I have explored many ways of keeping a trading journal over the last few years, including a straight-forward diary, loose-leaf by company, web pages, Google docs, and so on. I usually return to the diary format, but it is not ideal if I want to search for companies. My latest experiment into keeping a journal is to use a text file, but keep it in a GNU recutils format . Think of it as a flexible (free-form, but with some structure) plain-text database format. Linux users will be able to install it easily from their repos. I am actually using it on cygwin on Windows 7, although it’s not straightforward to set up from scratch.
The diary is in the form of a single file, and the great news is that I can use recutils to query it as a database, and look for my company notes. Recutils is quite an obscure tool, I don’t really hear of anyone waving the banner for it, but it is an official GNU project. Given that it’s a free-form textual database, I would have thought it would have been more popular. Alas, no.
At the top of the file, I have the table schema, which is simple enough:
%type: Date date
%auto: Date Id
I have a tiny bash script where I can “add-trade $SYM”, and it will add a record for me, and bring up an editor (vim), putting the cursor in the approprate place. Very convenient for me, as the record id, EPIC code and date is all inserted automatically for me. All I have to do is add my notes.
I hasten to add that my solution works for me, and YMMV (Your Mileage May Vary) – which will basically mean that you will likely find it a terrible way of approaching things unless you have a UNIX/programmers mindset. Not to mention, it will be a solution that will work best if you’re on a UNIX-like system, although OS X may have to compile the code. I wouldn’t know, I haven’t used OS X for a number of years now.
Of more importance than the technology is what I actually write down. I am refining my approach, but my latest thinking is to try to explain the critical aspects of my decisions in ways that can be seen to be right or wrong in hindsight. The point of a trading diary is to improve your decision-making, and hence returns, rather than being able to provide a nostalgic trip down memory lane.
I think, too often, I have seen articles which don’t really help in the decision-making. I remember reading an article on the tobacco industry, which explained something along the lines that IMT (Imperial Tobacco) catered to the low end of the market, whilst BATS (British American Tobacco) were strong in emerging markets, where there was potentially a lot of growth, but that governments may be trying to curtail its expansion. It then went on about packaging and tax revenues and how that might affect company performance. In terms of giving an overview of the issues surrounding the tobacco industry, it was good. But the problem is, in the end, the article just kind of “shoots the breeze”; something something might happen, or it might not, or something different might happen. Fine, but ultimately fairly useless, because it doesn’t give me any points around which to make a decision. “If” this happens, “then” that will happen. If not, then it wont. OK, so which is it?
So, in my trading journal, I’m trying to boil things down to their essences. If I bought a share, presumably I liked something about it. What were those somethings? I also, presumably, disliked some aspect about the shares, but chose to ignore those aspects. What were they? With the benefit of hindsight, I will find out that some of my reasoning was right, some of my reasoning was wrong, and some of my reasoning was irrelevant. Ideally, I would hope to find that what I liked was both right and relevant, and what I disliked was irrelevant. I then have a record to judge what I thought were the critical aspects of the time against the results I achieved. I can then build up a better picture of what works, what is counter-productive, and what can safely be ignored. And that, to me, is the point of keeping a trading diary.
Enough talk, how about I actually give you a diary entry:
Date: 17 Mar 2014 08:31:20
Trade: buy MMM for £NNN @ 320p
+ * canslim
+ * momentum 98
+ * PV50 8.53%
+ * 50dma > 200dma
+ * positive trading statement on 05-Feb-2014, albeit v brief
+ * heavy director buying in last year
+ * pe 19.4
+ * shares down 5.9% on trading update (288p to 271p)
So you can see that I was interested in CANSLIM stocks with good momentum, but were trading close to a support level (that’s what the PV50 is about). The flies in the ointment were that it had a high PE (not untypical of a CANSLIM stock), and that the shares were down on the trading update – something that I generally don’t like to see; but which were facts that I chose to ignore.
Were the good points sufficient to send the shares higher, or should I have taken the poor market reaction to the trading results as a warning sign? I at least have a basis to make an assessment in the future.
It would be interesting to see what’s in others’ trading journals.