Alpha Theory 2022 Year in Review: The Year That Broke the Streak

For the first ten years of Alpha Theory’s dataset (2012-2021), a portfolio using Alpha Theory’s optimal position sizing outperformed clients’ actual sizing every year by an average of five percent. 2022 broke the streak with the average optimal portfolio down -14.6% versus the average client down -13.5%. Of course, a winning streak is nice, but a 91% win rate is nothing to frown about. Especially, since it looks like the streak will start again in 2023 as, year to date through August, optimal is up +10.1% versus the average client up +6.3%.

Because Alpha Theory clients operate with ~125% of leverage vs. the industry median of ~150%, per a 10-year study from Morgan Stanley, analyses are based on Return on Invested Capital (ROIC). ?

The goal of publishing our data is to convince active fundamental managers that there is a better way to size positions. Even our own clients leave returns on the table. Over the past eleven years, the compound return is twice that of their actual performance, at 215% vs. 101%, and over 2.5x that of the average hedge fund, 215%ii vs. 80%. (Sidenote: four percent additional return for eleven years doubles the returns. Isn’t compounding amazing?)

HOW OFTEN DOES IT WORK??

On average, returns from optimal position sizing have topped returns from actual position sizing for 10 of 11 years. But it doesn’t win for every client and every position. If we randomly select a client, optimal sizing is better 66% of the time. If we randomly select a position, optimal sizing wins 57% of the time. What we see in the results is the benefit of consistently applying process. The more time spent applying process, the more likely the process is to win.?

THE 90%

Every active manager has seen the comparison below that 90% of active managers underperform their benchmark (Table 2). It’s depressing.

Here’s the positive. There is a way to increase the chance that you are not part of the 90%. Our data suggests that fundamental research and stock picking are not the problem. The Alpha Theory Optimal return is a measure of fundamental research and stock picking skill because the optimal sizes are calculated using the price target forecasts and conviction levels scored by the analysts. The research-based sizing (Optimal) outperforms by 4.4%. An equal-weighted portfolio, which is pure stock selection and includes no sizing skill, outperforms active managers’ sizing by 1.6%. Active position sizing reduces returns!

The good news. The data confirms there is research skill. And there is an easy way to get even better returns from this alpha-rich research. ?

PROCESS ENHANCES PERFORMANCE?

Alpha Theory clients are a self-selecting cohort who believe in process and discipline. Below are some of the best lessons for turning process into performance.?

START WITH PRICE TARGETS

Alpha Theory research shows that ROIC for long equities with price targets is 6.2% higher than for those without price targets. Some investors chafe at price targets because they smack of “false precision.” These investors miss the point because the key to price targets is not their absolute validity but their explicit nature, which allows for objective conversation of the assumptions that went into them.?In other words, the requirements of calculating a price target and the questions that price targets foster are central to any good process.?

KEEP PRICE TARGETS FRESH

Once you establish price targets, keeping them fresh will further increase performance. We observed a notable increase in ROIC for long and short equity positions, where fresh price targets were associated with an ROIC increase of 2.9% and 3.6%, respectively, compared to their stale counterparts (Figure 3).?

KEEP UPDATING YOUR RESEARCH

To better understand the relationship between freshness and performance, we grouped daily position-level data from 2012 to 2022 into quartiles based on freshness, where quartile 1 is the freshest, and quartile 4 is the least fresh. We then calculated the mean ROIC per freshness quartile per day and plotted the cumulative ROIC by quartile (Figures 4 & 5).

Our research indicates that the positions with the freshest research generate the highest cumulative optimal ROIC. The freshest cohort (updating price targets approximately once a week) outperforms on both the long and short side. ?

The second and third quartiles are fairly close in returns (price target updates once a month and once every two months, respectively). These quartiles underperform the freshest and outperform the stale price targets which are updated every 146 days (about every 5 months).

When we explore actual and optimal performance by freshness quartile (Table 4), we find that across freshness quartiles, optimal performance tends to outperform actual performance. This suggests that the optimal algorithm is more skilled in its allocation to high ROIC names.

Additionally, the optimal algorithm works better with fresher research. In the freshest quartile of optimal long positions, we observe an annualized ROIC of 17.9%, compared to 14.4% in the lowest quartile of freshness. ?

For short positions, this relationship between freshness and performance is even more pronounced. Positions in the freshest quartile are the only positions to achieve a positive mean ROIC (positive is good for shorts in this case). The freshest quartile averages an annualized ROIC of 0.6% compared to a ROIC of -6.6% for the least fresh quartile.

This analysis reflects the mean performance of long positions in the all-manager portfolio from 1/1/2012 to 12/31/2022. As the data shows, the more frequently that research is updated, the higher the probability of making good investment decisions. ?

Finally, create a systematic approach to position sizing

If you want to be part of the 10% of managers that outperform their benchmark, let humans find great ideas and make forecasts then let systems optimize and implement the human insight. Let Alpha Theory help you and your team become part of the 10%.