Gannon On Investing

I agree with your analysis. Our company made some great returns on merger deals back in the 80's and early 90's. Since then, the spreads in these types of deals have become much more narrow and more difficult. It has been challenging to find new strategies that incorporate the risk control we look for, and sometimes the inability to look outside the box has hurt us.

Awesome post, itself an example of the original thinking it describes.

This is similar to the "Moneyball" philosophy-- Scouting players on characteristics that weren't looked at before. Since the book came out, those previously overlooked stats like walks and on base percentage are no so analyzed, that it almost makes the original Moneyball philosophy unusable.

http://www.youtube.com/watch?v=PzpojSVQ2rs

You must've written this post before seeing this pitch!!!

Geoff,

This is the first time I've posted a comment to your blog, and I just have two comments on this particular post.

1. A previous commenter mentioned Moneyball, but I don't think he went far enough. In his book, Michael Lewis describes the exact phenomenon you are talking about in the person of Chad Bradford, a relief pitcher. If you look at Bradford's stats in his first year in the league, with the White Sox, they are much better than average for a reliever. 3.23 ERA, NO homeruns allowed, and ~0.3 walks per inning. But Chicago hated him. Why? Basically because he had an incredibly original pitching motion that no one really respected. The funny thing is that this picthing motion has led Bradford to post incredibly consistent results across his MLB career, and allow VERY FEW homeruns. Also, his motion results in a slow pitch, which means a pitcher doesn't blow out his arm, and the guy is durable because of that. Yet, he's still been bounced around across the league because teams don't really like the motion. And even though teaching this type of pitch might allow pitchers to last longer and allow fewer runs, I've yet to hear of any minor league system looking into this. It's precisely the phenomenon you mentioned.

2. You said "They should always be rewarded for winning. They should never be rewarded for the way they won." I would qualify that if I were you. Bill Miller, Michael Mauboussin, and Robert Rubin have all written about the importance of actually doing the opposite - focusing on process and not outcome, and I won't go into the details here. I don't think that you and them contradict one another. I think the difference between the two opinions lies in frequency: If a process is repeated many times and ON AVERAGE, the outcome is quite good, it should be rewarded. Obviously, everyone's goal is to reach a better outcome, but focusing on process is key. For example, seeing Chad Bardford's pitching motion and statistics and blindly saying that it works would be faulty. Focusing on the process by understanding WHY it works (batter seeing the pitch in a weird way, increased durability of the pitcher, difficulty in hitting the ball out of the park due to the physics of the pitch) is as important, or more, than just knowing that it works. Yes, we want to reward people for winning, but we only want to reward them for winning intelligently, so that it's understandable and repeatable, and also to eliminate flukes from our analysis.

I agree with both points.

On the process versus outcome debate, I agree with the importance of process especially in looking at your own performance. However, a system that stresses process over outcome will lack variation and thus will attempt to solve new problems with old solutions. If a corporation stresses a specific process in some of its employees, or some of its functions, it will stagnate and will fail to find better ways of doing the same thing or new ways to solve new problems. However, if some principle is stressed instead – a principle that will remain relevant under all future conditions – there won't be a conflict between allowing variation and adaptation and sticking to that core principle. But, most corporate principles aren't really principles; they're strategies at best and tactics at worst. I think the same is true of a lot of investors and their principles. "Margin of Safety" is a principle – a lot of this other stuff isn't. If you'd consider changing it given the right conditions, it's not a principle.

However, I completely agree with the focus on process for individuals assessing their own decisions. I keep records of exactly what I was thinking (or at least what I was willing to write down) before I made an investment decision as well as excerpts from filings, some of the important ratios, and other information that seemed important to me at the time. A year later and then every two years thereafter I try to take the time to not only assess the performance of that one decision versus the other options then available (not just the S&P 500), but also – and perhaps more importantly – what really lead me to make the decision. Did I overweight certain factors? Underweight others? Did I really buy it because it had a low P/E, even though my notes stress qualitative factors, etc. I also keep some similar records of investments I seriously studied but failed to purchase for some reason. By comparing those to the ones that were purchased, I can get a better idea of how my mind really works in such situations and how I can correct or avoid misjudgments I'm likely to make in the future. So, in that sense, I definitely think a focus on process is a good thing.

Finally, I agree completely with the very last statement in your comment about knowing WHY it works and that it isn't a fluke. When discussing variation – whether among pitchers or anything else – we're usually talking about a record built up over a period of time involving countless individual encounters. A pitcher can be lucky in terms of strikeouts, walks, and homeruns for a day or a month. In terms of wins – sure, you can have a lucky year. But, you don't get very many lucky careers – and the more basic the data you look at the less luck will tend to alter the findings. The same is true of investors, although "luck" is a much more complex factor there, because you can ride a trend for a few years without any knowledge that you're doing it. Investors tend to make many fewer decisions and each occurs under materially different circumstances. Still, you don't look at a career like Buffett's and say maybe he was lucky. In the long-run, results are the best indicator of talent. So, simply ignoring long-run results, because they were achieved in an unusual way demonstrates a dangerous lack of curiosity.