Averaging down...

8th May '16 by marben100

38 comments 11563 reads

"Never average down" is an old trading maxim (as illustrated by the linked article), but is it a correct one?

Contrary to the article, I do quite frequently "average down"... however, I don't see it that way and, as with many maxims, there is grain of truth, which I will try to elucidate in this short article.

Let's consider what "averaging down" means: it implies that the reason for the purchase is to achieve a lower average buy price for the security you have purchased. That in and of itself is a pretty bad reason for a purchase. Why? Because it implies anchoring bias and could also be a consequence of loss aversion. Cognitive biases are the enemy of the successful investor or trader and we must constantly seek to fight them. There is no logical reason why the purchase price of past purchases of a security should have any bearing on a purchase decision you make now, and it is important to try to avoid being influenced by past trades you have executed in the security.

What does matter is any change to the valuation of the security due to newsflow or other factors (e.g. average sector valuation changes) and how any such revised valuation relates to the price that is currently offered for the security. I.E. has the price fallen or risen by more (or less) than is justified by the change to the valuation of the security? Under those circumstances a purchase (or a sale) may be merited.

Under those circumstances, I would consider a purchase to be a "top up", rather than an average down. The criteria for topping up are exactly the same as for making an initial purchase (with the one caveat that you shouldn't breach your own rules for over-concentration in any one security). Just as when making any other purchase, topping up can result in excellent gains - if your judgement is correct, or losses, if you get it wrong.

So, in conclusion, I consider that "averaging down" is a bad idea, but "topping up" can sometimes be a good one.

Happy investing!

Mark

Unlock with Email

Unlock with Google

Already have an account?
Login here

Rusty2

10th May '16

What is worse losing 30% on a loser or 30% on a winner?

I would say on a winner because you will lose more money.

Mark Simpson

Stop Losses can be a good thing, contrarian and irrational in the sense that an investor completely without behavioral bias does not need them. In fact they would cost them money since they would increase frictional costs due to trading. It's just no one is perfectly rational in their decision making.

So you can say that stop losses are a rational way for you to overcome your behavioral biases (undoubtedly a good thing) and they are not as widely used as they should be. It is just that most stop loss proponents over-reach by suggesting that everyone would benefit from stop losses.

I'm a big fan of using investing rules to overcome behavioral biases (see:

http://www.dangercapital.co.uk/2015/08/RuleYourPortfolio.html )

just that the particular rule of using stop losses doesn't seem to work that well for me.

GrindertraderUK

Herbie. losing 30% on a winner when I was 50% in profit means I have closed a trade with 20% profit. I have increased the funds in my account.

Losing 30% on a losing trade means I have decreased the funds in my account.

Disco Ball

I always believe we have cash capital and emotional capital. I obviously don’t want to deplete my cash capital and likewise I don’t want to chip away at my emotional capital. Holding onto an underwater position is the best way to reduce my emotional/confidence capital so I keep my loses small at no more than 20%. Holding a load of shares in the red is just depressing and what’s even worse is trying to explain to your missus, why you’re 40% down and why you ‘think’ your oil exploration/banking stock is bound to bounce back.

Thats one way to look at it. But if you look at your overall portfolio, a 45% loss on one share is more than 30%. I think people ignore winners that are falling too much, they don't bother because they are in profit but you are still losing money.

well I am not sure many investors with an investment/trade plan would allow a winning trade to fall by 45%.

Anyway I knew I shouldn't get involved in the stop loss argument. I am just glad I am part of the minority that use them successfully, cutting losses letting winners run and compounding my gains successfully.

Its not fall by 45% its 1 x 1.5 x 30% = 45%. All I'm saying is you should review your winners, maybe use a trailing stop loss of say 15%-20%, don't lose all those profits. Believe Robbie Burns uses a trailing stop losses of 10% which I found to be too small.

As far as I am aware, he doesn't use trailing stops ... I asked him.

Hi Ian..

Yes but the long-term winners are so far higher that stops don't really come into play much. On newer trades I definitely exit - you can always buy back lower!

cheers robbie

vik2001

theres 2 sides to this argument as herbie and grinder have both showed, guess it comes down to your judgment.
if I was in profit by 50% on a stock, I would set up a 20% stop loss on that profit (as per what I always do) if it goes up great, if not the stop loss took me out still at a profit.

I got done however using stop losses as well, eg as part of my NAPS I bought Cambria Automobiles, it went down by 20% and the stop loss took me out, a few weeks later as in today the share went up 10% on great results, which I was pissed of about, as the stop loss already took me out.
I could not have used a stop loss and averaged down which would have been great, but hey goes to show no one rule works.

OK from his book it seemed he did.

Yes its swings and roundabouts. I had a stop loss on Interserve (LON:IRV), @410, got stopped out, shares then went up to 445p before crashing down, now 298p, so that has worked for me. One thing a stop loss does is it takes you out where you maybe inclined to hold on and on waiting for a recovery which may never come, that's what Ed is saying, people hold their losers longer than winners. Even if I have a stop loss I often sell before it gets to that point, as in Trakm8 Holdings (LON:TRAK) I got out really fast then bought back in when it had bottomed out.

