What I learned in 2012 (and what became more ingrained)

Published on Wednesday December 19th, 2012 at 07:23PM by TonyIommich

As the year comes to a close, I wanted to write up the things that I have learned as a developing trader this past year:

* Studying charts to gather technical ideas is very important, as is observing the psychological effects of implementing these ideas.

One of my biggest mistakes before was getting frustrated with charting analysis and believing that simply trading the market live was the best way to learn. Bullfeathers. It is critical to study charts very carefully, and while gathering them note such things as: what time of day is it, what day of the week was it, what news releases occured during that day, where was the prior open/high/low/close, how volatile overall is the market. This helps you get an idea of what is going on.

The more charts you study, the better, though you also have to keep careful tabs on the context of the market era you are examining. Obviously things were much different during the subprime mortgage crisis and the initial eruption of the Euro crisis, as compared to this day when market participation is much lower by comparison. Back in those high volatility days there were more setups, greater ranges, nearly every major currency pair could be traded intra-day, and you could still eek out short term trades during the late New York and Asian session. Today's low volume markets require tighter stops and tighter take profits, and you must be way more selective. The only intra-day tradable hours these days are from London to early New York, save for an afternoon New York fed announcement.

To come up with trading ideas search for technical themes that make sense. They can be incredibly simple classic things such as double tops, head and shoulders, flags and pennants, basic support/resistance, and candlestick patterns. Simple as they are, they do work. However, what is critical is the context you see them in, the confluences that increase the probabilities. That is what gives you an edge. Obviously if you simply traded every single double top or head and shoulders, you'd end up with randomness and that is very bad for your account.

One favorite thing I like to do is imagine a super-trader who can catch all turns in the market. I look at each one of the turning points and try to imagine what gave the trader the idea to catch at least 50% of that turn. I work backward to find out what possible elements of confluence existed to provide an edge. Granted, I don't expect by far to be able to predict everything, because there are so many different trading themes that it is impossible to specialize in all of them. I'm just looking for 1-2 good moves a day that are related to several themes I can understand well, because that is all you really need to be a profitable trader.  

Note this is not the same exact thing as 'backtesting'. Backtesting is when you test very specific rules, e.g. sell on the cross of moving averages X and Y, noting your profits and losses and tabulating them to determine if a system is profitable (and here, computer programming is a big time saver). I am talking here about simply studying technical patterns and coming up with themes to use in trading.

Combined with exhaustive study you need to get your screen time. This is critical because without it, you will not understand the psychological issues involved with implementing the technical ideas you noticed in your research. It is vital to understand how your thinking works, so when you look at the charts you will think about how your mind works along with what you observe in the markets.

* Narrate and tokenize concepts.

As you are watching the chart (either live or after hours) I'm a big believer in narrating, because this helps you understand what the other traders are thinking. You have to try to come up with logic based on common technical understandings, which is what helps develop a strategy. Think, for example, where all the traders are who are about to get stopped out, and what is likely to happen when they do. Think of when traders are taking profits, why are they taking profit here and not holding for the move. Think of those who will come in to buy their profits. Why might they be right or wrong?

As you narrate you will begin developing your own vocabulary which will help you understand market action better. For example, some of my favorite terms are "jamming through the roof" and "crashing through the floor", when the market is going through resistance or support respectively, and forcing people out of their positions.

* Always think of both sides of the story.

Don Miller gave this idea to me: you have to consider both sides of the trade always. Otherwise, if you only see one perspective you are not only going to be surprised when you lose, but you will fail to gain a strategic insight into how the others think.

* Risk reward is a concept you have to consider in a technical and volatility context.

Yes, we all know that if your loser is 1/4th of your winner you can be wrong 87% of the time and still break even. But that is not how markets tend to work, especially in the short term where your spread takes up a significant percent of your profits.

Technicals determine stop points and profit targets, as well as overall market volatility. In August, the market volatility was miniscule. Your profit targets are not likely to be reached, so you have to shorten them appropriately (as well as shorten your stops). This underlines the need to think critically when approaching the market, not live by simple mechanical rules.

* Some strategies work great on paper, but are not economically viable.

I found a good way to counter-trend trade strong trends, but the most I could get out of many moves was 4 pips, and after the spread and late execution the most I'd net is 2 pips. That made it very difficult to effectively implement without falling victim to repeated stopouts.

* Just because a strategy doesn't work, doesn't make it fadable.

Lots of times people think that merely doing the opposite of what doesn't work is a great idea. The truth is that many times if a strategy is not effective it simply has no predictive value, in other words it is equivalent to a coin toss. Just because you are losing money with it, doesn't mean you would be making money if you went the other way necessarily. You can test this for yourself. Granted, there are times when doing the opposite is the right idea, e.g. I once found out that my desire to fight a strong trend actually worked much better when I went along with it. D'oh!

* No economically rational trader will ever publish a truly successful trading strategy.

In the days of the algorithmic trading craze, any strategy that can be effectively programmed (and yes, computers have gotten very good) would be immediately copied by the numerous quantiative hedge funds out there and soon rendered useless. Those strategies that are published and are available for sale are those that have largely lost their effectiveness (think Richard Dennis's Turtle Strategy, which has even become fadable according to the 'Turtle Soup' idea).

Strategies that can't be effectively copied by computers are highly discretionary, requiring a combination of screen time and qualitative analysis to put to work. Hence, such strategies can't be used outside the box but require repeated practice to get right. The good news is that you can use such published strategies (assuming you get them for free or close to free) to good use in developing your own.

