Hi guys,
Thanks for the encouragement! I guess now that I've whizzed through the last 20 years and I'm gonna get into the nitty gritty of what I'm up to, it'll be a bit less interesting to follow.
I love spreadsheets
If you'd have asked me a month ago, I'd have told you that I was only really interested in buying and selling shares, I'd heard of spread betting, but I thought it was probably a scam, I understood the concept of going short on something, but it seemed an added complication. My vision of trading was one of buying some shares, holding them long enough to actually receive the certificate (there is something quite nice about that), and then selling them once they've gone up a reasonable amount. That's what I was aiming to learn how to do properly. I figured if I learned how to do it properly, with stats and analysis and all that, I'd be able to trade calmly with confidence.
I had thought about other instruments in the past (only I didn't know they were called instruments). As I've mentioned before, I've got a natural attraction to the forex market - it kinda sings to me whenever I'm in the proximity of foreign currencies for practical, everyday reasons but I was under the impression that you needed to be either very rich or a professional trader to get into that. Gold and bonds and wholesale gas contracts I understood, but didn't think they were for little people like me. As for lean hogs or orange juice futures - I wasn't even sure they really existed, I mean why pork bellies? Isn't there better meat on the leg?
I'm telling you all this, 'fessing up to my quite recent naivety to explain why I've build the tool that I've built. It's based on actually buying - with hard cash - stocks.
I finished reading Pozzyp's journal last week - a great inspiration, both to get writing and to start testing out some of the theories and methods I've been reading about all over T2W. I was itching to build a spreadsheet. The only problem was that all of my costings and data were aimed towards swing trading stocks through a broker. Oh well, I can still use some of the concepts that I've learned about to pull together the bones of a model, which I can later build on to reflect more accurately what I've been learning.
I imagine that to some readers it may seem like the hard way to go about learning to trade, starting by building spreadsheets, especially while I've still only scratched the surface of technical analysis and how to use it. But by trade I'm an engineer, I like to know how things work, and since being a little kid I've always pulled stuff apart and then tried to put it back together again. For me, it's the best way to get into something.
History tells me I'm going to build a monster spreadsheet, so complex and intricate and full of idiosyncrasies that nobody can follow the formulae, and the bugger keeps crashing. And yet it will start from something so small and simple.
I've read in several places that many experienced traders minimise their risk exposure to 2% of their account. 2% sounds like an awfully low figure but I love awfully low risk figures! I decided I'd build a spreadsheet to calculate where to put stops based on a 2% risk. As I'm not sure of how much cash I'm going to put into my trading account, or how big a stake I'll put into any one trade yet, I left those as variables. 2% seeming quite low, I built in my % risk as a variable too. I tested it, fiddled around with it for a bit and got it working. I banged in some numbers and got a feel for what a 2% risk stop looks like on shares, it's obvious when you watch the numbers dance - the smaller the stake that you place, the wider the gap you can have with your stop for any fixed amount of capital in your account. Okaaaa-ay, I just learned something that I didn't know before.
What next? going short, that's what. I was much more comfortable with the concept after reading Pozzyp's journal. The maths isn't quite so intuitive, but what the hell, I literally am a qualified rocket scientist, surely I can master the sums behind going short! (Rocket science really isn't that that difficult, not like trying to learn Spanish or make shortbread, both of which have eluded me.) Cracked it eventually, I had to cross-reference a few sums to ensure that the answer would come out the same no matter which way around I calculated it.
Now what? Well, I was kinda eager to put it to the test with some real data, but that needed some thinking about. A few people have mentioned the theory that if you cut your losses quickly and let your profits run on, you'll come out an eventual winner no matter how randomly you chose your trade. I'm interested in testing that theory, because if it's true, it'll form a fine bedrock to a trading strategy. Still mostly thinking about stocks, I picked 10 company names that I'd not heard of from FTSE 250 and plotted their end of day bid, ask and volume. To truly test the random theory I should have modelled 5 going long and 5 going short, but that seemed a bit pedantic and not much fun. I waited until the next day and got two more points of data for each, a midday update and another end of day update. I'll let the market lead me - if three points of data show an upward trend, I'll go long, if three points of data show a downward trend I'll go short. It's still a very random way to pick my trades, and allowed for the fact that a few of them were range bound within those three points.
How to choose the value of my stake? I went back to my spreadsheet. The value that I'd need to invest was linked to the cost of the trade. Some early research that I'd done suggested a fairly average-to-worst-case cost of transaction would £10, so to buy and sell would be £20. I would need a large enough stake that a favourable price movement would cover the £20 cost per trade, without requiring an enormous step change in share price. I fiddled around with some formulae and build another set of sums. As expected, the higher the stake, the less the share price has to rise / fall to breakeven of the cost of trade.
Ahhhh... The higher the stake the less the price has to rise to cover the cost, but the higher the risk becomes, so the tighter the stop. Hmmm, this is gonna take some experimenting with. Eventually I came to the conclusion that £500 per stake was about the best balance - although the stops look scary close and the prices have got to move a long way to cover off the costs.
A week on and I've been stopped out twice and everything else is still in its first running, time will tell whether the profits will cover the losses or not. I suspect not, my costs are too high and my risk % too small for this kind of trading.
In the meantime, there's so much more that I want to model, and now that I've got documented decisions on where I went long / short or left alone I can replay the same experiment with different variables. I'm particularly interested in what would happen if I changed the cost base from broker trading to spread betting. I'd have opted for this straight away, but I don't understand the SB cost base enough to model it yet and I was too eager to wait for more research. Using typical SB spreads, does the stake come down, the stops move wider and the ability to breakeven get more likely - or does it go in the opposite direction, the leverage on minimum bet sizes pushing the risk up and the cost of failure? I don't know yet. I'm about to go back into my spreadsheets, so if I go quiet for a few days, you know where I am, I'm in the land of little grey squares!
Back soon,
Sal
