Many traders use stop-loss orders to prevent big losses on their open stock positions. These sound good in principle -- after all, if a stock's price is collapsing, shouldn't you have a safety net in place to sell at a certain point?

However, in practice, things aren't so simple. We asked three of our own analysts whether they use stop-loss orders in their own portfolios, and their answers may surprise you.

A downward-trending arrow superimposed over a stock ticker board.

Image source: Getty Images.

Dan Dzombak
I don't use stop-losses. You should sell a stock if your initial investment thesis proves incorrect, but if the stock drops for no good reason and your thesis holds, you should be buying at the lower valuation, rather than selling.

In our world of fast-spreading misinformation and flash crashes, you don't want to be forced out of a stock that temporarily drops if you had planned on holding it for the long term. In the case of a flash crash, if you bought the stock back, it would be akin to buying high, selling low, and then buying high again. You would also have to pay taxes on any gains from your forced sale, rather than continuing to defer taxes by simply holding the shares.

While flash crashes are infrequent, they can be caused by countless unexpected events -- and they can also be severe when they occur. One recent example is the Apple (NASDAQ:AAPL) flash crash on Dec. 1. The stock had been trading around $119 per share. In a matter of minutes, the stock plunged to $112 before recovering to $115.

If you had stop losses set at $113, your broker would have sold the shares somewhere below that level, just before the share price immediately recovered to $115. You'd have to buy back in at a higher price, and you'd owe taxes.

Dan Caplinger
I don't use formal stop-loss orders for a very simple reason: They don't work.

The theory behind stop-losses makes plenty of sense. If you had a perfectly continuous market that always rose and fell in an orderly fashion, then stop-loss orders would let you set maximum profit and loss tolerance with precision.

But in the real world, the market doesn't work this way. Every day, a stock can open at a price that's nowhere near where it closed the previous day. If a stock plunges far below your stop-loss order price, then the order will trigger -- but you'll get nothing close to the price where you expected to sell.

Moreover, stop-loss orders give smart traders a chance to take advantage of you. Examples like the one Dan Dzombak discusses are numerous and show how stocks can plunge and recover in short periods of time, running people out of the market at low stop-loss prices and then sending the stock back up toward where it started.

Having a mental stop-loss that gives you the freedom to react rationally to the specific situation does make sense. Turning that over to an automated brokerage process, however, invites more trouble than it's worth.

Jason Hall
In May 2010, I spent 17 days traveling with family, with very limited access to the market. Here's what happened while I was away:

^SPX Chart

I had put in stop-losses for several of my biggest gainers, afraid that I would "give up" some of my profits if the market fell apart while I was traveling. Sell orders triggered on a few, getting me out before an inevitable market crash. But that crash never happened:

^SPX Chart

That's the market since I returned from that trip. How is that cause to sell? It's not, of course, but this experience was a good reminder of the power of human psychology.

I spent months convinced that if I just held out a little longer, I could get back in at a better price. Here's what I missed out on by setting those stop-losses:

^SPX Chart

I did eventually reinvest in both Apple and Starbucks (NASDAQ:SBUX) within a year of selling, but at a higher price than I sold for. The worst part? These were -- and are -- companies that I love for the long term. A stop-loss was good only for my broker, and it was an expensive lesson in opportunity cost for me. Lesson learned.