One book I like to this end is Bob Volman's Forex Price Action Scalping. This book focuses on the EUR/USD exclusively, which is what I spend 99% of my time on, and doesn't rely on indicators which I hate. The concepts Bob talks about address trend following and range trading on an intra-day basis. Bob focuses on 70 tick charts, which are pretty much equivalent to a 30 second chart during the liquid European and early New York sessions. While he gives you specific setups to work with, including trade management and psychological guidelines (very important), I find him more useful as an overall guide on how to approach intra-day price action trading as opposed to using his strategy out of the box as is (whatsmore, I find the $50 a month that is required for a 70 tick feed with 1 pip increments to be a total waste of capital that doesn't provide any additional edge).

* Day trading: one asset or multiple assets.

One of the different approaches to intra-day trading is if you decide to focus on one asset exclusively, such as my focus with the EUR/USD or other's focus with specific futures instruments such as the S&Ps, or to trade multiple instruments like stock day traders who pick out either sector ETFs or individual stocks that are 'in play' (are making new highs or lows, usually due to fundamentals).

The advantage of the single asset approach is that you get to know a specific instrument very, very well and understand how it moves technically and fundamentally. Your mind is free to focus on one thing without being spread out over multiple instruments, and arguably that allows you to get better at understanding just that one thing. The other advantage is convenience. If you just focus on the EUR/USD for instance, you have virtually no fixed costs (no data feeds) and you can keep all your capital in one account. If you decide to branch out into stocks and/or futures, you have to be concerned not only with data feeds but different levels of margin. For example you could not trade stocks, cash FX, and S&P futures out of one account.

The disadvantages of the single asset approach are that your trading opportunities may be more limited by comparison to a multiple asset approach, and you will have to specialize in multiple trading strategies versus just one or two in order to increase your trading opportunities (for example, be equally good at trend following as well as range plays).

If you want to trade intra-day and take the multiple asset approach, clearly there are greater opportunities in stocks as opposed to currencies. I've only found, especially as of late, that the EUR/USD has sufficient volatility for intra-day trading. Most other pairs have average ranges far below EUR/USD which make them useless for anything but long term trading, save for the occasional country specific news release.  The less traders participate in a market, the less liquid it is and the fewer solid trading opportunities exist (basically, we need people to be involved to trade off their moves). This explains the tremendous popularity of the S&P ES futures contract for futures traders.  

One of the biggest problems these days for traders is that markets have become increasingly correlated. Part of this is due to the traditional correlation that comes in times of global economic strife where the majority of people's concerns reflect macroeconomics as opposed to individual microeconomic dynamics that differ across instruments. Part of it is due to arbitrage activity by quant traders which is keeping the markets more in sync with each other (something that used to be the exclusive domain of discretionary day traders). Consequently, the traditional benefits of trading multiple assets are diluted when instruments travel in the same direction. Also, if liquidity dries up in one market, these days it is likely to dry up all across the board since money managers are thinking more in terms of 'risk on, risk off' rather than asset class A vs B vs C. That makes it hard to seek shelter from diversification by going to multiple asset classes both on a long term time frame as well as an intra-day basis. However, despite this fact, there are always individual events that affect specific asset classes and introduce liquidity and participation, often breaking correlation. These are where the 'in play' opportunities lie.

* The quants are not in control.

For a while I kept hearing all the complaints from frustrated day traders who insisted that the high frequency traders have taken over the market and the market simply is not tradable on a short term basis. The truth is there are still many traders who successfully make a living from day trading stocks and other instruments, they have learned to adapt to the new players in the market. It is also important to remember that as quants with HFT boxes enter the arena, their competition isn't simply human: they are also competing with their own kind. Hence, the crazy profits that HFT machines used to make are now smaller given the competition. Also, these HFT boxes will always leave behind patterns in the market, allowing discretionary traders an opportunity to jump in on their game.

Even if you're reading this and are jealous that the HFT guys are so powerful, don't be so happy for them. First of all, just to get the mathematical skill set required to become a competitive quant trader, you need to not only do the standard four years of college but also a PhD which entails an additional five years of having your head buried in the books with math formulas and C++ code. Once you have the knowledge, you have to invest in solid computer hardware and expensive data analytical software which require maintenance. For rapid-fire strategies you have to pay for very quick and secure internet connections, and have redundancy systems in place. You also need access to a brokerage that can execute, although in many cases even that won't do: you have to access the exchange directly which is very costly. To top it off, you need people to help you handle all this work, including traders with experience that can help you develop trading ideas.

While the pay of a quant can be in the neighborhood of $200,000 and more, due to the large economies of scale required for their strategies to be successful they are always pretty much tied to an institution such as a bank or a hedge fund. The competition involved requires that you work your tail off, many quants are busy for 12 hours a day and even work on weekends. Its hardly the glamorous thing its made out to be.

  • Go to Pipstradamus's profile

    On Monday December 24th, 2012 at 12:05PM by Pipstradamus -

    Wow. Awesome lessons. Tnanks for taking the time to write this Tony.

  • Go to TonyIommich's profile

    On Monday December 24th, 2012 at 04:57PM by TonyIommich -

    My pleasure Pipstradamus! Here's to a great trading year!

  • Go to SpacePip's profile

    On Wednesday December 26th, 2012 at 08:46PM by SpacePip -



    Awesome read!

  • Go to buttercake's profile

    On Friday December 28th, 2012 at 03:17AM by buttercake -

    Wow, Tony is back!

Leave A Comment

Please sign in to comment.

  1. Login
  2. Sign Up

Our Sponsors