A few weeks ago I got a letter from a reader who actually exists and who isn’t my cat. I’m too lazy to go back and read it again (but hey, email me, yo!), so I’ll just give y’all the gist of it.

“When should you sell a stock?”

Nice work getting to the point, letter writer. I like the cut of your jib. Here’s five Nelson dollars.

There are many schools of thought here. Let’s go through a few of them before I tell you kids how I run my multi-billion thousand dollar portfolio.

Buy and hold forever

This is pretty self-explanatory, right? These types of investors buy stocks they think are great long-term businesses. The intent is to hold them for so long that disposing of them will become the duty of whoever owns the estate. To them, the ideal time to sell a stock is never.

Timmy, go ahead and scatter Grandpa’s ashes and then you can have his Coca-Cola shares. No, I’m not sure why he wanted this done in the parking lot of a strip club either. Just hurry up so we can go home.

When a better opportunity comes along

Say you own a portfolio of 10 stocks because you’re one of these guys who would rather add to your top idea than put new money into your 11th best idea. All of a sudden, something happens to your 6th best idea and you think it’s dogshit.

At the same time, you discover a new stock or maybe you warm up to your 11th best idea. Whatever, it doesn’t matter. So the 6th best idea is punted and the money is moved into something new.

Back to break-even

This strategy is practiced commonly on stocks that are losing money. Once it gets back to break-even, the investor will get rid of the dog and move on with their lives, original capital intact.

This is a terrible idea. The market doesn’t care what you paid for a stock. What matters is its future business prospects. If your long-term thesis is still alive, then continue to hold or even think about averaging down. If it isn’t, get rid of it faster than the hooker’s body the morning after. You sick freak.

It’s May

People actually sell in May and go away. That’s dumb and they’re dumb.

Valuation

This one actually makes all sorts of sense, especially for value investors. You buy something when it’s cheap and you sell when it gets expensive. The issues with this process are a) expensive stocks can always get more expensive and b) how do you determine if a stock is expensive? And what valuation methods do you use? Should book value be the best metric? Earnings? Free cash flow?

Figuring out when to sell a stock is hard, yo.

It’s gone up a lot

But how much is a lot? Is 20% enough? 50%? 100%? Ideally, you’ll want to combine this with another factor, like valuation. If a stock goes up a lot and it has an expensive valuation, then maybe look at selling. Maybe.

My selling process

What you’ve all been waiting for, assuming you didn’t hit back four minutes ago. Here’s how I determine when I’m going to sell a stock.

(Note: much of this is copied from the guys from Contra The Heard, two of my favorite value investors)

Since I’m usually trolling the 52-week low list (sometimes even the multi-year low list), I’m waist deep in a lot of stocks that resemble turds. My investing process involves wading through a lot of crap before I find good stuff.

One thing about beaten-up stocks is they offer a whole bunch of potential upside if things go right. If something falls from $25 to $5, it takes a 400% return to get back to $25. I’m looking at those kinds of businesses.

Once I identify something I want to buy, I’ll set a target price. This isn’t a very scientific practice. All I do is check out a 10-year chart and pick somewhere about 25% less than the all-time high.

Take TransAlta, a stock I bought last August for $6.66 per share, mostly so I could say “the devil made me do it.” Here’s a 10-year chart.

TA 10 year chart

I picked that nice $20-$25 range as a nice selling point. The stock spent three years in that range after the Great Recession, which I thought was an achievable goal. I picked $22.50 as my ultimate target price because it was right in the middle of the $20 to $25 range.

I told you this wasn’t very complicated.

If TransAlta hits that target price, I’ll get a return of 238%, plus dividends. Even if that takes a decade, I’m pretty happy with it.

What I like about setting a target price before I buy something is twofold. One, it serves as a constant reminder that it’s a long-term play. And two, I’m less tempted to sell after a quick 20%-50% move to the upside.

The problem with my when to sell method is it doesn’t cover what happens if a stock goes down. That’s more complicated. Mostly, I try to identify whether my original thesis is still intact. If it isn’t, I should probably just sell the stock and move on.

But sometimes I’ll hang on. By being patient and not firing something out the door the instant you fall out of love with it, you can sometimes milk an extra 10 or 20% out of it. The problem with that is it’s a whole lot easier to fall back in love with a company after it moves up nicely. It’s kinda like Stockholm Syndrome.

Let’s wrap this up. Selling is much harder than buying. I think you can give yourself an edge by thinking about when to sell a stock before you buy, rather than after. Your experience may vary.

Tell everyone, yo!