Getting burned in August

1st Sep '18 by Muggles

35 comments 5412 reads

Ouch….

It’s been a bad month. The portfolio has lost almost 6% of its value, and 6 of its 20 shares have hit sell rules. In the same period the FTSE has lost only a little over 1%. Further, while overall stockranks have gone down, they have only lost 2.2% this last month.

So what’s been going on?

Sells
Majestic wine
Ocean Wilsons Holdings
Plus 500
Hansteen Holdings
Pennon
CMC markets

Sell analysis
So, is the issue that my sell rule is too tight, or my buy criteria aren’t good enough…. Or just bad luck?

5b8a9583272afScreen_Shot_2018-09-01_at_1

From reviewing sells and reasons, drop in M rank was a good predictor, except for NWF. Maybe rather than 80, a little more leniency to 75 or 70? However, being off a 52 week high wasn’t a great predictor. Usually it felt appropriate as I’d just lost some past gains. However, out of 7 times (I’m excluding CMC as only just happened) I was tapped out by 52 week high, in 5 cases I’d have been marginally better off staying in, in once case it made no difference, and in the other I’d have only lost a further 1%. While the time scale is limited, and it is possible these shares fall further and have just had a slight rebound, I think the 52 week high rule hasn’t worked properly.

I do think, however, there is something wrong with my buy rules. 6 of the shares I bought at the end of June were sold within 2 months, plus two from last month.

What is going on?
I’ve had several holdings tank on, frankly, not even bad news – just not as good as expected. My assumption at the time was that these are fundamentally strong companies, but that are overvalued and so the slightest wiff of company results or market conditions not going their way has cause the price to drop. However, I don’t think it is all about the sell rules. Yes, this has been a bad month for the FTSE and stockranks generally, but the fact I’ve been hurt 3x as much suggests to me that this is at least partly about how I’m selecting stocks.

I’ve tried to see a pattern…

Unlock with Email

Unlock with Google

Already have an account?
Login here

Ramridge

4th Sep '18

Hi Guys - re. Darvas
Oddly enough I came across Darvas's method a few months back and I have been experimenting a bit.

To summarise, Darvas came away with the concept of a box, known as Darvas Box, in which a share price is allowed to bounce at will inside the box over a certain period. If the SP breaks out of the box upwards then it is a signal to buy, and if downwards then a signal to sell. The length of the period can be anything from a few days to a month, but typically 20 days.
The width of the box can also arbitrary, but I have found people to use around 10%. Darvas also said that he preferred volume to be low before the breakout.

This is very close to Minervini's VCP method. I have back-tested Darvas's method and indeed it came close to replicating the VCP pattern. Of course Minervini's method also stipulates other parameters which Darvas does not mention. So roughly speaking, Minervini's method is a super-set of the Darvas Box.

Since I personally use Minervini's system, the Darvas Box does not add any value for me so I have put it in my bottom drawer.

Was he a fraud? I doubt it. Minervini's performance has been independently verified. Darvas on the other hand did not put himself under similar scrutiny. In my book that does not make him a fraud... maybe just careless.

mmarkkj777

Hi dscollard,

I agree. I do both (invest and trade), with a different parts of my portfolio. I have some long term buy and holds with wide stops, or non at all (e.g. Amazon, Nvidia, Apple, Berkshire Hathaway, Airbus).

But as these are very passive, with very little to do other than keep a watching brief and a six monthly review, I also trade and it is then when the closer stops come into their own.

As an example, I recently bought a stock that was rising. Poor fundamentals, but it really had momentum. as soon as it gained 7% I set my stop to 4% below the current price, which meant the 3% gain which was then locked in covered the spread, dealing fee and stamp duty combined. From that point on it was a risk free trade. My only possible loss then could be an opportunity cost if I got stopped out then the price rose again. I'm much more comfortable with opportunity cost than a £ sterling loss. I moved my stop up 4 times before being finally stopped out. I did similar 2 days later with the same stock. I made well over 30% combined in a few days.

It obviously doesn't always work. I got stopped out of Sosandar (LON:SOS) earlier today by a mid-morning dip only to come back in this evening to see that the shares had recovered to over 7% rise for the day. But rather this than to lose a large sum while I'm out.

At the end of the year I'll compare my trading to my investing results and see if all the effort was worthwhile. I've been investing for quite a while, so maybe not a fair comparison, but an important one in terms of future capital growth.

timarr

In the absence of evidence the starting assumption must be that Darvas wasn't genuine. Would you allow a self-certified optometrist to start sticking things in your eyes? Still, you need more qualifications to be a plumber than an investment manager.

