Sunday, May 6, 2012

Warren Buffett quote on filters

As I've previously mentioned (HERE) how important I think filters are in deliberate practice and achieving expertise in investing, this quote from Warren Buffett during yesterday's Berkshire Hathaway Annual Meeting really struck a chord with me:
"Charlie and I have a number of filters that things have to get through before we'll think about them."
He elaborated a little more and--to paraphrase--if an idea doesn't meet certain filters they don't want to spend time thinking about it and would prefer if the person speaking/giving the idea would not even finish the sentence they are speaking when they realize it doesn't pass the right filters. 

Being in the investment business, we read other people's write-ups and hear other people's ideas all the time. This can be a very useful thing because a lot of these ideas can be very good, the people are usually smart (probably smarter than us), and it is another tool to search for potential things in which capital could be put to work. But the human mind is made to fall for stories and miscalculate the odds when a good narrative is in place, as has been usefully described by the work of Nassim Taleb and Danny Kahneman, among others. 

And that is why I think filters are so important. By not even thinking about things that don't pass certain filters, you'll probably miss plenty of good ideas, but you'll also avoid plenty of good stories that turn out to be bad investments. And as it takes a 100% gain to make up a 50% loss, for example, it is probably much more important to avoid the losing investments than it is to try and pick every winner. Or as Howard Marks often says, if you avoid the losers the winners will take care of themselves. So whether potential investment ideas are your own or those of others, I think the most important thing isn't necessarily the number of things you look at, but rather knowing when you should stop looking at that idea and move on to something else before your own psychology makes you see things that may not really be there.