| |
 July 2009 Small Ball P3 and P4 Hit a Lot of Singles, but Few Homers
Investing -- particularly quantitative investing -- is a lot like baseball in that regard. Unless you've got an absolute return strategy that's supposed to offer positive return in all markets and under all conditions, odds will tell you that over the long-term you're probably better off trying to stay ahead incrementally rather than constantly swinging for the fences. As is the case with baseball, investing power hitters also have the most strikeouts, too.
Then again, despite how frequently you edge your benchmark, a really bad month or two can put you well behind -- so far it may take months or even years of winning those small battles to catch back up. That's not to say long-term investment models should go for broke, but rather just a statement of fact.
To see a couple of good examples, look no further than quantitative Portfolios 3 and 4. Both were designed to track and incrementally exceed the S&P 500. No gaming here, their universe of potential holdings is limited to the present constituents of the index. If you look at their cumulative returns relative to the index, however, you'd probably conclude they've both been horrible failures. Without a doubt, they haven't kept up with the benchmark over the past nine years, much less exceeded it, but like the baseball team that falls behind early after giving up a grand slam, P3 and P4 have been chipping away for most of that time. Despite trailing in the overall game, how have they been doing in the smaller, games-within-a-game? This is what they were expected to win, so its a question worth asking.
The Teams Inception date for both models was July 1, 2000, so it's not too hard to see why they were both shelled in the first inning. Stocks were just about at their bull market high, so P3 and P4 had little chance to score before the overvalued market collapsed. Both models must always be fully invested, so there were no ringers to bring in. It's been an uphill battle ever since.
Nevertheless, P3 and P4 have had some good months -- and years -- along the way. Looking at each reoptimization period, P3 and P4 have beaten the index 79% and 71%, respectively. In the baseball analogy, that's like winning six out of every nine innings. Generally that's enough to win the game, or at the very least, will put you in position to win the game. Of course it doesn't matter how many innings you win if you let the other team post big scores in even just a few innings. Frequency can quickly be overshadowed by magnitude if you're prone to falling behind by large margins in single innings. Obviously this happened to P3 and P4 back in 2000-2001, but it really hasn't been a problem since then. Charts 2 and 3 show the distributions of the difference between the models and the index for each of their reoptimizations. In both cases, the blue bars indicate the frequency and magnitude of underperformance. The number of data points differs because of the models have different rebalancing periods. P3 has always been reoptimized every two months while P4 originally had a one-year holding period which was reduced in 2003 to six-months. As a result, it has three 12-month periods and eleven 6-month. P3's two worst periods occurred early in its history, falling behind the index by 22.3% between December 14, 2000 and February 15,2001 and by 24.6% from April 16, 2001 and June 15, 2001. The third worst comparative return occurred in the first two months of this year. Aside from those, all other losses to the index were less than 8%. That's not pitching a shut-out, but it's certainly not serving up a multitude of homeruns, either. In fact, without those early losses, P3 would be challenging the index on a cumulative basis.
On the positive side (green bars), P3 has exceeded the benchmark 56% of the time by 4% or more. For you baseball fans, that's five out of every nine innings. The point is, when P3 beats the index, it's not just edging it out. P4 shows a slightly different pattern. It's biggest loss to the index came right out of the gate, between June 15, 2001 and June 14, 2002 when it fell behind by 13%. Since then the worst loss to the the index occurred at the other end of the time series: -7.7% between December 12, 2008 and June 12, 2009. P4 is besting the index by 4% or more roughly 42% of the time, but unlike P3, it also had a grand slam of its own, beating the S&P 500 by 18.8% from June 13, 2008 to December 12, 2008. That helped pick up a great deal of lost ground in just two months and put it briefly ahead of the index. P4 has a more "normal" looking distribution than P4 because it more closely resembles the traditional bell curve. Nevertheless, it's virtually completed its comeback.
The Last Pitch After being out of sample just under ten years, both P3 and P4 are still "young" models. We've tried to avoid making changes too dramatically or frequently, yet performance has improved from those first few months of the 2000-2002 bear market. One might be tempted to conclude they do well in up markets, but will tend to fall behind in down ones, yet the numbers don't bear this out. Since the start of the current bear market, both have outperformed the index roughly 50% of the time, and in those instances where they've failed, losses to the index have been roughly equivalent to gains. All in all, this suggests the two models are on the right track. After falling behind so badly at the outset, it takes time to overcome the deficit armed with only incremental gains. But that's how the models were designed to work, and it seems that's precisely what they're doing. Yes, it's frustrating to see a cumulative deficit each month, but it's also encouraging to see it shrinking little by little. Like the baseball team that gave up several homeruns in the first few innings of the game, it takes a lot of singles, stolen bases, and sacrifices to cut into the lead. Nevertheless, if you do it consistently, it will work. That's the goal for P3 and P4, and that's what they've been doing. Search this site! Just enter you key word or words: Get current quotes or follow your own custom portfolio,
courtesy of E-Line Financials:
|
|||||||||||||||||||||||||
ART OF THE BEAUTY OF BASEBALL is that it's really a series of games within the game. Each at-bat is a duel between hitter and pitcher. Each half-inning is a battle bounded only by three outs. Nevertheless, a team can win most of these little battles, yet still lose the game. Timely hitting and a little luck often trumps consistent execution. Again, that's the beauty of baseball.
