Answering a reader’s questions!

By: ispeculatornew
Date posted: 08.20.2010 (4:15 am) | Write a Comment  (6 Comments)

      Post a Comment

Good morning to all of you. Today, I decided to answer to a comment that I recently received because it was very well formulated and also many of those questions have been asked either in private or on this blog, it is related to the stock picks that we make. Here is Nathan’s comment:

Hello, I recently started reading your blog and have found the commentary interesting and engaging.

I was wondering if you could please go into further detail regarding how you go about formulating particular trades.

Specifically, what is your methodology/rationale for pairing up two particular stocks?

That is, I understand why you might be long a particular company. And why you might be short a different company. You then seem to pair the two positions, and I’m not clear on how you came about the pairing.

I am assuming your trades are based on the spread in performance of the two positions — is this correct? Do you take equal positions in both sides of the pairing?

How do you pick the second stock for a paired trade? Is it a competitor in the same space that you think won’t do as well, or is it simply a company in a different space that you believe will perform in a non-correlated/independent fashion from the first company? Or is the ‘short’ company selection a method of trying to hedge against possible negative performance of the ‘long’ company?

[I did try to find an explanation in previous posts but was unable to clarify to my satisfaction]

Thank you for your attention and blog postings.


That is certainly a loaded comment, I will do my best to answer as well as possible but feel more than free to ask more questions. Initially, I wanted to give more hindsight into how I pair up trades.

The easiest and preferred way is to trade two stocks within the same sector such as going Long Priceline against short Travelzoo. These type of trades become a “bet” on which of the two stocks will perform better which takes out a lot of the external factors. If one stock is mis priced relative to another, doing a pair trade is an attempt to profit when they become back in line. Because of the small number of internet stocks, I often have to look deeper.

The way I generally proceed is to look at my dashboard to find 1 or 2 stocks that either look very attractive or expensive. They would be the stocks that I would be looking to pair. Then, I will look for a stock that is similar in nature. For example, if I am looking to do a trade against Amazon, I would be looking for ecommerce companies that trade at a comparable P/E, has similar growth, etc. Once I find that, one might look very cheap or expensive compared to the other.

In most cases, I am able to find a stock that can be paired off reasonably well but not always. If you look at the trades that we did this year, you will find some trades that are well paired off (Priceline & Travelzoo, Apple & Blue Nile) but also others that are not as clear such as (Research in Motion (RIMM) and Monster WorldWide (MWW)). That would be a case where I was unable to get a good pair trade and decided to simply take what I considered to be an overvalued and an undervalued stock.

Why do I pair trades and how?

I do pair trades because it gives me no net exposure to the market (more or less), which means that a 20% increase or decrease in the market will have a very limited impact on the portfolio. There are positive and negative aspects about pair trading or “delta neutral” trading but personally it’s a type of trading I enjoy because I am not trying to predict the overall direction of the market but rather a small universe of 2 stocks. I would not do only pair trading but as far as speculative trading, it has been a good recipe for us.

When I initiate the trade, I would go long and short the same amount, and thus be market neutral. As soon as the stocks start moving, I start getting a small exposure and that will remain the case until the trade is actually closed. For example, if the stock that is long gains 10% while the other does not move, I will be net long. If the initial bet was 2000$ on each side I would have 2200$ long and 2000$ short. It does remain very limited though.


Pair trading does have its share of traps however and for me, the Chinese market was one example. Going long a Chinese stock against an American one that is similar in nature was a trade that I enjoyed a lot and worked very well for me. Why? Because the Chinese stock usually displayed higher growth, and a smaller P/E for a comparable size. Seemed like an easy trade. But this year, the Chinese markets have suffered greatly and apart from Baidu (BIDU), most Chinese dot com companies have suffered declines. Which explains why I try to make different types of bets. I would never do 5 trades long of Chinese companies against American ones to avoid such traps.

Bigger bets

If you buy and sell for 2000$ of stock, you will have spent 0$ as you can imagine. However, you do need money to initiate such a trade because your broker will need to protect itself in case I lose money. Generally, they will require about 70% of that value as collateral. In this case, if I have 1400$ in my account, they will let me buy and sell 2000$ worth of stocks which gives me some type of leverage. That is why a 7% movement in the stock actually represents 10% or so for me because of this “leverage”.

No secret recipe

All of this having been said, pair trading has been working for me but like every other method, hard work and discipline are the keys to success in my opinion.

If you liked this post, you can consider subscribing to our free newsletters here


  1. Comment by Vito — August 20, 2010 @ 6:33 am

    This is interesting.

    I thought it would be more expensive than 70c for each dollar. Let me explain with an example.

    When you go long $2000, that is clear, you need $2000.

    However, when you go short, naively speaking, you are effectively borrowing enough stocks worth $2000 and selling it, but you will need to buy it back when you null your position.

    This means your potential loss on the short position is uncapped as opposed as for a long position.

    Drawing sums, I would have thought you need at least 2000 + 2000 * factor, where the factor is >1 and big enough for a broker to cater for a big grow of price of your short.

    This unless the double trade is effectively seen from the broker as a single one?

    I would be interested to know which broker(s) let you do this type of trade. I am for example with etrade, and you can go singularly long and short but without leverage?

  2. Comment by John — August 20, 2010 @ 2:38 pm

    That is really interesting!! It’s a great post. I understand better your strategy. WOW if it worked for you, why couldn’t I hehehe Do you time (technical analysis) when to enter in your trade once you have decided the trades and the directions? Thanks!

  3. Comment by OneDay — August 20, 2010 @ 2:44 pm

    would you pair trading with ETFs again? You’ve done it once I think with countries.

  4. Comment by Frank — August 20, 2010 @ 2:56 pm

    I heard that you can’t short financials and other securities. where can you find that list. how’s does it work for pair trading in that case?

  5. Comment by Zavi — August 20, 2010 @ 3:13 pm

    I’m interested with convertible arbitrage managers. Can you explain in a future post their strategy and the delta hedging concept?

    Also, practically, how does dividend and interest income impact the market neutral of the portfolio?

    Thank you.

  6. Comment by IS — August 22, 2010 @ 3:21 pm

    @Vito – When you say “more expensive”, I think it’s important to agree on the fact that going long and short has some risk. Let’s use an example.

    If you guy 2000$ of stock and sell the same quantity, here is what you would have initially

    Cash 0$
    Long $2000
    Short $2000

    At that point, you can see margin requirements at a broker such as Interactive Brokers here:

    Basically, I think opening the position with a reasonable amount of money, maybe $1500 would be sufficient to avoid closing out too early if the position goes against you.

    If it did move against you, you would be paying interest on the “margin required”, which is not that significant.

    Not sure if this answers the question?

    @John – Yes, I do look at technical factors but not really to decide on a position, more in order to avoid stocks that have strong momentum (example was shorting Baidu last year, a big mistake).

    @OneDay – I have stayed away from it lately, but you are right, I did do a trade on Canada vs Russia last year!

    @Frank – Very true. It’s not a big issue for me since I’m involved in technology stocks but yes, for some specific stocks you must track restrictions. Usually, the end responsibility will be the broker’s so he will advise you if you try to short a stock that you are not allowed to.

    @Zavi – Yes, great idea, will try to discuss convertible arbitrage in a future post.

    As for dividends, they would impact the return on a few stocks such as Microsoft. When you are long, it helps your return and does the opposite when you are short.

    And you are 100% right, each trade is “market neutral” when it is started, but as soon as stocks start moving, they no longer are market neutral. It does have a slight impact on returns but not significant enough so far since I usually have trades going well and others that aren’t.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.