Darvas' ideas originated with Jesse Livermore, who was entirely genuine and made several fortunes. He also went bankrupt several times and eventually killed himself. Still, no trading method is entirely foolproof. But his Reminiscences of a Stock Operator, dating from 1923, remains one of the best books on trading ever written.

Hi Ramridge,

I wouldn't even say he was careless, as when he was trading he was doing it in his spare time from all over the world whilst professionally dancing with his sister (not a full time trader like Minervini). Why would he put himself under scrutiny. He had many of the Telegrams he had sent to his brokers, etc. and specific examples of trades he made. He could have painfully put it all together retrospectively, I suppose, but my view is they just didn't want to believe him at first.

I agree, his methods, although still valid and work, have been overtaken by more advanced methods. I've just started reading Minervini.

What I got mainly from the book, strangely, was the way he evolved his trading mentality, the bad times and how he overcame them.

But I just use snippets from here and there and don't try to replicate anyone else's methods completely.

Hi Tamarr,

Guilty until proven innocent?

He was a part time private investor for himself, not a fund manager.

ricky65

Probably at worst, Darvas added another zero onto his profits. That would be a bit disingenuous, not necessarily fraudulent imo. Worth noting that they never tracked down all his brokerage accounts.

Anyway, I think his book provides good lessons for stock traders.

5th Sep '18

Guilty until proven innocent?

That's how most of Bernie Madoff's clients lost their money :)

But it's a false analogy: no one's on trial, but someone is being touted as an advisor for private investors to follow. If I use an advisor of any kind - accountant, osteopath, hurling coach (it's a long story), builder, etc - I either want to see their qualifications or some other proof of their competence or I'll ask for a second opinion.

He was a part time private investor for himself, not a fund manager

If you write a investment advice book, marketed to private investors, making claims about how much money you've made then evidence is required. Otherwise you're just another moonshine merchant.

This site is all about evidence based investing - so if someone starts making claims about investing methods we should ask for the evidence. And if it can't be produced we should discount the claims. It's just commonsense.

Its not a false analogy.

You said a man who is sadly no longer with us was ("probably") a fraud, without evidence. You said he made a fortune from his book. I asked you if you knew how much, because I don't and I know he died a long time ago. I don't know who gets the proceeds from his book..

I said it was proven that he made at least $216K dollars from trading in the 1950s. Part time, while dancing on tour.

His methods have been proven to work. Other, later methods probably work better.

This site is partly about evidence based investing.

This particular part is about sharing ideas and debate (unless someone has commercial interest).

Have you read his book?

It's not really an investment advice book as such (not in the same as Ben Graham's Intelligent Investor is).

It's more a recollection of how he traded, the pitfalls. How he got over confident. Listened to too much noise from the market and how he learned to hold his nerve and not sell too cheaply, in addition to the techniques he adopted. I got more from the former than the latter.

I haven't read Jesse Livermore (but I've heard of him). I'll add his book, Reminiscences of a Stock Operator, to my reading list. Thanks for the tip. I'm interested to learn why he went broke so many times if he was a great investor, and how he came back from the brink each time.

On here, I'm just a private investor. No axe to grind, no commercial interests to "tout", as you put it. So I'm entitled to my opinion. And to the very best of my ability, I will only say things that have helped me and I think may help others. I thought that was what this section was for.

Maybe you should go after Tom W. They gave the Darvas book away for free at the UK Investor Show two years ago.

This Darvas theme came about due to some discussions about wide stops and close stops and their relative merits and uses. As we are well off the original thread, perhaps we should end the debate here. I'm sure others may be getting bored if they are reading it thinking its about a bad August :-)

If you wish to continue this debate, I'm happy to, but maybe off-line, or start a separate thread about Darvas etc.

Fair enough. The finding of the NY court throwing out the Attorney General's case agrees that Darvas wasn't offering advice, or indeed even proposing a usable investing method:

https://www.leagle.com/decisio...

The important fact is, however, that the book does not "advise", does not pretend to foretell advances in stocks. It does describe a method used in specific transactions, actual or fictional, which turned out successfully for the author. I seriously doubt that this is investment "advice" within the reasonable intendment of the Federal or State securities statutes.

