Binary Buy/Sell Decisions

[APG12]

Over the last few days I've been thinking about my tendency to approach buy/sell decisions in a binary fashion. When I conclude that a company is undervalued, I will buy my full position (includes room to make a second large purchase should the price fall significantly) in one swoop. If the stock drops but not significantly, I generally don't add to it. If it drops significantly I may make one more large purchase. When I believe that the stock is fairly priced, I sell my entire position.

 

On the other hand, an investor for whom I have tremendous respect does his buying and selling quite differently. When he determines that the security is undervalued, he will begin to accumulate over a longer stretch of time- buying on the way up and on the way down. Similarly, when it's time to sell, he sells some at one price and some at another, slowly working his way out of the position and funneling the cash into other opportunities. He seems to view the situations as much more fluid than I do.

 

Assuming that you're not constrained by volume, how do you guys approach your buy/sell decisions? Do you immediately make a security a full position when you've determined that it fits into your portfolio? Is there a reason to favor a fluid approach as opposed to a black and white approach?

[xtreeq]

Good topic agaglio and I would also like some input from others ont he Board .... personally, I buy/sell in full or half positions but I think the best thing is, in theory, to know how volatile the stock is - if it is not very volatile on a daily or weekly basis, it makes little difference how you enter or exit - however, unless its a large cap, chances are the price differences can be dramatic even from one day to the next (+/- 5, even 10%!) ... then it pays to buy and sell in much smaller packets, so to speak, but that requires time and patience, which, I suspect, most of us lack to some degree

[hyten1]

i usually tip toe in, i don't have the balls/courage

 

usually over time, i build up a position, its usually due to the fact i need time to feel comfortable, time to read read read, time to see how others feel, etc.

 

this is not the best, i have miss out on lots of potentials

 

hy

[scorpioncapital]

From experience, I find that jumping into a position too fast can be costly if you aren't entirely convinced of the value of the purchase price.

[rpadebet]

I follow a 3 stage buying process which kind of suits my personal psychology/ability and it mostly protects me from my short comings

 

I tend to look for businesses which have decent ROIC and currently trading at a below average earnings yield. My primary selection criteria is if I can think of plausible scenarios in which the business can die. Personally qualitative factors are more important and if quantitative factors are not too bad, I get very excited about it. I haven't had much luck the other way around. If I find such a business, I immediately want to have an exposure to such businesses.(I find it very difficult to control that urge  :)). So I follow a 3 stage approach.

 

1. Once I find a business I like, I read one recent 10K, 2-3 Q's, listen to couple of quarterly calls, read up discussions on message boards, blogs etc and if I am still excited about it, I open a starter position after 1-2 days. Typically less than 1% of the portfolio, most times a round lot of shares. This is a big shortcoming of mine and I have learnt to consider it as a cost of doing this business. I have to learn to not get too excited about many businesses and minimize this cost.

 

2. With a starter position, I feel like I have some exposure so I can concentrate on my research. I read a few more 10K's, find out about the competitors, read their reports, listen to their calls and try to assess the industry and competitive position. Typically what happens is I find myself liking one of the competitors more and if that's the case I follow that company. However after a week or two if I still feel good about my idea and none of the qualitative factors seem compromised, I open half a position (about 4-5%).

 

3. In order to make it a full position, I force myself to wait for 1-2 quarterly results so that I can understand in real time how my thesis is playing out. If nothing material changes and if the price is still decently attractive, I make it a full position of about 10% ( in my experience very few companies get to this stage, because either the price would have moved or I become aware of previously un-thought sustainable competitive threats.). If I cant make it a full position because of price, I sometimes sell OTM puts around my price and hope to get put. These, in a way, partially fund my starter positions.

 

My sell decisions are mostly based on whether I find other attractive opportunities and how much capital I have left to invest. I clean up starter positions first, half positions next. Full positions I take them down in 2 steps.

 

I only have positions in 10-20 stocks (because I feel I can remember and seriously follow only so many). I have a similar number on my watch list and I usually just skim their reports and conference calls. I cant seem to remember important details about more than that. I consider that another constraint on my ability, so that's why I have chosen the model I follow now.

[TwoCitiesCapital]

I used to buy heavily up front and basically establish a full position on the first purchase. I have since learned over the past 5 years this has been poor for two reasons:

 

1) Generally I'm early and there is soon a better opportunity to invest.

 

2) averaging down is psychologically harder after having established a full position and watching it drop by 20-30%. Even if you make it past the psychological barrier, it hurts that much worse if you're wrong.

 

I generally establish a position over time buying 2-5 separate times as I learn more about the company, grow more confident in its value, and see catalysts immediately on the horizon. I also generally sell in 2-3 phases to pair exposure of fairly valued equities but lower loss aversion in the event of a continuous run up.

 

This process can work against you sometimes but seems to have worked favorably for me and it seems much better than one time buy or sell decisions with the main downside being increased transaction costs.

[frommi]

I make these decision purely by the chart and price action. The longterm decision to buy a stock or sell a stock comes from valuation, but for me it makes no sense to use this information for timing. So for timing i am relying on technical analysis with support/resistance levels and elliott wave chart patterns. I am probably not good in both analysis types, but through the combination of these methods i think i can be better than average.

 

When the technical and fundamental situation is crystal clear for me i buy a full position in one swoop. If its not, i wait most of the time or if the fundamentals are very good and i want to have a feet in the water i start with a half position and buy the other half when the situation is clearer (most of the time after a drop).

 

But when your timing decision isn`t adding alpha it makes more sense to smoothly cost average in and out. In this case the number of buys/sells should depend on your transaction costs, since this is the only variable you can control.

[siddharth18]

I tend to think that value investors are always early - in both buying and selling. But even if you disagree with this notion - buying everything at once implies that you are timing the bottom or the most attractive entry point, which I don't think is possible in most cases.

 

The risk in NOT buying everything at once is the possibility that the price slips by making future purchases more expensive - but how often do you buy something when you know WHEN exactly something will go up in price?

 

The advantage of buying in different blocks/chunks is that between your first and second purchase, you have a time to think, reflect and reconfirm your thesis. To me this is important - psychologically speaking. Buying the second block or third block is more like buying something same for the second or the third time. For some odd reason, I tend to be much less biased (and much more attentive) when purchasing the second or third block, rather than the first block.

[SharperDingaan]

Our initial buy is usually around 30-40% of what we want, there-after we add in successive equal tranches as the thesis plays out. If we're right we're getting fewer shares on each tranche as the price is rising, if we're wrong; we get more shares compensating us for the timing uncertainty in our thesis. We use technical signals to decide when we buy.

 

Our initial sell is usually around a 25-35% gain, & we sell only enough to remove any leverage. There-after we're sitting on it until we've somewhere better to deploy the funds at a better expected future return. In most cases we're selling cycle A at around 75% of peak, & buying cycle B at around 25% of trough. Seasonal trades are a round trip sale of season X in the busy season, & a purchase of season Y in the slow season.

 

Long term value approach, with seasonal trading overlays. Seems to work well, but doesn't do much for volatility.

 

SD