Trader Talk

[RobbieG]

I get a lot of communication (especially lately for obvious reasons) about trading and trading strategies. For those who don't know, I own a private equity hedge fund and trading software development company. I'm not going to go into any real nitty gritty stuff, but many questions can best be answered by yourselves in the same way I might if you understood how I view markets and trading and the industry in general.

The ability to predict market direction is the biggest Wall Street lie that to this day is still going on. People misconstrue success with knowledge. Most of the time fund managers are right they just got lucky, but yet their career validity is based on making people believe that they won those trades because they are smart. What a joke. Any market always does what it wants to at all times. Don't let anyone EVER tell you differently. They are either ignorant or not telling the truth. Period. All the big insitutional dogs know this and it their business to make sure you don't know that. Otherwise, why would you pay them to do something you can do yourself? All markets are perfect and imperfect all at once. As such, no formula or intellect can consistently be successful other that what I call controlled omniprescence. You have to be available if the market you trade should decide to move, and apply a trade management strategy that will maximize getting a large percentage of what it gains. Conversely, if you are wrong, which I'm proud to say I am at least 50% of the time, you have to employ a strategy that limits the amount you lose which is consistent and precisely quantifiable. In sum, FOOLS try to be right. Let the market do whatever it wants to and just be ready for whatever that is and know exactly how you will react when it does. If you consistently win more than you lose and you are right 50% of the time you will make money. Period. Then lever up (but not too much, Bear Stearns...) and you can make more with quantified risk in the long run. However, you don't always win in the short run. Nobody does. The reasons for this are best explained with gambling theory.

Most investors aren't comfortable with the idea that trading is gambling. Like it is a dirty word or something. But yet I can honestly say that I'm sure I wouldn't be able to make consistent money as a trader if I didn't apply gambling principles to the game. As a card counter, there isn't a casino in Vegas that will let me play that way. The reason is simple. In the long run I'm going to beat them. My trading is the same thing except that my casino lets me play. Mathmatical expectation is a simple concept, and it is flawless if properly applied. To use this gambling analogy to what I said about "being right" in the previous paragraph, trying to take a trade only when you think you will win is the equivalent of wanting to play blackjack with a 48 card deck, or only hitting 16's if both cards are red. Huh? You want to play basic strategy which is the trading equivalent of following simple money management and risk quantification rules on each hand (trade). The expectation to a card counter comes from betting more when they have the nuts. Very simple. In trading it is the same. Let the market tell you when the conditions are favorable to bet more, not before the trade which is just prediction, but after the trade is on. That way you are not predicting, but instead you are interpereting. Big difference. It is amazing to me how many traders don't know the difference between those two words.

Once you have devised a consistent and repeatable method to do all this, discipline is the key. You have to do the same thing every time and not let emotion get in the way. Look, to return to the blackjack scenario, you can be playing and while the deck is extremely favorable to you the dealer still pulls 10 blackjacks in a row and busts you for the day. It happens and it sucks, but trust me, you will come back tomorrow and bust them almost every time that happens. I have seen it many times. The math works both ways. You have exactly a 50% chance of pulling any two cards which are more favorable than what the dealer has on every hand regardless of what the count is. But by using a tiered betting strategy that takes into account how many tens are out and the order to act you change the expectation of returns while keeping the expectation of outcome the same. Only the most disciplined guys can wait it out and you have to have faith in what you are doing. In other words, don't lose the count! In trading it is the same and the "count" is whatever you have decided are good ways to interperet (not predict) what is happening (not what will happen) You never know when the market you trade is going to wander off course and mess with your math in the short run but you must not allow it to make you alter what you do or you are dead. In my world we call those black swans and they are always there but you can't let them scare you. But if you are always in and bet like a card counter you can make money in any market you trade.

