Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
I am currently reading Kevin J. Davey's book on algorithmic trading and so far it has been very inspiring and motivating.
Kevin mentions a couple of times that trading edges tend to dissapear. I am trying to understand why. This is a very general question so I am not looking for a system trading based answer per se.
To me, the main reason why an edge may dissapear, is because of changing market conditions. But material changes don't seem to be very frequent. For example, I don't think that events such as the 2007-2009 financial crisis will pop up again in the very near future. To me this reason by itself is too long term to take into consideration for everyday trading.
I believe that I am missing something and that more subtle and more frequent changes may be enough to invalidate a given approach.
Therefore my question is: what events tend to alter a trading edge?
Others could be catching on to the edge you found, and trading it away.
Maybe the edge took advantage of a big market player who decides to leave the market (I've heard this from a few people)
New players enter, and maybe their net trading impact wipes out your edge
Maybe it was never an edge at all, but just something that looked like an edge, but was really just random luck (a million monkeys at a million typewriters might eventually write a Shakespeare sonnet, but that doesn't make the monkey a good writer)
The market might change. The Euro systems I had in the book did pretty well until mid 2014, when volatility started dropping in the Euro. That hurt both systems, and neither has really recovered (even though volatility has returned)
**************************
In my mind, it doesn't really matter why a trading system could falter. What is important is that you plan for it, and if/when it happens, you turn the system off before it wipes out your account...
I think the correct term for this is planned obsolescence... although I hope no system ever fails!!!
Markets never change. It goes up. It goes down. It goes sideways. You have small range. You have large range. High volatility. Low volatility. Just learn to trade in each of these instances. Forget about the lagging indicators and just stick to volume and order flow.
Trading: Emini ES, Emini RTY (TF), Crude CL, Eurex DAX, Euronext CAC40, EuroFX 6E, and Hang Seng HSI
Posts: 47 since Mar 2011
Thanks Given: 125
Thanks Received: 61
Markets changes often every year for many different reasons such as changes in key institutional players and assets, changes in volatility, changes in global economics (e.g currency wars, interest rates, debt, regular schedule economic reports and so on), Geo-politics (e.g. government elections), changes in trading rules, illegal trading activity...all of which causes changes in the psychology of the markets.
Don't be mistaken. Changes in the markets can be bad and good. Thus, new edges can form just as fast as those that disappear. Just as important, an edge can become stronger or weaker but still remain an edge.
The key then becomes the trader. A trader being able to recognize the change and then adapt to it.
If everyone started to trade with volume and order flow (or any other method for that matter), any edge that may exist in such a method would eventually be traded away.
HFT is a good example. High frequency firms have taken away a lot of edge away from the retail scalping crowd. Their actions have changed the trading game for many traders.
Maybe a particular edge will never go away, and that is awesome if true. But most edges seem to dissipate over time, so the key in my mind is to be ready for whatever changes are necessary.
Thank you for your answers guys. I didn't expect the author of the book to reply to my questions! Saw that you also got a thread specifically designed for that Kevin - thank you, I will post there in the future.
Now onto the discussion.
Interesting question.
To me an edge is any entry, exit, position sizing, money management technique that would provide better results than a random approach AND generate a net profit in the long run.
Most of my trading practice so far has been in discretionary trading, simple things like breakouts and trend following. These approaches are better than a "monkey do, monkey see" approach in an intraday environment. The evidence for this is the small statistical sample deriving from my experience: basically, I think I got better by reading books and observing price action. I am pretty sure that this is what Kevin's "monkey see, monkey do" test is supposed to be reveal: does his set of rules perform better than a random entry/exit? I didn't have the chance of designing a statistical test around my discretionary techniques, but I believe, from what I saw, that they are indeed better than a random approach.
The approaches that I mentioned - breaksouts, trend following - are designed around principles that are inherent to financial markets and I don't see them dissapearing any time soon (like the fact that it's better to not trade against a trend while it's in force). This, I think, this echoes mrmuggins post in this thread:
However, the specifics of a given strategy - such as the relation of a pullback to a trend in order to determine whether the trend will resume - vary from market to market and I would think that these might indeed evolve quickly. The performance of your set of rules will therefore change if you don't adapt them. So I guess that this is the "edge" that most traders refer to: the performance of your specific set of rules, not the general principle underlying your technique.
Now onto my initial question which I have refined in light of the abovementioned conclusions: what are the specific reasons that might alter the performance of your set of rules? (now that I ask the question in this fashion it seems sort of useless but let's keep going!).
Wrbtrader's and Kevin's posts both cite similar reasons. Those are the answers I was looking for (thanks). I agree with most of them, but there's one that still bugs me a little bit.
Let's take a hypothetical situation where I developped a hugely popular trading strategy for a given instrument. This would mean that a lot of people enter and exit trades on the exact same moment by following my rules. On entry, there would be a lot of follow through. This would mean that the more people trade my system, the better off I am.
You seem to say that, on the contrary, a given system will not stop yielding results if a lot of people start trading it. I don't really understand why.
EDIT: just saw that you mentioned this issue on the previous post.
Let's say you found out that a 50 bar breakout was a terrific way to enter the market, and a 20 bar breakout (in the opposite direction to your position) was a great time to exit. So, you start trading it, and things are going well.
Word gets out on what you are doing. Maybe you tell people. Or maybe some really smart people somehow reverse engineers what you are doing. Maybe others discover your approach through testing.
So, in the beginning, people will probably trade your exact method. So, let's just say when you entered an order before others caught on, and your small 5 or 10 lot was easily absorbed by the market. But now, maybe 1000 buy orders are all hitting the market at the same time. What happens? There aren't 1000 sellers at that price, so something has got to give. The price moves up to fill those orders, giving most people a slightly worse fill. Same thing on the exit. So, your system may still work, but is not as effective.
Then, over time, people figure out ways to "jump in front" of your signals. Maybe it is a 48 bar breakout they enter on, or something where you now board the train after it left the station. That will degrade your system too. Or maybe it is someone with a colocated server, who just jumps in front of your order by microseconds. Or maybe someone burns the code on a board, so its signals are generated a microsecond before yours are.
So, what will you do? You'll probably change your approach in response. You'll adapt, maybe slightly, maybe dramatically.
In that respect, the market never changes. Each participant in the market is always trying to beat the other participants, whether they intend to or even realize it. That has always been the game.