'Nuff said ;-)

justinian

Just one more thing....I think many consider the Stockranks as helping to choose individual stocks and thereby end up falling into actively trading them which borders on the technical analysis side of things. That is not 'wrong' as such as there is no right or wrong in trading and investing whatsoever, only results and a statistical opinion as to how lucky you were in achieving any particular set of outcomes.

The way I look at my StockRank based portfolio is about "maintaining the integrity of the asset". By that I mean one has selected a particular set of features with the idea that these will outperform the market as a whole, minimising the chances of not achieving that by ensuring one has a sufficient number of stocks. (Stockopedia has in the past come up with a figure of 25 beyond which there are rapidly diminishing advantages in trading more.)

So, I say to myself, if I want to outperform the market over the next year and where I would feel comfortable owning them, I need to buy 25 stocks with these features. At the end of the year some stocks will win, some will lose. No stock represents more than 4% of the portfolio so one hopes that any individual 'catastrophic' losses are ameliorated by only being a small portion of one's 'fund' and that others will outperform to a greater degree on average. (You could of course reduce the risk further by buying even more stocks so they represent an even smaller %.) Then, at the end of the year, one would review the stocks in the portfolio to ensure they meet the characteristics one is looking for, eject those that don't, buy new ones that do and set the whole thing running again for another year.

I don't use stops on my Stockrank portfolio because all the research I've ever seen has shown that they cost more than they save. And the movements of individual stocks are just so damned large over any reasonable length of time. One relies on not putting all one's eggs in one's basket instead. Plus I use a Bull/Bear indicator on an ongoing basis that I use to tell me when to move to cash.

When I am trading, I do use stops, Guaranteed Catastrophe STOPs, as I hate the risk of gaps. And that is another form of diversification....using different systems with different assets over different time periods. The purpose is simply to smoothe one's overall equity curve.

The Phoenix

Hi personally I find the other Livermore book "How to trade in stocks" even better its old as the hills but in my opinion still the best book written on position sizing and money management.
He generally lost when he listened to others and broke his own rules also had a love of women which I think didn't help plus a destructive personality.

However he was probably the great trader of all time and the original momentum master would make a great film about his life one day.

Regards

Hi Justinian,

That seems like a well balanced and methodical approach.

Like you, I use a different approach in my investing, compared to my trading.

I have only just started to use stockranks to help select my stocks, so I won't know for a while if it helps, but so far it seems to.

Can you explain the Bull/Bear indicator? How often do you move to cash? Just during the major falls, or also on small drawdowns.

Hi Clerea.

I wondered about his bad periods that have been mentioned, and how a great investor could go down so many times.

I guess sometimes even genius traders have trouble with the psychology side and have a kind of self destruct button where all defensive strategies and risk avoidance temporarily goes out of the window.

At lease they have the genius to turn it all around again, though.

Personally, I worry if I loose even relatively small amounts. It happens a lot and I get over it, but it hurts.

I'm definetely going to read Livermore's books.

iwright7

6th Sep '18

Could I just add that the audible book Reminiances of a stockmarket operator, is wonderfully narrated. I have listened to a least 3x on long car journeys. One part particularly that stays with me is revalation he had when talking with a old time trader nicknamed Turkey, about the wisdom of holding in a bull market.

I use just a few variables in a Macro. It doesn't trigger very often. I'm looking to catch full Bulls and Bears. The last time was the "earnings recession" in 2015 and before that the Financial Crisis.

The chart shows the Bear phases in grey. Using the S&P500, the 2 lines represent buy and hold and going to cash in Bears.

https://4.bp.blogspot.com/-QPL7uzvu9Ko/W5DZsD5VSPI/AAAAAAAAUcI/1WGv964JiNUG0zMc7htTBaWfxvJkzqrkwCLcBGAs/s640/bull%2Bbear.png

There are a lot of economic and stock indicators that you could use. US unemployment for one. Many have successfully used the 200 dma on the S&P500 to avoid the big drops (but it does whipsaw a little and I don't trust it as it relies on a specific shape of the stock curve).

I should also give a word of warning. I'm concerned the future might not look like the past. Markets have a habit of laying the best laid plans to dust.

Quite simply, he was way over leveraged. Phenomenal on the way up, but carnage when one loses. He also relied on things that were later outlawed from the markets. Many 'legendary' traders could no longer make money when their activities became illegal.

But without those flaws and his resulting success, we would never have heard the useful bits. Like about 'sitting' making your money.

