good strategy if you are a guy who can ride it properly.

 

If the objective is to be out by the end of the day, then I prefer to admit failure as quickly as possible and start chopping. I would rather take the small loss and decide if i want to get back in higher or lower, if the trade is still right. I said earlier in the thread, there is no better barometer that your entry might be wrong than negative equity on a trade.

To do this you have to accept that you will potentially be buying or shorting at a worse price after taking a loss.

 

Regards,

MM

[zdo 20]

re "If the objective is to be out by the end of the day, then I prefer to admit failure as quickly as possible and start chopping. I would rather take the small loss and decide if i want to get back in higher or lower, "

 

Yes. Many systems should simply be out way before price gets to a place where "doubling down" (and associated maneuvers) is considered. Example: Under certain conditions, PTJ would take a bunch of tiny losses, then load up. etc.

[zdo 20]

zdo, when you're scaling in and the market is below the intial leg, do your subsequent entries always mirror the same size as the original position or do you ever play with the size of of subsequent legs (e.g., 1, 2, 4, etc.)? Reiterating, when the market is below the opening leg.

 

jackb,

 

it's system dependent. For example: over in Traders Logs

http://www.traderslaboratory.com/forums/f103/v-dow-spread-7706.html

I have been loosely chronicling two trades. In the SI/Dow spread I put on shorts in the

Dow in 25% increments (into strength and or weakness) of the total allocation to the trade (ie equal amounts). Also, (sort of off topic) am journaling a short Yen position, where the initial entries where only a tiny percentage of the size of the recent and any subsequent additional entries I will ideally do at lower levels… the result of progressive scaling is much lower average price than if it had been equal scaling in additions. (Note these long term trades are a work in progress…not playing to my typical strengths … that’s why I’m ‘journaling’ them some. )

In ndx day trading I do up to 7 equal sized entries (and btw don’t do them routinely but only under certain conditions…) hth

[bathrobe 10]

You are much better off stopping out and re-entering if you still like the trade.

[Patuca 13]

in day trading doubling down has saved me from a loss many times more than causing me a further loss. however, it all depends on where it takes place. if price reaches a support level and hold and turns bullish then it makes sense to do it. The basic idea is price has to travel less distance back up to my breakeven point. And it should be before price reaches my puke point stop. For me it makes sense to use an intraday somewhat flexible stop but have a trade management stop at puke level. Of course one could argue just let the intraday stop take you out. How many of you have seen your stop get knocked out and seconds later price is back up and you are in the money if you wouldn't have gotton taken out. With all the HFT going on nowdays the computers algo's create some problems. I prefer to use an intraday trading stop that is adjustible within certain guidelines and a puke point stop that nevers gets moved.

Edited July 21, 2011 by Patuca