I'm oversimplifying obviously for illustrative reasons and you can't just roll out of bed and implement this stuff. I'm not soliciting either BTW. I don't cross those boundaries. I wouldn't take your money if you had millions and you put a gun to my head. But I know a lot of people here are trusting their money to managers who may not know what they are doing. It makes me feel good to help others and so I try and take something I know more than just a little about and educate friends. I do this in the hopes that you will educate yourselves enough to keep from going broke in these crazy markets.

If you take nothing out of this but this next statement, you will have come almost as far as you need to. And that is at the very least invest/trade in instruments in both directions. That is to say you must be abe to bet on the market going down as well as going up. These days that is easier than ever for individuals as many of the most popular indices are represented by ETF's (exchange traded funds) that are also available as reversed instruments whereby in buying a "short" version of said fund you are betting that it is going down instead of up. Good luck and good trading...

[Toadtorrent]

Interesting read. Thanks Robbie!!

[RobbieG]

Thanks. You are welcome.

Interesting read. Thanks Robbie!!

[Chronus]

Very very interesting... where can I learn more? And I mean that...

[cornerstone]

Interesting insight Robbie.

[capice]

tnx for sharing RobbieG, great insight view.

[RobbieG]

Thanks guys. I should also probably disclaim that professional trading (gambling) is not for the faint of heart and requires skill and dedication. I don't mean to oversimplify the process, but I can't stand to see people needlessly lose money at the hands of managers who would have you believe that they can pick stocks, time the market, or both.

This brings up an addition to the article that I need to address. Stock pickers love to cite Warren Buffet as evidence that stock picking works. That makes me laugh. Mr. Buffet is the greatest investor ever, but not the greatest trader - big difference. The most important thing about him to remember is that when he invests in companies he always buys a MAJOR stake in them. That changes the rules. With that comes juice. There is a big difference between picking a stock so to speak that will do well and buying a hundred shares and buying 10% of the deal. The rules change. Plus, everything he does becomes a self fullfilling prophecy because he did it. The fact that Berkshire Hathaway grabs an interest in something creates a buzz that makes everyone else want to own it too. If Buffet bought it, I want it too - that kind of thing. Again, the important distinction is the difference in outcome when you are actually buying a company as opposed to buying the stock in a company. BIG difference. Buffet is an investor that buys companies, not a trader that buys and sells stock. I just get annoyed when people mix apples and oranges using him as an example, and then say fundamental trading is great because Buffet does it. Plus, I should know a thing or two about him. One of my partners is a champion bridge player and he has been Buffet's math partner in many tournaments. The Oracle of Omaha can play some bridge I'll tell ya', and he is quite a gambler too I might add. But he doesn't gamble in the market. He bets on sure things and he makes them sure things by controlling the whole company. And that my friends ain't trading. The moral of the story is if you are in investor, don't invest like a trader, and if you are a trader, don't trade like an investor. Buffet is neither really. He is an owner who is very good at controlling what he owns through strategic trading and effective management. Don't get me wrong, Warren is an incredible genius, but even if you had his brains, it takes a tremendous amount of capital to accomplish what he does. Billions, not millions...

OK, I'll get off my soap box now - I promise...

[P4GTR]

Thanks Rob, good info.

[fat.tail.event]

I agree with many of the high level themes being communicated here...have a plan and stick to it, protect yourself when you buy a position (stops or hedges). I think it would have been helpful to also include position sizing insight, meaning never over commit capital to one idea/trade and avoid concentration in correlated assets.

But there is one theme throughout that leans towards quantification of odds in global markets, using the casino and card counting examples, that needs more reinforcement. Robbie mentions randomness and black swans, these are important...I will get to this in a second. Our team spends hours a week examining "fat tail events" (low probability, but highly devastating) with our traders/portfolio managers. MY WORD OF CAUTION: If you run a model or system or plan etc., sit down every single day and question its validity and ability to handle major shocks. Here's the problem... When you step up to a table to play a casino game you know your odds. If astute enough, you can chart the probability of outcomes at some tables over time. Put simply, the table and deck you choose is a contained system that is not susceptible to outside influences. Global markets are the exact opposite. There are infinite variables for which to account and no model/system will ever be able to.

A lot of Nobel prize winning professionals even fail to take notice of this. They develop models to explain the world using "bell curve" scenarios; it allows them to quantify odds of something happening. These models work most of the time and their operators eventually become married to those ideas after seeing repeated successes in predicting/explaining events. But when events occur outside of 'normal distribution' odds (the one in 10 million event) they become momentarily confused before saying...this cant happen twice, or three times in such a short span of time...so they risk more and more capital (remember they are married to the system). You just read the heart and start of almost every spectacular hedge fund blow up.

Folks, this is why these "once in a hundred years" financial crises occur every 5-7 years now. The "once in a hundred years" part comes from some economist's academic paper that uses a flawed model like the above and the world event at hand doesnt quite fit into the research. Its like sitting down at the blackjack table and getting 6 aces dealt to you in a row. Impossible in a contained environment, but we dont live in one.

Someone asked where they can learn more about trading. Start with a book like Trading for a Living by Dr. Alexander Elder. Its not about making a business out of it, but rather a comprehensive introduction into every area of money management. Anyone with an IRA, 401K should read a book like this at some point. The black swan reference comes from Nassim Taleb. Read his books Fooled by Randomness or Black Swan. I guarantee you that you will look at the world and so called experts a lot differently.

[b16a2]

Some really good information here. Trading has always been somethin I'd love to get into afte graduating, and this has certianly wetted my apetite! Thank you Robbie and FTE for taking the time to post some very informative posts!

[RobbieG]

Ah yes, Fat Tails. I haven't heard that talked about in some time. Bravo. You are 100% correct in your comments, but please understand that the point of the article was to oversimplfy really to answer general questions which are routinely asked to me. But I do appreciate your additions to the thread. Fortunately, we also don't encounter much fat tail risk simply because we never hold any positions overnight. Anyway, it was really just a general explanation of some basic gambling principles and not meant to be a full explanation or construed as hard trading advice. That said, let me add to the article by explaining how some of your concepts depend on the trading timeframe and how quantified risk is related to it.

My alpha fund is setting records each month because the current volatilty plays into the model by increasing the size of winning trades while holding the average losses (risk) constant. In fact, the fund returned over 100% on equity in the first 10 trading days of October with a risk model of 1% per trade, and trading only one market which is the S&P Emini futures contract (ES). The fat tail events hardly ever come into play with (any) intraday models such as ours mostly because reaction times are greatly diminished. This is because news events are on a clock and very predictable during the trading day. We never have postions on at news times, nor do we create new ones around those times. Our average time in trades is also so small due to the short term TF's we take triggers from. We are in a trade on average for less than 15 minutes. It really takes a black swan on top of a black swan to play out to de-rail our intraday expectation as a result. The likelyhood we would be in a trade on devastating news overlap and not be able to exit is small. Getting wacked and blown out on both the stop and the hedge is unlikely, but even if we did in that one trade the recovery would be quite quick and robust.

The price for this inherent safety we have is limited position sizing. That is why more large funds don't (read can't) do what I do. And yes, position sizing relative to triggers is very important and we are no different than most funds in that it is a big part of the success of the strategies. you just can't trade a hundred million in equity with my model. You couldn't get your postions and hedges on and off fast enough. All told, my fund plus our institutional clients who white label my software have only $30M employed in multiple markets. We cap the investor returns with no fees, so we don't need much of a surplus to earn for oursleves fairly significantly with equity which is quite tiny in institutional terms. My own fund is capped at $7M equity and trades a max size of 700 contracts per trade 3-8 times in an average day.

I could (and might) write a book on the myth of asset diversification as a risk management tool. It is the other giant Wall Street lie aimed at getting people to buy more stuff IMO. Nearly all assets, including ones that are inversely coorolated are coorolated. Coorolation is coorolation and that is why I hate the word. Just because instruments move in the opposite direction doesn't mean they are uncoorolated. I always found it funny for instance that people would use say, the long bond as a hedge for an equity index trade. A better example of an uncoorolated pair might be say, sugar and swaps as opposed to equity and bonds as a simplifed example. Sure equities and bonds are inversely coorolated, but other than arbitrage really, where is the gap in the hedge? Anyway, I believe in employing what I call directional hedging strategies which involve hedging with non-coorolated strategies in the same instrument and direction, as opposed to hedging using these supposed non-coorlated markets in the opposite direction.

In sum, we have many models but they all employ the same basic principals which is really nothing more than we don't ever assume overnight risk. The black swans are always overhead, but you can't shoot them if you are sleeping when London opens. At 9:30 AM we are bright eyed and bushytailed and the shotgun is loaded. Case in point: A swan wacked us last week for 110 ticks in one day actually. Ouch. nearly 14% drawn down in a day. I shot the bastard and we went on to make over 200 ticks over the next two days. Those big numbers have nothing to do with me. It is just the volatilty. Like I said before, I am no genius, but the fact is a lot of great traders are getting killed right now because the overnight volatility risk is nearly untradeable. The pain threshold for stops and hedges is just too tough to fade. Meanwhile, our average loss intraday is around 7 ticks a trade. I could use 70 overnight and I still might not like it...

In conclusion, I am well aware that my numbers are silly big right now but so are the intraday ranges so it is expected. Of course the returns will mirror that when your system wins more than it loses. In a more typical month (like last summer for example) our average win might be something on the order of 10 ticks a trade and the loss will run 8 ticks or so. In October 08' month to date, the average win is 26 ticks and the average loss is 13. That explains why we are making so much unlevered money this month. The market dictates it, not me. I'm just well set up intraday to maximize pocketing what the market gives when it does and minimizing what it takes away when it does. As the numbers get bigger so will my profits. The bad news is that it will end and we will go back to earning a third as much when it does.

I hope that clarifies a few things and I appreciate the contributions. Keep them coming!

I agree with many of the high level themes being communicated here...have a plan and stick to it, protect yourself when you buy a position (stops or hedges). I think it would have been helpful to also include position sizing insight, meaning never over commit capital to one idea/trade and avoid concentration in correlated assets.

But there is one theme throughout that leans towards quantification of odds in global markets, using the casino and card counting examples, that needs more reinforcement. Robbie mentions randomness and black swans, these are important...I will get to this in a second. Our team spends hours a week examining "fat tail events" (low probability, but highly devastating) with our traders/portfolio managers. MY WORD OF CAUTION: If you run a model or system or plan etc., sit down every single day and question its validity and ability to handle major shocks. Here's the problem... When you step up to a table to play a casino game you know your odds. If astute enough, you can chart the probability of outcomes at some tables over time. Put simply, the table and deck you choose is a contained system that is not susceptible to outside influences. Global markets are the exact opposite. There are infinite variables for which to account and no model/system will ever be able to.

A lot of Nobel prize winning professionals even fail to take notice of this. They develop models to explain the world using "bell curve" scenarios; it allows them to quantify odds of something happening. These models work most of the time and their operators eventually become married to those ideas after seeing repeated successes in predicting/explaining events. But when events occur outside of 'normal distribution' odds (the one in 10 million event) they become momentarily confused before saying...this cant happen twice, or three times in such a short span of time...so they risk more and more capital (remember they are married to the system). You just read the heart and start of almost every spectacular hedge fund blow up.

Folks, this is why these "once in a hundred years" financial crises occur every 5-7 years now. The "once in a hundred years" part comes from some economist's academic paper that uses a flawed model like the above and the world event at hand doesnt quite fit into the research. Its like sitting down at the blackjack table and getting 6 aces dealt to you in a row. Impossible in a contained environment, but we dont live in one.

