The Problem With Patterns

I recently had a funny experience. In late October, financial stocks took another dive (at one point my BMO and NA were down about 10% from my average purchase price). The yield kept climbing, and I just couldn’t believe Canadian banks were going to go out of business, so I was tempted to buy more. Speaking with a friend, I quickly talked myself out of this (I’m not working right now, so the purchase would have come from a LOC, which probably isn’t the smartest idea).

Fast forward a week, and the stocks have almost gone back up to my average purchase cost (NA was $50.50 on October 23rd, and its $54.20 and climbing as I write this on Oct. 31st). I’m looking back, seeing that my feeling was correct, and that I missed out on some rapid increases over 8 days.

So next time, I’ll trust my gut and buy? No I won’t.

Humans has a remarkable pattern matching ability that helps us figure out what’s happening in our environment, and predict what will happen in the future. In many areas this is very helpful, but in a few it can cause weird behaviours or hurt us. Optical illusions play on our pattern matching to confuse us. Gambling preys on this pattern matching where we think we’ve figured out what’s going to happen (and trust the feeling as we lose all our money).

If NA had been guaranteed to make a speedy recovery and shoot up in value, institutions and other investors would have been buying it while it was dropping and would have never let it go as low as it did (they wouldn’t leave money on the table). There was some risk that it would drop further, or it would have stayed at its lower value for an extended period. Its price reflected all these possible future outcomes.

I was interested in my own reaction when it went up, because first I thought “I should have bought”, then I thought “I should buy in a similar situtation in the future (since I was right this time)”, then I thought “wait a second, this is the *EXACT* feeling that gets people into trouble!”.

Unpredictable things are unpredictable. Once you start feeling that you have an intuitive instinct about what’s going to happen (or worse, some arcane technical analysis formula that will predict it) you’re turning into a dangerous gambler. Beware that siren’s call, as it could lead you to your doom.

11 replies on “The Problem With Patterns”

I thought you didn’t care about your stocks prices?

You explained previously that you were only concerned with yield. If this is truly the case then your only fear should be a dividend cut. A dropping stock should be bought if you do not believe a cut is looming.

I’m not trying to be annoying, I’m just trying to figure out your strategy and implementation habits.

mg, I could be wrong but I think that’s precisley the point he’s trying to make.

Sometimes your mind tricks you into thinking you should change your strategy…

If that is the point he is trying to make then he made a mistake in his implementation of his strategy. He could have improved his yield if he would have bought NA lower to avg. down.

MG: I did do that (avg. down), the problem was I’ve run out of money to buy ;-). I was debating going (further) into debt to continue averaging down (and get that sexy, sexy high yield it has right now).

telly: You’re exactly right.

Ok, I guess I’m not being clear here, and I apologize.

Let me explain better:

The way I read your post, you saw NA fall 10% from your original purchase price. You thought about buying the dip, but you did not.

My point is that if you are truly investing for yield and not for share appreciation, you should have bought the dip. Whether or not the funds had to come from LOC is beside the point. When the stock dropped the yield increased. I thought that you once stated that you were investing for yield only. If that is the case then NA is a better buy the lower it goes (barring a div. cut)

Where am I wrong here?

It sounds like no one is “wrong” – but rather you have different perspectives?

Buying on dips is one thing. Buying on dips with borrowed funds is another. Buying on dips with borrowed funds with potentially long-term uncertainly as to the effect of the ABCP fiasco on overall profitability is yet another.

I’m facing a similar internal conflict with Starbucks. They are off substantially – but I wonder if this is a short term dip in the grand scheme of things. Or is it? The reason I say that is McDonald’s coffee strategy is apparently paying off and parents no longer need to make two stops on Saturday morning – one for the kids and one for “their” coffee. They can get their gourmet coffee at McDonald’s.

By the time I make up my mind – SBUX will probably have returned to form of course! 🙂

What about buying call options the next time that happens? That way you know exactly what your total downside exposure is and you will only be out whatever the option contract cost if it doesn’t recover – and if it does – you’ll have had a levered effect way more than a 1:1 LOC loan. ?

Additionally, National had a relatively large exposure to the ABCP market – and given that the long term implications at that time were unknown – I can see how it would cast doubt on a quick price appreciation.

And further – since it was exposed – there would be *some* fear that a dividend cut would be necessitated if the ramifications were wider spread and affected profits going forward, no?

Buying a dividend yield financial stock which isn’t in financial trouble isn’t really risky is it? Price doesn’t affect dividend payment- if the dividend payout ratio and cash flows are healthy I am not really sure if you are entering into an unmanageable risk.

Buying it on a LOC when you are between income streams is risky.

TMW: Yes, exactly. If I had had some cash waiting to invest, I would have bought NA and not thought twice about it.

MG: It was the temptation to abandon my strategy and buy NA with my rent money that was a bad idea (even though it would have worked out this time, nothing says it would have next time).

Sorry if I was/am unclear…

Leave a Reply

Your email address will not be published. Required fields are marked *