PhilHofGrowYourDough

Well, if you buy stocks that are showing signs of breaking out then a stop loss makes complete sense as a breach tells you that your initial selection criteria appear to have failed. Time to get out and find another!

I don't want to sit around for months as a stock stagnates or worse sinks. I want a quick positive move and for it to last for an extended period, i.e. > 1 yr

Edward Croft

I don't want to sit around for months as a stock stagnates or worse sinks.

Absolutely - one of the costs that nobody ever seems to talk about is opportunity cost. If you hang onto losers you are gambling that they will come back to parity (or better). But the reality is that most loser stocks tend to keep losing for quite some time. Whether that's due to momentum, disposition effect, overhead supply or other effects doesn't matter - it's a fact.

The cost of remaining in an ill performing stock (aside from the risk of further losses) is a cost of opportunity. Why wait for loser stocks to recover when you could be redeploying and riding the coat-tails of institutional money flow into better prospects ? As private investors we're nimble... we don't have to be head in the sand camels. Nick Train and Terry Smith might 'never trade' but they manage huge amounts of money and don't have liquidity - thankfully most of us don't have this problem.

There's a lot of people who criticise this kind of approach as 'not serious' or 'not proper investment' or 'akin to speculation' - but they tend to be the ones hanging onto their loser stocks for years and wondering why they never outperform the index.

Tis just the humble opinion of a recovering head-in-the-sander...

What about your winners how long do you let them fall before you sell, is it still when they get to -10%?

cig

13th May '16

I've run a simulation of a pure (trail) stop loss strategy on Quantopian to test the hypothesis that "loser stocks keep losing for quite some time" on average.

Within the limitation of the exercise (data 2008-2016, SP500 only, possible bugs in my script) it seems unconvincing. It seems to hug the index most of the time, with a slight drag probably due to trading costs (that the platform simulates), without reducing volatility in a visible way. It even seems to outperform a little in strong uptrend periods, which is unexpected. Perhaps it is too naive to capture negative momentum effectively and ends up capturing some upside momentum (by virtue of eliminating stuff that is obviously down?).

In 2009 it just sold everything which turns it into a broad market timing rule, which is prone to miscalibration where it ends up selling too close to the bottom and goes back in higher.

Code: http://pastebin.com/T1ttcE5b

Some suggested stop losses are good for leveraged strategies, but placebo risk management can be worse than no risk management as it brings complacency without the actual protection. If I wanted an economically leveraged stock rank strategy, I'd truncate the universe for intrinsic leverage (e.g. debt > equity or beta > 1, or something like that) and then pick the best stock ranks from there. No risk of being wiped out that way.

mpat89

15th May '16

Apparently there's emperical evidence showing high beta (i.e. leveraged companies) stocks are systematically overvalued as investors pay a premium for high beta as a means of gaining leveraging.

Steve L

Beg to differ about "opportunity cost" it's an argument based on sleight of hand. If you bought into a stock thinking it's going to go up and it doesn't, why do you think you will then be able to sell out at a loss and pick out another that's just about to continue to go up? "Timing is of great importance to the speculator because he wants to make his profit in a hurry, The idea of waiting a year before his stock moves up is repugnant to him. But a waiting period like this is of no consequence to the investor". Ben Graham, The Intelligent Investor, chapter 8. With a basket of stocks some will move quickly, some will move slowly, and just accept that that is so.

Paul Scott

There's a place for Stop Losses - if you're trading (which is what most people actually do, when they mistakenly think they're investing).

Speculative shares - yes, a stop loss is a great idea. It means that you get out of story stocks when, when the market has realised that the story is bogus. This can save us all a lot of money, if we're stupid enough to buy story stocks in the first place.

Value shares - more tricky. After all, who cares what essentially random factors made you buy? So why does it matter if the price goes down 20%? All that matters is whether your research was right or wrong. IF your research was right, it's now spectacularly right, 20% cheaper, surely?

So, for me. Each buy or sell decision is entirely independent of any previous buy or sell decisions. That's the key point in my view.

Anyway, whatever works for each person - keep doing more of it!

Key point though - most people think they are investors. In reality, most people are actually traders. They are reacting to relatively short term price movements, and judging their performance on that. That makes you a trader, not an investor. What people are actually doing, is trying to anticipate short term market price movements. That's not investing, it's speculating.

Regards, Paul.

17th May '16

Yes, but same (and probably worse) for manually leveraged investors, they underperform as a class their unleveraged peers. Just saying if one really wants leverage, the intrinsic variant is likely safer and cheaper. Leveraging Stock Ranks in particular includes some protection, e.g. should pick up the less overvalued among the higher betas.

dscollard

couldn't agree more: I have 2 distinct methodologies and rules for trading and investing with documented plans for each. On occasions I will "trade" investment positions in my portfolio to take advantage of price dislocations and the relative difference in timescales. Trading has rules on volume, liquidity, ATR, technical behaviour: all the kinetics and momentum of price movement: investing is much more about the intrinsic value, prospects and the underlying macro to support thematic fit and future value

Averaging down...

Unlock the rest of this article with a 14 day trial

About the Author

Marben100

38 comments

Unlock the metrics for all these stocks and more...

Unlock the metrics for all these stocks and more...