There are multiple ways to approach equity investing and, ultimately, the pursuit of alpha. While many strategies rely on market direction or discretionary stock selection, a different approach centers on systematically capturing the underlying drivers of returns embedded within equity markets. At Gothenburg-based ORCA Hedge, a six-person team has built a strategy around this premise, seeking to capture factor premia that have historically delivered excess returns across different market cycles, while acknowledging that these premia are neither stable nor persistent.

“We, like everyone else, have no magic way to see into the future,” acknowledges Mikael Andersson, co-founder of ORCA Hedge. “That means the core is about getting rewarded for risk that is already priced in, but we do it at scale by targeting hundreds of risk factors simultaneously.” Crucially, the performance of individual factors is highly regime dependent. Factors such as value, momentum, or quality can experience prolonged periods of both outperformance and underperformance, often driven by macroeconomic conditions and shifts in market structure. Rather than maintaining static exposure to such premia, ORCA Hedge combines a large number of factors within a risk-controlled framework that dynamically adjusts to changing environments.

“We, like everyone else, have no magic way to see into the future. That means the core is about getting rewarded for risk that is already priced in, but we do it at scale by targeting hundreds of risk factors simultaneously.”

Mikael Andersson

Delivering Equity-Like Returns Through Dynamic Factor Exposure

At its core, the strategy is a long/short equity approach driven by systematic factor exposures, with overall net market exposure kept low or close to market neutral. However, the ambition is not to deliver market-neutral returns in the traditional sense. Instead, the objective is to generate equity market-like returns over time, but with significantly lower volatility and reduced downside risk.

Although officially launched at the start of 2025, the strategy is the result of several years of development, testing, and refinement. “It took our team almost four years from idea to validated strategy and completed fund setup,” says Andersson. This extended incubation period highlights both the complexity of building a robust systematic framework and the uncertainty involved in developing a strategy reliant on multiple interacting signals. 

Over time, the team has also strengthened its capabilities beyond its technical core, adding financial and commercial expertise, while benefiting from the support of its partner FinServe Nordic. “In hindsight, it was a huge risk to put this many years into a project with an uncertain outcome, but at the end of the day we’re happy with the result,” notes Andersson. “So far, we have a strategy behaviour that is aligned with our internal simulations and trading.”

From Factor Selection to Portfolio Construction

The investment process begins with constructing long/short portfolios across a broad equity universe, where stocks are evaluated based on their exposure to a wide range of factors. “Stocks with strong characteristics within a given factor are ranked against those with weaker characteristics,” explains Andersson. However, rather than treating all factors equally, ORCA applies a proprietary screening process to identify which return drivers are currently most effective in the prevailing market environment. This filtering step is critical, as it transforms the approach from static factor investing into a dynamic allocation process that adapts to regime changes.

Once selected, the relevant factors are grouped into broader themes using machine learning techniques, allowing the team to manage risk at a higher level of abstraction. These themes represent distinct sources of risk premia, enabling more efficient diversification and portfolio construction. The team’s technical background, largely rooted in IT and data science, plays a central role in this process. “Machine learning adds most value in allowing us to manage a large set of risk factors, e.g. bringing order to the ‘factor zoo’ and in risk management,” Andersson explains.

“Machine learning adds most value in allowing us to manage a large set of risk factors, e.g. bringing order to the ‘factor zoo’ and in risk management.”

Mikael Andersson

Portfolio construction is implemented through a systematic bottom-up process, where each stock is assigned a weight based on its contribution to the overall factor exposure. Importantly, this approach deliberately avoids traditional company-level views. “The most important thing to understand is that the long and short books are not constructed based on individual company views, but rather as a way to create the strategy’s desired risk factor exposure,” Andersson emphasizes. In this framework, stocks function purely as building blocks, vehicles through which the strategy expresses its exposure to selected factors.

“The most important thing to understand is that the long and short books are not constructed based on individual company views, but rather as a way to create the strategy’s desired risk factor exposure.”

Mikael Andersson

The resulting portfolio is both highly diversified and dynamic, typically consisting of around 250 long and 250 short positions, and is rebalanced on a monthly basis. “At each rebalance, the portfolio construction process reassesses the entire stock universe,” explains Andersson. Whether a stock remains in the portfolio or is replaced depends entirely on its relative ranking within the factor framework. “If a stock continues to rank high relative to the rest of the universe, it will remain in the portfolio across multiple rebalances. If it gets a lower score, it may be replaced by another stock that better contributes to the desired strategy exposure.” As such, the concept of a traditional holding period becomes less relevant, as positions are maintained only for as long as they serve the broader portfolio construction objectives.

A Scalable Framework and a Diversifying Portfolio Component

The strategy’s return drivers stem from the aggregation of systematic alpha signals derived from this extensive universe of risk factors. However, ORCA does not position itself as a pure factor-premia harvesting strategy. “The goal is not to simply harvest traditional risk premia,” Andersson emphasizes. “Instead, the strategy applies a proprietary scoring and filtering process to select the factors that are currently most effective.” This dynamic selection process introduces an additional layer of alpha generation, as performance depends not only on factor exposures themselves but also on the ability to identify which factors are likely to perform in a given environment. “Our secret sauce is that we’re able to extract more alpha the more risk factors we take into account,” he adds.

“The goal is not to simply harvest traditional risk premia. Instead, the strategy applies a proprietary scoring and filtering process to select the factors that are currently most effective.”

Mikael Andersson

ORCA, an acronym for Optimal Risk and Capital Allocation, is designed as a broader framework capable of supporting the development of additional strategies over time. “ORCA is a framework where we are able to quickly iterate and create additional strategies with quality-assured and live-tested components,” Andersson explains. This modular approach reflects a clear awareness of alpha decay, a common challenge in systematic investing as strategies become more crowded. “Our long-term plan for the fund is to carefully diversify among multiple strategies and by that avoid potential alpha decay,” he adds. “We believe that our strength lies within building technically challenging strategies, where the implementation itself is a first barrier. By overcoming that barrier we enter a space where competition for alpha is less intense.”

“Our long-term plan for the fund is to carefully diversify among multiple strategies and by that avoid potential alpha decay.”

Mikael Andersson

The use of derivatives to implement the strategy introduces an additional layer of capital efficiency, leaving excess cash within the portfolio. This capital is currently allocated to short-term bonds, providing an incremental return that complements the core strategy. However, this component is not central to the investment thesis. “Basically we currently treat returns from this part of the strategy as ‘cherry on top’ and we do not depend on it to deliver a product above baseline,” explains Andersson.

From an allocator’s perspective, ORCA Hedge is best understood as a diversifying component within a broader portfolio. Its low correlation to traditional asset classes, including equities and alternatives, enhances its role in improving overall portfolio efficiency. “The strategy aims for equity-like returns over time but with lower risk and dependence on overall market direction,” Andersson concludes. “The strategy has a low correlation to all major asset classes including alternatives, making it a true portfolio component.”