Someone asked where they can learn more about trading. Start with a book like Trading for a Living by Dr. Alexander Elder. Its not about making a business out of it, but rather a comprehensive introduction into every area of money management. Anyone with an IRA, 401K should read a book like this at some point. The black swan reference comes from Nassim Taleb. Read his books Fooled by Randomness or Black Swan. I guarantee you that you will look at the world and so called experts a lot differently.

[chronoluvvv]

sticky this thread already !

[RobbieG]

Yes, random numbers are NOT random. Interestingly, we actually PROVED mathmatically that in a craps strategy we were working on for giggles and profit. This will blow your mind. Dice throws are totally random right? Wrong. In craps, we proved that a new shooter who has made it to throwing a point roll is just as likely to throw two passes as he is to throw twelve. Could you make money knowing those odds? We did (and do). Meanwhile, a shooter who doesn't get enough consecutive passes to make a point roll is more likely to crap out than he is to make a point. How can that be? Random numbers are not random, they run in streaks. For some reason success seems to breed success and vice versa. Weird.

Regarding authors, yes, Dr. Elder has written maybe the most valuable basic book on trading for trader and it is required reading for anyone serious about trading.

Given the current debt crisis and its ties to OTC bank trading and the swaps/currency arbitrage markets, I would like to also recommend a book called Traders, Guns, & Money, by Satyajit Das, which will scare the sh*t out of you if you didn't already know how banks pass paper around. Check it out...

I agree with many of the high level themes being communicated here...have a plan and stick to it, protect yourself when you buy a position (stops or hedges). I think it would have been helpful to also include position sizing insight, meaning never over commit capital to one idea/trade and avoid concentration in correlated assets.

But there is one theme throughout that leans towards quantification of odds in global markets, using the casino and card counting examples, that needs more reinforcement. Robbie mentions randomness and black swans, these are important...I will get to this in a second. Our team spends hours a week examining "fat tail events" (low probability, but highly devastating) with our traders/portfolio managers. MY WORD OF CAUTION: If you run a model or system or plan etc., sit down every single day and question its validity and ability to handle major shocks. Here's the problem... When you step up to a table to play a casino game you know your odds. If astute enough, you can chart the probability of outcomes at some tables over time. Put simply, the table and deck you choose is a contained system that is not susceptible to outside influences. Global markets are the exact opposite. There are infinite variables for which to account and no model/system will ever be able to.

A lot of Nobel prize winning professionals even fail to take notice of this. They develop models to explain the world using "bell curve" scenarios; it allows them to quantify odds of something happening. These models work most of the time and their operators eventually become married to those ideas after seeing repeated successes in predicting/explaining events. But when events occur outside of 'normal distribution' odds (the one in 10 million event) they become momentarily confused before saying...this cant happen twice, or three times in such a short span of time...so they risk more and more capital (remember they are married to the system). You just read the heart and start of almost every spectacular hedge fund blow up.

Folks, this is why these "once in a hundred years" financial crises occur every 5-7 years now. The "once in a hundred years" part comes from some economist's academic paper that uses a flawed model like the above and the world event at hand doesnt quite fit into the research. Its like sitting down at the blackjack table and getting 6 aces dealt to you in a row. Impossible in a contained environment, but we dont live in one.

Someone asked where they can learn more about trading. Start with a book like Trading for a Living by Dr. Alexander Elder. Its not about making a business out of it, but rather a comprehensive introduction into every area of money management. Anyone with an IRA, 401K should read a book like this at some point. The black swan reference comes from Nassim Taleb. Read his books Fooled by Randomness or Black Swan. I guarantee you that you will look at the world and so called experts a lot differently.

[Shundi]

Damn how'd I miss this one!

Another thing is that Buffet almost always gets Preferred stock which is paid a higher dividend and is entitled to liquidation proceeds before common stock - in essence: even if whatever company Buffet is invested in tanks, he gets paid before 99% of the other share holders

Interesting stuff Robbie...