# Trading Efficiency, from Tom Ryan

June 6, 2007 |

Last night while swimming laps I was thinking about efficiency of movement through the water. Actually this started last Friday happy hour when I took a pile of people from here to the pub and they happened to have the America's cup on TV. Several of the folks here are part-time sailors and while watching the kiwis fend off Italy through tack after tack a seed of a thought started germinating in my mind.

Efficiency is a critical component of all kinds of systems involving movement. In fluids for example we know that flow can be steady or unsteady, laminar or turbulent, and uniform or non-uniform. The more the fluid flow is steady, laminar, and uniform, the less energy per unit mass it takes to move the fluid. Unlike solids, however, elements of a fluid mass may move at different velocities and be subject to different accelerations. Still, we have three fundamental anchors: the conservation of mass (continuity), the principle of kinetic energy (flow equations), and the principle of momentum (from which the forces exerted by fluids can be established). What I am most concerned with however in this gedanken is efficiency, therefore energy relations. The most basic energy equation for flow in a fluid mass with boundary constraints, say in a channel or a pipe is (Energy at point 1) + (energy added) - (energy lost) - (energy extracted) = (energy at point 2).

In swimming, energy is added via the muscles, energy is lost due to drag, and there can be a slight effect of energy being extracted depending on the movement of water in the pool. For example, in racing at elite levels measurements have shown that the racers who are lagging behind the others can get slowed ever so slightly as they approach the wall if the wave front generated by the lead swimmers hits the wall and is already bouncing back into them. But in any event the main concern here is with drag.

We know based on the Froude relations that drag of a body or vessel through water is a function of the ratio of the length of the body to its width and depth. In general experiments show that the higher the ratio of length to width, the less drag that is exerted, which is why racing hulls tend to be long and narrow.

In swimming the practical example of this is the use of the method of front quadrant swimming, which was popularized in the 1980s and has recently come back into vogue as people examine Ian Thorpe's technique. (At least as a term or methodology.) It's beyond the scope of this post to discuss all of the minor details of FQS (you can Google it) and there is always some controversy when it comes to swimming technique but the main gist is to maximize and maintain the ratio of your body's L/w in the water in order to lessen drag. This is accomplished by swimming mainly on the sides of your body, using rapid but fluid transitions from one hip to the other, with a long body position in the water that is created by keeping one hand out in front at all times (not straight out in front of the surface of the water but in the front quad in front of your head).

I can say from experience that working with FQS can drastically reduce the number of strokes required to cross the pool, in my case by about 15-20% (3,4).

This leads me to contemplate trading efficiency. To me trading efficiency is not the same as volatility of p/l that is really what is captured by Sharpe, but rather is using the least number of contracts to capture a given level of profit. We know that in trading, as in swimming there is considerable drag in commissions, account fees, and the bid/ask spread which has to be overcome.

Also, as with swimming, I have noticed as I get older that it is not so much the losses that annoy me (although they can definitely hurt at times) but rather the periods where I am churning and getting nowhere. As daytraders we have terms we use to describe this annoying feeling, like "not in gear," "last to get the joke," "the broker is my best friend this week," etc, all of which capture that feeling of trading inefficiency.

It also begs the question as to whether trading efficiency as a metric has any predictive value, at least for traders following a systematic approach to positions. Or, conversely, is there a metric that might be developed for the market movements as whole, say over the course of the week, to assess the degree of market efficiency or the degree of market efficiency as it moves between certain goal posts (1500-1520, 1520-1540, etc)?

We have discussed before how the market often has a tendency to turbulent behavior during the course of the week only to finish nearly unchanged on the week and Victor has proposed that this is one way the system maintains itself, by enticing people to do the wrong thing at the wrong time and thereby make a contribution to the upkeep.

Here are a few additional considerations: Swimming underwater creates less drag, so there are some rules limiting the time, say one stroke, underwater in swim racing. I have a lot of experience swimming underwater trading and in the ocean when I wipe out surfing, but it seems to be part of the game and is a necessary component that might be factored in to your equations.

I like spending only the optimum time underwater, ideally none, but only a breath or two. A two-wave hold down is serious and approaches drowning. Some time underwater is necessary to pop through waves, but also to build a position, as it's impossible to get a full boat on the bottom tick every time.

The second consideration in water is planning. At a certain speed and with certain hull designs the boat starts to plane up on the surface with a break out from the limitations of the hull length/width speed formula into a new equation. When surfing, the commencement of the plane from paddling through the water is the start of the 'ride'. The length of time in a trade getting to the profit ride part seems very important in efficiency terms.

The longer and faster profit/ride, the less time underwater slogging, the greater efficiency, especially tallied over a large number. Paddling a surfboard through the water, getting onto a plane, and jumping up, while at the same time not missing the wave, takes incredible timing and strength. This is a key to trading as well, to time the opportunity, to have the strength to plow through water, sometimes under water, and get to the plane, and ride the ride, maneuvering through the changing face of the wave. Maneuvering around the sections, under the curl, until the wave ends and before hitting the reef takes skill.

Just like every trade requires such strength and courage to enter with the risk of going underwater, skill to read the changing market turns and getting the maximum distance. These things could be quantified in performance stats, but lost opportunity must also be added in as a variable, and lost opportunity has been the key variable in the recent cycle.

## From Vincent Andres:

Here is a humble suggestion. Let each trade be evaluated automatically by your computer according to some criteria (many have been given here). There may well be around 10 criteria. Evaluation may be a note between [-100, +100].

After enough trades, have a look in the criteria space to see if there is some clustering or if the distribution is random or not. Also, this is nothing else than an alike computerized version of Chair's "keep in a notebook" recent advice.

`SELECT * FROM wp_comments WHERE comment_post_ID = '1714' AND comment_approved = '1' ORDER BY comment_date`