I KEEP A PILE OF

worthless stock certificates on my desk to remind myself at all times that investments are tools to

nothing more. They're not your friend, your lover or your family. Stocks are simply pieces of paper, of which, no matter how much research we've done or how much we like the stock, it's our job to sell. In my portfolio nothing is sacred. Even favored names can quickly get kicked to the curb.

As regular readers know, it's my belief that the best way to dump stocks is via the use of stop-loss orders, a basic investment technique I've been espousing long before Martha Stewart made it famous. To review, a stop-loss order is placed below a stock's current market price. Should the specified price (or anything below) get traded, the order is immediately executed at the market's best available bid.

Regardless of whether you take a fundamental or technical approach, stop-loss orders should be an integral part of every trading discipline. They succeed simply by design: By placing one, you're quietly acknowledging that, yes, even great stock picks can end up as lousy trades.

Stop-loss trades are the bedrock of disciplined trading. Yet the most common problem with stop-loss orders is that they never get executed. Instead of actually placing orders, most folks use "mental stops," expecting that once a certain price is hit they'll be available (and able) to make the trade. Few ever are, making mental stops about as useful to a portfolio as mental sit-ups are to abs.

So even with XYZ down 25% from the purchase price, hitting low after low far below where a stop should've been placed, many traders will hang on, naively certain that a bottom and rally aren't far off. Is it denial? Is it delusion? Either way it's expensive to your bottom line.

All brokerages offer orders on a "good-until-canceled" (GTC) basis, meaning that your stop can be placed weeks, even months in advance. If the market ever trades the stop price, your order is immediately filled. So one need not spend each day huddled over the Ameritrade account. Place your orders and let the market, as I often say, take you out.

Once you've committed to trading with stops, the next challenge comes in knowing just where to place them. Some people opt for a technical signal, like placing stop orders at a security's 100-day moving average. Some use a mathematical approach, selling stocks if they decline 15% from the purchase price or a recent intraday high. Regardless of the approach, the most frequent grievance I hear is from people who complain that their stops always get hit, only to have the market reverse and continue higher.

The real purpose of a stop order isn't to save a few dollars, but to ensure that you move on from a trade when a market's trend has legitimately changed. And yes, when you place a stop order a mere 3% below the current market price, it will get hit. So because most stocks, just like a rodeo bull, will try and toss you off before moving higher, it's much preferable to trade a smaller position with a wider stop rather than a larger position with a tighter stop.

By now it should be gospel: Size kills. And with 15% of your portfolio invested in Taser or PetroKazakhstan, a tight stop is all that protects against the possibility of catastrophic loss. At the same time, by placing stop orders a mere 3% to 5% below the stock's current price, you're just asking no, begging for the market to churn you to pieces. Investors end up repeatedly getting stopped out and getting back in, inevitably increasing the position size with each new buy. If you're looking for an easy way to lose money, this is it.

For the highest probability of success, one should aim to take big chunks, not bite-sized pieces, out of any market move. To that end, if you're bullish on XYZ, I believe it's much preferable to trade a smaller position and use a wider stop in order to avoid constantly getting flushed out of your hand. For example, if you were going to use a -10% stop with a 5% position size, I think it's even better to use a -20% stop with a 2.5% position size.

Nobody knows the future. And by trading a smaller position with a wider stop, you're actually giving the market a chance to either validate or negate your outlook. Stops placed on uncontrollably large positions end up hanging traders up on the everyday volatility, not protecting them against the possibility of a serious shift in trend.

As we wrote a few years back, trading decisions are rarely black and white more like shades of gray. And for larger positions, I advocate using multiple stops, a handy approach that works especially well with large, winning trades that might need to be pruned, but not killed altogether.

Say I'm long a big slug of XYZ. With the stock at $50 a share, my first stop-loss order might be near $45, representing a -10% drop in the current market price. At that point, I'd sell a quarter or a third, anywhere up to half of my position. Then I'd wait and watch, setting yet another stop order to protect me should the market decide to continue lower once again.

To me, it's the most rational approach. As the market weakens, I've reduced my risk. But by keeping a portion of the trade in my account, I give the stock a chance to rally back with my holdings still largely intact. If XYZ marches back up to $50, I've still got skin in the game and, depending on my outlook, can quickly re-establish a full position. It hurts to sell losers, but it took me many thousands of dollars to realize that success in the market comes not in fighting the trend, but following it.

A final thought on stops: Although there's no concrete data available, trader superstition suggests that you never place stops at "obvious" prices such as $10.00, $25.50 or other round, commonplace numbers that are likely to appeal to the herd. The rationale? The public is lazy, so when Ma or Pa Kettle establishes a stop, you can bet it's at $30.00, not $29.87. Because markets tend to cluster and churn where the herd's stops rest, I try to pick slightly more unusual prices to avoid getting tangled up in the public's clumsy tracks.

Jonathan Hoenig is managing member at Capitalistpig Hedge Fund LLC. At the time of writing, Hoenig's fund may have held positions in the securities mentioned in this article.