What do trading algorithms do for the markets? Walter and Darren discuss various opinions and how it affects your trades.
Download (Duration: 15:53 / 14.5 MB)
In this episode:
01:38 – the hedge fund everyone loved
02:07 – the insane “mother of all losses”
03:26 – the dinosaur asteroid & the markets
04:12 – how calculating risk is comforting
06:29 – technology = the easy way out?
08:30 – the good things about algorithms
09:08 – scalpers & trading algos
10:25 – tried and tested approach
11:08 – the markets before & after algos
13:08 – reports on the effect of algos on the markets
13:37 – Doctors warn of the danger of trains
14:25 – should we worry?
Walter: I don’t know, I mean there’s a lot of good things about the liquidity, the fact that, you know, you’re able to get in on tighter spreads because there’s so many more trades. I saw a stat the other day that in one year, the amount of trades in the stock market, I think it was-
INTRODUCTION: Two traders, Darren and Walter, pull back the curtain on profitable trading systems, consistent money management, and profitable psychological triggers. Welcome to the two traders podcast.
Walter: Welcome back, it’s Walter and I’m here with Darren, and Darren today we’re going to talk about the trading algorithms. These trading algorithms usually make the news every time there’s a big move in the markets, and we’re just wondering Darren if we can get your opinion on what do trading algorithms do for the markets, have they changed the markets, do they make it more difficult on the little guy, what’s going on? and what are your opinions about these algorithms?
Darren: Yeah, I mean I’m not really an expert on algo-trading, Walter. But it’s like a few little stories that I sort of latched onto that make sense to me and the way that I trade. For me the big one is the Black-Scholes model. Which was Myron Scholes who won a Nobel Prize for his kind of work that a way you could eliminate risk in the markets by some sort of dynamic hedging. And basically it was like some intense mathematics, like some really clever stuff. And, then he went on to start this hedge fund and obviously everybody wanted to invest in it, and make a use of this model. And of course they made loads of money, absolutely loads of money, everything’s brilliant, and there’s no risk at all.
And then there was like an unexpected event, and they lost a load of money. But you know they redid their calculations and it’s okay let’s keep going, and it’ll all come out in the wash and it’ll be fine. And then there was like another unexpected event. And then, within I think it was like four months, the lost like fifty billion, or something insane. And they nearly brought the whole bloody financial system in the world crashing down. And, what’s important for me to take from this, is this fact that you cannot remove uncertainty in the market by trying to quantify it, and that’s what I believe in a hundred percent. And so, although you can build an algo that will be profitable, you can’t change the fact that the markets always can be uncertain. So I think really it’s a level playing field, basically the market for them is the same as us. It’s something uncertain and if anything really, we’ve probably got an advantage over that. Because it’s what we believe that’s important, and if we believe that we can remove the uncertainty and we can’t, then we’re going to become unstuck sometimes. But if we’re aware of that uncertainty, we can better manage our risk, and really that’s the only thing that we have control over. So that’s my kind of view on algos, what’s your take on it?
Walter: Just hearing you talk about it reminds me of the idea of asteroid hitting the earth. We can calculate the probability that we’re gonna be smashed by an asteroid, and go the way of the dinosaurs, right? But, that doesn’t really remove the threat of the asteroid slamming into the earth. In fact, if you pay attention to astronomy, you’ll hear these stories about “oh, here comes an asteroid, and it’s gonna make a near miss again and we didn’t even see it until a couple weeks ago”, or, these sorts of things. So, clearly there’s the risk out there, and the scientists are aware of the risk, but they still don’t necessarily see it coming. If that makes any sense. It just popped into my head when I heard you talk about this because I think it is kind of a parallel idea. This idea that if we can quantify risk so I have a 0.335% chance of risk of ruin for my account, so therefore I feel better about trading this system you know, or whatever with these given parameters. So it’s like a soft thing to lie down on when you can quantify risk, isn’t it?