Gromley

I have been mulling over your original post for a couple of days, wondering whether my initial thoughts were a bit knee-jerk and muddled. I have convinced myself that they were not (other opinions are available!) So here goes.

Reading back a bit, I noticed that you were the author of “A letter to my panic stricken self”.

So, tongue ever so slightly in cheek, I do hope that letter was your first port of call?

In it you set out , rule 1 don’t change the plan ; although you did give yourself permission to tinker around the edges a little.

And that brings me to my initial thought which was to wander why you concluded that the August result (painful as it was) indicated something was broken or wrong?

Down 6% in a month certainly would seem quite extreme to me, but if you were making 2-3% for the other months of the year, you'd be quite content overall wouldn't you.

So was the result so far outside your "envelope" to indicate something broken?

I don't know enough about your approach to even guess at the answer to that, but from memory the main NAPS approach has had a draw down of 15% at one time.

I might have misinterpreted, but it seemed to me that you were looking for what to change as opposed to looking for whether anything needed to change.

I probably cannot help to much with that as I would still classify myself as 'a learner' with regards the very best use of the StockRanks and their components.

I do have a few observations though on your sell criteria.

I would echo the thoughts of others that a 5% stop-loss, looks much too tight. Such tight stops can be right for traders, but if I understand correctly your approach is to be a bit more active than 'NAPS' but I didn't think you were really attempting to "trade".

Based on the numbers you posted, your sells do look reasonably sound, but, I wonder if you are viewing this too early. The real test is surely comparing the price when you sold with the price at the end of what would otherwise have been your holding period? In saying this I would echo the comments suggesting it is worth thinking about whether you are reviewing too frequently.

You also use a fall in the Momentum Rank as a sell signal. It seems that you are far from alone in using a fall in the SR or a component as a sell trigger.

I can see the logic : if at any given point it time the 'best' stocks to buy are those with the highest StockRanks, then it seems to follow that those that have seen a fall in the SR (or component) must be the ones to weed out.

But is it really evidence based? Despite the logic of the above view I do have to note that the SRs and their components seem to predictive even a year out, so it might not automatically follow that those whose SRs fall are unattractive holds. I really don't know the answer to that and if anyone has done the work on that I would be grateful for that as that might save me some work!

(I just did a completely unscientific review of stocks that were SR 90+ at the start of the year and those whose SRs had fallen to 75 or below by late March had lost c. 15% compared with near break-even for those that had held at 75+). Over the next three months though, those with SRs that had fallen recovered c. 15% compared with a c. 6.5% recovery for the more resilient group.

I would not claim that proves anything (although the most obvious conclusion would be to sell those stocks that will see a decline in SR in the next three months) but it might at least say that sell rule is open to question or additional criteria. I would particularly also note if you are using falling/fallen momentum as a sell trigger, momentum is said to have negative correlation to share price performance in the short term (but I haven't seen the "short term" defined).

I only have questions at this stage, no answers; but I hope that is of some use.

Muggles

Hi Gromley

I'm a little embarrassed to admit it, but you are absolutely right. I didn't go back and read my earlier post... which is rather frustrating given that I'd only recently written it! I do remember being aware of it when I was writing the post (and making the decisions I outlined), but I didn't think to actually re-read it and apply it. I think the reason was that I'd imagined writing it to myself when the market is in the middle of a sharp decline and I'm struggling to know whether or not to stay in, rather than simply under performing the market with a high number of sells. Talk about blind spots!

The question then becomes, was it still right to change? Interestingly, considering this post now, having realised I didn't follow my process, it suddenly feels much more subjective and less logical.

Having not looked at my portfolio since writing, I've felt I've got a little valuable distance from it, and my biggest concern is not that it loses money in the short term, but that I keep having to sell shares. This has been an issue for several months now. I don't want to be a short term trader, I want to be a medium term investor (although I suppose trader by some definitions). I am concerned all these trades are going to eat up any profits, and that I'm going to and up with spinning my wheels but going backwards.

In terms of profits/loss, I started a year ago almost with £60k (not real money, this is me trying to make mistakes and learn before I put my real money on the line). I was up to about £70500, up around 17%. Since then, it fell to around 10-12% (all excluding dividends and costs). I suppose in the grand scheme of things, writing it out now, cold, it doesn't seem like much. It's still more than I'd get in a bank. I think part of the panic was a drive for 20% gain in the year. I got fixed on this (I think off of my Naked Trader reading) and so I think it made things about hitting a target, rather than getting the right process.

