Tag Archives: sabermetrics

By Rob Bennett I learned about Sabermetrics (the empirical analysis of baseball) by reading Bill James’ Baseball Abstract many years ago. In those days, it was a curiosity. James would argue that a hitter who hits .260 and walks in 10 percent of his at-bats is better than one who hits .290 and walks in 2 percent of his at-bats and the “experts” would dismiss his work as so much foolishness. Today, of course, Sabermetrics has revolutionized the sport. Valuation-Informed Indexing is the Sabermetrics of investing analysis. Once upon a time, we all knew that the stock market is efficient, that price changes are caused by economic developments, that investing risk is stable, the timing never works and that stock returns cannot be effectively predicted. Then this crazy Shiller fellow came along and stood everything we once thought we knew about stock investing on its head. Well, that’s in fact not quite true as of today. But we are getting there, slowly but surely. We are in the early years of a “revolution” (Shiller’s word) in our understanding of how stock investing works. Valuation-Informed Indexing (the model for understanding how stock investing works rooted in Robert Shiller’s “revolutionary” [Shiller’s word] finding that valuations affect long-term returns and that stock investing risk is thus variable rather than constant) is the first true research-based investing strategy. Buy-and-Holders claim that Buy-and-Hold is a research-based investing strategy. But if the valuation level that applies when you make a stock purchase is 80 percent of the story, as the last 34 years of peer-reviewed research shows, it’s not possible to develop effective strategies without taking valuations into account and it’s the first rule of Buy-and-Hold that valuations may never be taken into account (timing doesn’t work, remember?). I came across an article in the Wall Street Journal (” Bill James and Billy Beane Discuss Big Data in Baseball “) that reminded me of one of the most exciting aspects of these revolutionary breakthroughs in our understanding of a field of human endeavor: Revolutions change everything, not just the stuff that we were seeking to change when we began the investigations that led to the revolutions. James started out making the case for on-base-percentage as a better metric for assessing hitters’ skills and arguing that the relief pitchers who close games are not as important as most of us once thought they were and that the best hitter should generally be placed higher up in the line-up. Today insights developed by Sabermetricians are used to inform decisions regarding all sorts of matters that were not on the minds of the pioneers. Most teams use fielding shifts today; that change was brought on through the use of Sabermetrics. Sabermetrics is being used today to prevent injuries to players. Sabermetrics can be used to assess when is the right time to move a player up from the minor leagues. And on and on. So it is has been with my 13-year study of Valuation-Informed Indexing. In 2002, I was posting at a Retire Early discussion board and we all wanted to know when we had saved enough money to hand in our resignations to high-paying corporate jobs. We turned to the safe withdrawal rate studies that were responsible for the infamous “4 percent rule.” I noted one day that those studies do not contain an adjustment for the valuation level that applies on the day the retirement begins. Oopsies! Thirteen years later, the 4 percent rule is universally reviled and most of us are still too ashamed of the mistake to acknowledge that we have sent millions on their way to experiencing failed retirements by our reluctance to correct the mistake we made promptly and openly. But that was really just the first wave of knowledge generated by our decision to start taking valuations seriously. I remember the day when one of my critics demanded that I say what the safe withdrawal rate was when calculated accurately. I didn’t know. It’s easy to say that a study that fails to consider valuations cannot possibly get the numbers right. But I am no numbers guy. I knew that the correct number had to be something significantly less than 4 percent. I guessed that it was perhaps 3 percent when valuations are high, being sure to tell people that I was speculating. There was enough interest in the question that some people offered to work with me to come to develop more precise responses to the question “What is the correct safe withdrawal rate today?” I learned that the safe withdrawal rate can drop to a lot lower than 3 percent. Try 1.6 percent (the number that applied at the top of the bubble). If I had been asked in the early days how high the safe withdrawal rate can rise, I would have probably said that it could rise to something in the neighborhood of 5 percent. Not close! The correct answer is – 9 percent! That’s the safe withdrawal rate that applied in 1982, when valuations were at one-half of fair value. It took me a long time to let that one in. 9 percent! That means that someone with a $1 million portfolio can take out $90,000 per year to live on with virtually no risk of seeing his retirement money run out before he dies. Who’d a thunk it? And that’s still not all. We’ve learned that stocks are not as risky as bonds (for those willing to take valuations into consideration when setting their stock allocations). We learned that economic crises are caused by bull markets. We learned that one form of market timing (long-term timing) ALWAYS works and in fact is required for those seeking a realistic chance of achieving long-term investing success. We learned that stock prices do not play out in the pattern of a random walk AT ALL in the long term, that we always see about 20 years of steadily rising prices (with lots of short-term price drops mixed in, to be sure) followed by 15 or 20 years of steadily dropping prices (with lots of short-term price rises mixed in). Once a revolution gets started, you never know where it is going to take you.