Darren: Yeah, yeah, definitely. I think that ultimately, we all start out with this kind of idea that the goal is somehow, to remove the risk. And we kind of believe as well when we talk about algorithmic trading that it’s someway less risky than discretionary trading, but essentially we’re all trading the same market and it’s all controlled by humans, so really you’d have to go down this kind of efficient hypothesis thing as well. Where everything is already priced in. The news comes out, the price corrects instantly and there’s no kind of room in the market for slack.
I had a back trader talking about the algo part of their operation, and he was saying that they still need someone to go in there and make an assumption about whether the market’s going to be trending, or ranging. And then set the algo off doing it’s thing. And, I think it’s completely true, we’re all basically at the mercy of the market. If you want to be the sort of trader that says “okay, when A, B, and C happens then I have a seventy percent chance of winning and if I place my stop here, this is the ideal place and my exit will be there” then you want have it all pre-defined on what’s happened before, then maybe algos are going to affect you, because that uncertain movement in the market when you zoom in so close, and try and be so quantify [00:06:00] it to an Nth degree like that, then you’re going to keep coming unstuck. There’s ways we can use as discretionary traders to make it not an issue for us.
Walter: Yeah, exactly. I think there’s this sexy sort of, technology is sort of this thing that we look at as a way of improving our lives. And so, as traders, especially traders who are sitting down and really doing the hard time in front of the computer, they might see an algorithm or a robot as a way out, or an easier way. Sort of like you know we’ve got one of those vacuum robots, you just let the robot go, you turn it on and then you leave the house and you come back and your place is vacuumed. You know what I mean? So, and that’s kind of I think how people look towards technology is, technology is supposed to help us. But I think you bring up a couple of good points. Number one, and this is one I’ve always argued, is that when traders come to me and say “look, I’m gonna turn this into a robot, and that way I don’t have to worry about it”, my reaction is “no, you’re still gonna worry about it”. Even if you are trading in the A, you’re the one that turns it on and off, and you’re the one that babysits it, and it sounds like the same thing the banks are doing.
So, one, it isn’t totally completely like a robot that’s sitting off in the corner trading for you. Two, with these algos, I think that the thing that really it may affect, is more the velocity of the moves in the short term, right? So, if I’m a three minute chart scalper, maybe I will see a little bit more violent moves as algos start to filter into the market, and into my market in particular. Maybe. Now I don’t know if that’s necessarily a bad thing, it probably depends on the way that you know, the style of trading that I use. However, I would argue that, even if, ninety percent of the market is flooded with algos in terms of the higher time frames or the swings in the market and those sorts of moves, I don’t really see how that would make a difference in the way that I trade in and a lot of traders I know. You know what I mean? Like I don’t see how, how an algorithm, because [00:08:00] most of them are high frequency trading algos, or at least that’s how they’ve been characterized in the media.
So what that means is, the bulk of the moves that they influence are going to be those sort of micro moves, Right? And the shorter time frames. And it’s to, while it might give a little bit push to the daily range, or how fast it can accelerate in the range of the day, I don’t really see how it can change the moves that I trade. You know what I mean? And so, I don’t know. There’s a lot of good things about the liquidity, the fact that, you’re able to get in on tighter spreads because there’s so many more trades.
I saw a stat he other day that in one year, the amount of trades that in the stock market, I think this was in London, and this was back you know when algos were just starting to come into the markets. They said that the year prior, all of the trades done in the whole year were done in one month the following year. You know, because of these. So that’s gonna really improve liquidity, that’s probably going to pull down the spread on a lot of markets. So I think there are a lot of good things about this. The only real weak spot I would see is if I were a low time frame trader. If I was some sort of scalper, then I might be concerned, but again, I might also see that I can get a chunk of bigger moves perhaps. Whereas maybe before I had to wait a little bit longer, now that the algos are here maybe it pushes the market a little bit. You know it gets a little bit more slippery I guess is the one way to look at it. But I’m not sure about that. So, I don’t really see how algos are going to affect those guys who sit back and look at sort of the swings in the market. If that makes any sense?