I think that change to pure QM, based on data from the historic SR data, shouldn't hurt. Maybe there wasn't anything with my buy rules and it was just bad luck? The high turnover of shares is, I think, a legitimate issue, so getting the right sell rules is something I need to work out. However, did I have enough data to suggest my old ones were wrong?.... Possibly not. I don't want to change anything again until the next review at the earliest, and even then, I think it would be good to try and do nothing if I can, as the start of my process of trying to do less and review less regularly. However, reading all these views on stop losses has been really interesting and useful.

...Maybe I need some rules for changing the rules? Perhaps "no changing the rules until they have run for at least 3 months", or "any rule changes have to be written down and left for at least one month before being implemented". Something to think about next review

Your research on SRs is really interesting! How did you go about gathering the data? Is there a tool, or was it a long evening switching between stockopedia and excel? I wonder whether the same holds true for momentum rank? I would have expected it to be more sensitive to short term decline, but I'm not sure if it is predictive as a lot of the rank is relative to current price. Same could be argued for Value too. Maybe Quality would be better long term? I don't have any data, just a theory.

I know it was a side point, but you mentioned the idea of price at the end of a holding period. I've so far had these as open ended holds - hold until they meet a sell rule. However, maybe having a set hold period (unless it still meets buy criteria at the end of the period) might help when reviewing rules, and again force my focus to the medium term a bit more.

As you can probably tell, you've given me a huge amount to think about. There have been loads of great responses already, but I really appreciate such a detailed and thought out response, thank you.

7th Sep '18

I'm glad you found it helpful. I wasn't sure if I was being too cheeky, but I know how easy it is to get too close to a problem.

I agree that moving from QVM to be more focused on QM should really hurt. From looking at the Stockranks history. since inception QVM90+ has achieved 18.3% annualised compared with 17.6% for QM90+. If you look at the broader 80-100 bands in fact QM comes out slightly ahead a 16.7% compared with 15.9% for QVM. So there is not really a lot to choose between then based on the long run data available so far.

Looking at the year to date so far QVM80-100 is up 3.6% with QM80-100 better at +6.35% . So QM has been doing better than QVM recently although that could just be a fluctuation, it's certainly true that both have been underperforming compared with recent years (but still outperforming the FTAS - down 3.81% on the year).

There is probably a risk of reading too much into that, generally the historic data shows that both have been good approaches to follow. (In passing though I would note that whilst QVM has not been particularly successful bu QM has been better, with a suggestion that maybe Value has been arbitraged away by more efficient data, so maybe QM is a bit more future proofed?)

Anyway I do think that the holding / selling rules are probably the right thing to focus on.

The quick view on SR movements that I gave was just based on a couple of snapshots that I had for other purposes, so it's effectively a single observation which I can't easily replicate. However as I also have the data for the individual Q,V & M measures I can a least offer a little more.

I had 130 stocks at the start of the year with QVM>=90 and this was the dataset I was referring to above. Of these 56 also had M>=90 and these collectively were down by 4.3% by my late March reference point (perversely worse than those with M<90 - up 0.7%).

Those that had seen the MR reduce below 80 had fallen 13% (compared with 2% loss for those whose MR had held up) . Those that had retained MR>=80 went on to recover 8% over the next 3 months, whereas those where the MR had fallen recovered by 12%.

I would stress again that this is obviously a single observation, at a particular time and on smallish data set; so I wouldn't want to draw any significant conclusions, but I can at least say this doesn't appear to offer any positive evidence for fallen momentum to be a good sell signal in the short term.

I have read though a number of people suggesting they do use sell signals like this, so it would be interesting to see how much 'real life' evidence as to whether this is working or not.

That's given me plenty of food for thought too!

Nick Ray

I tend to use Volatility rather than Value in my screens these days. It works in a similar way. A stock with high volatility is one which the market is finding it difficult to value and vice versa.

You do need to take into account that volatility as a measure of uncertainty of valuation needs to be compared against the overall price growth though. (e.g. 10% volatility on 100% growth is OK; 10% volatility on 10% growth is not OK.)

This screen shows one way to do it, with a limit on maximum volatility and a ratio between Price Change and Volatility. All the numbers in this screen can be adjusted to taste of course!

I find this avoids the "value trap" stocks which are cheap for a good reason. It is quite conservative in its tastes though. It is not going to leap on an early stage undiscovered gem for example.

Getting burned in August

Unlock the rest of this article with a 14 day trial

About the Author

Muggles

35 comments

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

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