Apr

20

A depiction of Antigone trying to bury her disgraced brother whose body had been denied a proper burialIn order to destroy the myth and hero status of a villain and stop others of the same vein from following in his footsteps, the dead and bruised body should be displayed in all its glory. (see this story from heraldsun.com.au about gangland killers)

In trading the same can be said. Hang that P&L sheet from your bathroom mirror. Blow up that chart and stick it on the fridge. As you clean your teeth everyday and pour your OJ, you will have to remember that big spike down in P&L and what on earth you were thinking when you put on the risk. Gone will be the feeling of "it's a new day, a new dawn," and wipe that smile off your face. It will continue to make you do the work and stay in touch, without memories of past grandeurs.

Maybe give yourself a heavy punishment, e.g., a morning exercise routine — 200 sit-ups, 200 push-ups, 30 minute run — until you get the P&L back on track of pre-madness levels. (And at the same time something good could come out of the negative.) 

Ralph Vince writes:

Craig,

I don't know if I would do it with P&L statements, but your post illuminates something I have been thinking about a lot in recent years. At first, it was just a nagging problem, became an obsessive one, and ended in an epiphany which has changed not only how I approach trading, but my results– dramatically. And it is the simple question that often nags us (and, when it does, the discussions usually end unresolved, or "resolved by convention") of "What is a trade?"

Some years back, I had the statements of a famous long term trend follower, and was reverse-engineering his system (because I am sneaky that way). This particular fund always had a position in a market, either long or short. These were futures positions, so, of course, there was the rollover issue going on.Did a rollover constitute a "trade?"

There was a fellow a few years back who would put on, effectively, two positions when the time came to put on one. When half the position saw enough profit such that the stop on the remaining half of the position was hit would result in a scratch, he would exit that half.
Is that one trade or two?

Is a long term trend following system an accumulation of day trades?

Is what appears on a p&l statement really a trade? Is what the software you are using to research your trading ideas give you a trade that is the same as what SHOULD be considered, by you, a trade?

Here's the solution I came to, and why it was significant to me. I determined that a "trade" is, in effect, anytime I have on a position, sized according to the moment or equity at the moment. For example, a long term trend following system IS a cumulation of day trades (sans commission on most of those trades) if, each day, the position is resized based on the current capital in the account (and this is not to say that size varies proportionally to capital — it can even vary inversely, but it varies). In effect, anytime the compounding aspect of consecutive "trades" is affected– that is a trade.

If I trade 10,000 QQQQ always, and never vary my size…it is one trade.

Why was this rule important to me? Because, the results of my trading are not the p&l statements, but rather the multiplicative effect of this stream (in fact, I regard each trade by the term I and others call an HPR, which now makes it a relevent mathematical entity). So I am racking up HPRs into a type of queue, if you will, and their multiplication together IS what I am making or losing, and does show me the drawdown I am really experiencing.And from that, I can now go and craft my trading around creating "trades" (HPRs) to append into this queue……and it is the mathematical "shape" of these HPRs that becomes important, because that IS the effect of my trading. Rather than having arbitrary "trades" handed to me, and looking at where I am at time T, and saying "Oh, look what happened there…that's odd," I am able to steer things much more in the direction I wish them to travel.

Lastly, as I said, it doesn't matter if quantity is varying in direct proportion to capital or not– in fact, it is the very function of HOW it varies, in this respect, that is the REAL puzzle of money management (for example, direct proportional variation is simply a "fixed fractional" approach, wherein one fraction would result in optimal geometric growth). But optimal geometric growth may NOT be someone's criteria– and it is the very function of how your size varies relative to your capital that must be crafted to satisfy what someone's criteria is.

Gibbons Burke replies:

Interesting insight, Ralph.

So, if I understand you correctly, a trade is any action which changes a position's risk connection to the portfolio, including initiating it? Or is a trade considered when an HPR can be calculated, that is, when an already-initiated position is changed?

As an example, how do you account for the fellow who uses the old floor trader's trick of taking half the position off the table when his objective is reached, and letting the rest ride. When the position is initiated, do you initiate two trades in your tracking system, or does the single initiating trade #x get split into trade #x.1 which is completed logging an HPR (holding period return), and trade #x.2, which remains active? What if profits are take on position x.2 when another objective is reached - you then get trade x.2.1 which is realized as an HPR, and trade x.2.2 which continues to ride?

How difficult is it to reconcile your way of trade accounting with brokerage statements, and reports which must be made to taxing authorities?

Ralph Vince replies:

Until I am flat the tradeable, I consider it as one trade. One I am flat– I have an HPR (to add into the queues of HPRs) The reason is that the initiation of any position in a tradeable is a function of the current state of compounding/equity.

So, for example, I buy 1000 shares of XYZ. IT goes up, I am still in it but now I buy 1200 XYZ (I have a total of 2200, the subsequent 1200 buy a function of the equity at the time). Thus, I would consider this two trades, two separate HPRs. The first one would be on 1000 shares and it would be considered closed when I added to it (!) The second position now, 2200 shares…ad infinitum or my own physical demise (whichever comes first!) Similarly, if I buy 1000 shares, and I say, I am buying 1200 shares, irrespective of account equity at the time, but rather based on account equity at the establishment of the first part of this position, then, this is one trade when closed out.

Chris Cooper writes:

Ralph's way of looking at the trades makes sense when the reasons for increasing size are due only to some function of your equity. But often, changes in your position are due to other causes. For example, one might trade 1000 shares of XYZ when the price reaches one standard deviation above a moving average (Bollinger Band), and 1000 more if and when the price reaches two standard deviations above the average. These should also be considered two separate trades, but entered roughly in parallel, not strictly serially as in Ralph's example. The parallel case implies different characteristics in modeling risk and reward, whereas the serial case is more straightforward.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search