Today I'm going to discuss a fundamental investing distinction between trend-following and predictive strategies. Occasionally I'm reminded that some of these foundational investing building blocks need to be revisited from time to time. So while this is is basic information, it's also really important.
Most investing approaches fall into one of these two camps, though there are countless variations of each. On the one hand, you have investors that are following the market's lead and trying to move in the same direction as the market: these are trend-followers. On the other hand are investors who try to anticipate what the market will do next, independent of what it's doing currently: these are predictors (that's my term, not an official investing label).
It's important to note that a predictive approach may yield the same result as a trend-following approach. But it often will not. For example, a predictive approach today might say that it makes sense to be fully invested in stocks because seasonal and historical patterns suggest the next six months will be strong. Or that it makes sense to be fully invested in stocks because of an expectation that earnings will grow in the coming quarter. Either of these would produce investing instructions that are currently similar to a trend-following strategy, which says to be fully invested in stocks simply because stock performance has been strong recently.
But a predictive approach might also say to get partly or completely out of stocks because today's high valuations, or an expectation that a trade war (or further Fed rate hikes, or a hundred other factors), are going to cause stock prices to fall. There are countless variations, but if you're looking at something that is expected to happen in the future and it's based on anything other than a continuation of the current market direction, it's a predictive approach.
Note that both trend-following and predictive approaches can yield results that aren't necessarily tied to timing the market. For example, our Upgrading strategy is constantly rotating us among different types of funds (even within our risk categories) based on what has been performing best lately. Similarly, a predictive approach may stay invested in the market, but advocate shifting between types of stocks (growth/value, large/small, domestic/foreign, one sector vs. another, etc.) based on what the person or system believes is likely to happen in the future.
SMI uses trend-following strategies, specifically ones based on performance momentum. There is a huge volume of academic research supporting the idea that recent past performance is predictive of the near-term future. Naturally there are many differrent ways to apply this idea. SMI has several different strategies, some of which look quite different from one another. But underlying them all (except Just-the-Basics) is a trend-following/momentum philosophy.
Importantly, this is not new. SMI has been running Fund Upgrading in something close to its current form for more than 20 years. Sector Rotation debuted as a live strategy almost 15 years ago (November 2003). Dynamic Asset Allocation is the relative new kid on the block, but even that strategy is coming up on its sixth anniversary.
Momentum investing — which, again, is just a particular branch of trend-following — works. The research widely confirms it. Even the efficient-markets crowd acknowledges that “Momentum is a big embarrassment for market efficiency” and "the premier market anomaly." So why don't more people use it?
Consider this explanation from a predictive writer I came across today:
My problem with trend followers is that it’s ridiculously simplistic. You want to buy the stocks that are going up, not the stocks that are going down. That is literally the dumbest thing I have ever heard. If it were that simple, everyone would be doing it. The trend followers say that it is that simple. No. Either there is some sophisticated technical analysis going on, or some higher-level reasoning, or something. I have experimented with this. I have bought stocks that are going up. Doesn’t work for me.
Wow, there's a lot to unpack there. But bottom-line, the author nails the primary complaint regarding trend-following and momentum: it just seems too easy — too simplistic. In a world where complex analysis makes people seem smart and gets them paid, it's embarrassing to admit that simply following the market's lead produces better results than a team of economic and market analysts with expensive degrees.
Admittedly, SMI's strategies don't just "buy the stocks that are going up." You might be able to argue there's some "sophisticated technical analysis" or "higher-level reasoning" involved with the research behind SMI's strategies. But not only are we not trying to make things complicated, we deliberately try to keep our systems as simple as possible. The simpler the system and the signals it produces, the lower the chance that we've "optimized" the historical research and biased its results — and the better the chance it will continue to work as we expect in the future.
At the end of the day, the criticism of trend-following/momentum boils down to the simple accusation: It shouldn't be that simple.
Implications for trend followers
If you're going to follow a trend-following strategy (like SMI's), there are two implications to be aware of:
- You're going to run into skepticism of the sort discussed here. This often comes across as condescending. Understand that typically the person arguing in favor of the more complicated system is trying to keep their self-image intact and that image is threatened by such a simple approach yielding better results. In other words, when you hear/read this sort of thing, recognize it's about them, not you!
- As a trend-follower, you're never going to identify the market tops and bottoms in advance. In fact, you'll never be positioned ideally at any market turning point. By definition, if you're following the trend and the trend changes, you're going to be on the wrong side for a while until your system catches up and shifts you. That's just part of the deal, so know and accept it in advance. Doing so will help — it will bother you a lot less when it happens.
The most practical application of this last point currently is the knowledge that at the end of this bull market, we're probably going to be temporarily invested in aggressive stock investments while the market starts declining. If you know in advance that we're going to give back some portion of our bull market gains when the next bear market starts, and that this is part of the long-term design of the system, it should help you not be as concerned/upset when it happens.
The trade-off for giving up some of those gains at the end is being able to stay invested right through the end of the bull market, which is a luxury most predictive systems don't enjoy. We'd rather give up a little at the end than potentially give up a lot by predicting the end of the bull market way too early and missing out on those gains. We've seen this flaw of predictive systems literally throughout this entire decade-long bull market. Many investors distrusted the early bull market and missed the big gains early on, while others distrust it now and have missed out on the returns of recent years. Our way keeps it simple and has gathered all these returns, but the cost is knowing we'll give some gains back when the market finally cycles from bull to bear. We can easily live with that tradeoff.