Darren: Yeah, I agree with you. I think it’s, I mean a lot of people want to trade with a very tight stop, and you know, they want the high win rate. And I think if you’re trying to achieve that, on the lower time frames, then yeah, this high volume of short time frame trades are probably going to make that nearly [00:10:00] impossible. But I think traders who perhaps use the longer time frames to get into their positions have a bit more room to move. Or if they are using a tight stop loss, they’re at least letting their trades run so, although a lot of them are going to be taken out, the entry level, the few that win, are going to cover those losses. So I think you know, the sort of tried and tested approaches to trading still hold true. But, you know, people are attracted to the short time frame, tight stop, high win rate systems, and I think that perhaps algos just make it harder for them because if you watch some certain levels very closely, price always tends to sort of push through before it returns and these are places people want to put their tight stops and I imagine a lot of them are getting taken out and struggling because of that.
But you know when I look at the markets, they don’t seem to be acting any differently to charts from years ago, so I can’t see any visible change to the way markets move, really. And I think that’s because they’ve always been controlled by humans, and they still are. Really even though you say, even the algos, they’re based on human’s feelings and perceptions of the market. So, I don’t set it as being a problem or changing the market.
Walter: I mean that’s it, that’s the bingo right there, you hit it. We’re still humans, we’re still humans in the markets, and that’s kind of the deciding factor here. What’s making the market move are human decisions, so. And the other thing that I just want to point out is, the algos have been fingered as the dirty trader in the room ever since you know, 1987. As far as I know so, this has been going on for thirty years. So [00:12:00] people who are new to trading and worried about this, I think that it’s important to remember that this is not a new thing. These things have been in the markets for thirty years, at least, maybe longer. And so it’s not something that I would say is a concern, unless, unless you’re one of those people who’s bent on trading the lower time frames, and then it’s only a possibility that it might be affecting you. So I don’t really see this as a bad thing, I think that ultimately we’re going to see tighter spreads, which is essentially the equivalent of lower commissions for most traders in Forex. So, I don’t really see this as a bad thing at all, I just think it’s a- And maybe will push some traders away from, you know, lower term scalping to the higher time frames. And I think that’s a great thing too.
So, I don’t really see this as the big bad guy that a lot of people are at the moment, could be wrong. But this is something that I think isn’t really a concern. In fact, I think that if it does- You know there are some reports out there usually there are government reports and I usually don’t put a whole lot of value in what they have to say, but, basically we’ve heard things like it may be that the algos make it easier for the marks to go fast? right? Now, as a trader on the higher time frames like say trading the daily charts let’s say. If the markets go really fast, that’s not usually a bad thing, right? I mean, if it goes fast and hits your stop, okay, so be it. But if it goes fast the other way and goes towards your profit target I don’t really see that as a bad thing. So, I just think this is another case of people blaming technology. You know, when the trains came about- do you know the story about the trains Darren? And how the doctors were warning people not to ride trains?
Darren: I remember something briefly about it. Remind me, weren’t their heads were going to explode or something?
Walter: Yeah they were-
Darren: Travelling 30 mile an hour.
Walter: Yeah, so the doctors were concerned that, if yeah, if they went over twenty miles an hour that people would suffocate on the trains. Because you wouldn’t be able to breathe with the air moving so fast, and that sort of thing. So these are things that pop up and people are concerned about technology, and I think it’s a healthy concern, but I don’t necessarily think that it’s something that traders should be worried about. So, my general feeling on these algos- and remember algos are simply ideas put into code, you know, human ideas put into code so that they can be executed faster than the human can. So, there isn’t really much I think to be worried about here. So, that’s really my take on it Darren, do you have any final thoughts?
Darren: Yeah, I agree with what you say Walter. I think just we all believe that with certain technology that the more complex something is, we seem to think that that’s got greater ability to achieve it’s goal. And I think that the reality is, the more complex you make something, the easier it is for it to fall apart, it becomes more fragile the more complex you make it… that’s Nassim Taleb’s idea, its one of his ideas. It’s an idea that I think really rings true. I think just keep your trading simple, don’t try and complicate it. It doesn’t have to have some amazing mathematical formula, there is no “secret code” to break. I don’t think we need to worry about algos. Keep it simple and yeah, you should be alright.
Walter: Thanks Darren, and we’ll see you next time.
Darren: Okay, thanks Walter.