A Turkish delight

Each year for over two decades now, we’ve made a prediction for America’s Super Bowl Champion by applying our quantitative investment research to the teams that comprise the National Football League (NFL). Despite a relatively modest run of correct predictions over the past few years, our long-run performance has held up admirably and has given us the gumption to ply our trade elsewhere.

So, after years of not-so-subtle nudging from our London colleagues, we’ve decided to apply the same methodology to the original version of football. In other words, we’re swapping out the pads and pigskin for diving and direct kicks. This year, Man City of the English Premier League (EPL) and Inter of Italy’s Serie A square off in Istanbul, Turkey, to determine which one will take home club soccer’s most coveted trophy.

A Champions League of their own

As part of the Systematic Edge arm of Allspring Global Investments, we’ve made a living by researching and applying trends that often permeate different markets and asset classes. So, it naturally made sense for us to see how our comparatively naïve interseason mean-reversion heuristic applied to Europe’s counterpart to America’s Super Bowl.

At the risk of being pedantic, and to appeal to those with “Ted Lasso” or “Welcome to Wrexham”–level exposure to European football, the UCL is kind of a big deal. In fact, on our weekly calls with our London-based colleagues, we are reminded ad nauseum that the UCL Final (NOT the Super Bowl, and it’s NOT even close) is the most-viewed standalone sporting event in the world. It’s an immensely popular, grueling gauntlet of matches played between the best collection of teams in Europe, and it’s done so concurrently with its own domestic league and cups-ties throughout the year. With that out of the way, lets dive into the data.

Words on the bees

Similar to our work in the NFL, we set the stage for our prediction by quantifying the performance of all teams in Europe relative to market expectations. These “UCL Alphas,” as we’ll refer to them, are structurally similar to an alpha that one might compute for a security in that they represent an asset’s (team’s) performance relative to market (wagering) expectations.

To help illustrate this concept, let’s take a look at the 2022–2023 season for the Brentford in the EPL.

Table 1: UCL alphas for all 20 premier leagues teams

We’ll assume that for each of the Brentford’s 38 games in 2022, a hypothetical bettor named Tony Ivanovic placed a $100 wager on the “money line” for his team to win. For every game the team lost, Tony would lose $100. If it won, Tony would collect back his $100 for that game, plus an additional sum based on the team’s win probability per market expectations.

After the results for all 38 games were calculated, we added up all of Tony’s winnings and compared that dollar amount with the $3,800 total hypothetically wagered throughout the season. Anything less than $3,800 implies a negative Alpha, and anything greater is a positive Alpha. For 2022, applying this process to Brentford’s results during the season would have delivered $6,441 to the bettor—a gain of $2,641 that computes to a league-leading 69.5% UCL Alpha (as shown in Table 1 for the Brentford’s 2022 Alpha).

What to expect when you’re expecting

To help drive home the concept of performance relative to expectation, let’s put two additional teams from the EPL into focus: Bournemouth and Chelsea. Chelsea was expected to contend for a place in the top 4 with new ownership and a historic shopping spree in the transfer window while a newly promoted Bournemouth side was just hoping to hang on to a place in the Premier League for another year, as the team was the odds-on favorite to be relegated. Perhaps most emblematic of these differing expectations is the fact that Chelsea was favored in 30 of its 38 games while Bournemouth was favored in only two. Thus, when both teams finished the season with 11 wins, this was simultaneously seen as a revelation for Bournemouth investors (26.7% UCL Alpha) and a dumpster fire for Chelsea investors (-49.2%).

Chart 1: Cumulative alphas: Bournemuth and Chelsea

Italian stallions

Panning over to Italy’s Serie A, there was a changing of the guard as Napoli dethroned the usual suspects of Inter, AC Milan, and Juventus for its first title since the late Diego Maradona left the club over 30 years ago.

Table 2: UCL alphas for all 20 Serie A teams

Judging by its Alpha (35.1%), this outcome was a surprise to many, but perhaps the bigger surprise was Monza (39.3% Alpha). The club was formed in 1912, yet the 2022–2023 season marked its first appearance in Italy’s most prominent soccer league. Most notably, Monza took all 6 points from its encounters with Juventus, who could have really used those points after being docked 10 of them for general shadiness. The penalty will leave the Juventus faithful on the outside looking in for the 2023–2024 UCL tournament.

Chart 2: Cumulative alphas: Inter Milan and Monza

Death, taxes, and the “low-volatility anomaly”

We’ve been researching and exploiting the low-volatility anomaly for well over 20 years at Allspring Systematic Edge, and we never turn down an opportunity to explore new frontiers for its efficacy. Time and time again, we’ve found that the anomaly transcends its original application to the equity markets, and professional soccer is our latest positive test result. For each of the major European leagues, we placed every match into one of five groups based on a team’s odds-implied probability of winning. At the end of the season, we tallied up the P&L for each group. Spoiler alert, it works—and it worked swimmingly during the 2022–2023 season. As is typical nearly everywhere, investors appear to overpay for lottery tickets/longshots (-12.23%) in soccer wagering, while a lower-paying—yet more consistently hitting—wager on the favorite proved more fruitful (6.55%).

Chart 3: Low-volatility analysis 2022–2023 season

If it ain’t broke, don’t fix it

Naturally, the first stone we turned over in looking for an edge in the UCL final prediction was inter-seasonal mean reversion. This signal has worked rather well in the NFL postseason (60+% hit rate), and as we often find in financial markets and the world at large, patterns tend to generalize. Over the past 10 years, the UCL final contender with the greatest success relative to expectation (higher Alpha) in its domestic campaign has systematically underperformed its odds in the UCL final. This year, both UCL finalists found it difficult to exceed expectations, as they were favored to win in all but one of their 76 combined league competitions. Despite a relatively humbling third-place finish in Serie A, Inter was slightly more impressive than Man City from an investment standpoint (6.5% UCL Alpha). Consistent production from recent World Cup winner Lautaro Martinez proved enough to secure its spot in next year’s UCL and bring home the Coppa Italia.

Similarly, Man City rode the glorious ponytail of Erling Haaland to an EPL title—and, most recently, an FA Cup title—finishing the season with a 2.5% UCL Alpha. The market-implied odds for the Citizens to lift the trophy on June 10 sit at a possibly unprecedented 75.81%, yet as the lower-Alpha team in this matchup, we see the team as an even bigger favorite.

Look, we’ll be the first to admit that we’re hardly going out on a limb by picking Goliath to take down David in this matchup, but one of these two teams has mean-reversion and a real-life Viking with a low-volatility headwind on its side. Therefore, we think the smart money is on Man City to take home the trophy and complete the treble.

Table 3: UEFA Champions League results

Leave a Reply

Your email address will not be published. Required fields are marked *

You might also like: