Thursday, January 07, 2010

Managing Trading Risk: Time is Size

The recent post addressed the issue of adding to trades following adverse movement. Here is a trading faux pas that has cost me in the past: I don't add size to a trade that's in the red, but I have added *time*.

In other words, a trade will move a bit my way, then a bit against me, then back to scratch: back and forth for a while. Often this occurs in slow markets.

Normally, my trade ideas include an estimate of time: I should see the market move my way within a matter of minutes from my entry. If that doesn't happen and I stay in the trade, it's been a pretty good predictor of a losing trade.

One reason for that is that adding time is like adding size: it increases the variability of returns. If I turn a short-term trade into an intraday swing trade or an intraday swing into an overnight hold, I've effectively added to the size of the trade. Instead of adding risk on a promising trade, I'm adding it to one that has fallen short of promise.

Equally important, every trade idea with a target and time frame is assuming a particular level of market volatility. If the trade has not moved much in the allotted time frame, it means that the estimate of volatility may be off, which in turn suggests that market participation has declined. Again, this is hardly a good reason to add risk.

Some of my worst trades occurred when short-term trades turned into miniature investments. You don't have to add size to a losing trade to add risk; in the market time *is* risk.

.

5 comments:

Tahoe said...

excellent post, and timely as well. This is a challenge that I too have faced. There is debate around "giving" a trade time to show its character versus "waiting" for it to show the character you are expecting. This can be a fine line as we do not want to "pre-judge" the character. But in so many cases, waiting falls into the hope strategy that I try to stay away from. Excellent post. Thank you!

TraderSmarts said...

Great post as usual Dr Brett. What is the saying: "Never let a bad trade turn into a bad investment" Applies equally to a good trade: "Never let a good trade turn into a bad investment"

BalaB said...

"Normally, my trade ideas include an estimate of time: I should see the market move my way within a matter of minutes from my entry. If that doesn't happen and I stay in the trade, it's been a pretty good predictor of a losing trade. "

Exactly.

I will on occasion (and of course depending on conditions) add to a losing trade and neutralize the loss. So far, I've mostly had a positive experience. But if you asked me: "what are the specific conditions under which you would add to a loss" - I probably couldn't answer with clarity. I just know it when I see it.

(note: I'm a scalper so please take the above with a grain of salt)

JF Trader said...

Spot on! I've learned this from experience the hard way. Now, if it doesn't work right away, I'm out.

Adam said...

Brett ~

The most interesting problems in life are risk and time.
~Fischer Black

Having been bitten by hope: “It’s going to move, it’s going to move,” more times than one cares to remember, I made the discipline of time a core component of each trade put on a few years ago.

Each trade designed has a target return in an estimated time, with longer periods allowed for longs than shorts. Time periods are selected by examining near-term beta to the relevant sector, in the case of sector ETFs, to the S&P, and other factors. This isn’t an exact science and some curve smoothing is involved.

If a trade has not achieved its target return on the chosen date I am out, unless on that date it is showing a buy/short signal that would lead me to deploy another unit of capital.

Horizontal risk ~ capital deployed that is not realizing a return ~ eats into portfolio efficiency. Controlling this component of risk has helped my overall returns measurably over the years.

Thanks for posting on this interesting topic.

Adam.