Size Really Does Matter
Smart Thinking: A Skill Versus Luck Essay Series | Issue 8
Written by Michael A. Ervolini
Size Really Does Matter
INTRODUCTION
Position Sizing is a crucial fund manager skill. It determines if the alpha from great buys is effectively harvested and if questionable purchases are restricted in the damage they inflict. Alternatively, chronically underweighting the strongest positions and overweighting the weakest holdings is a headwind if not a surefire path to underperformance. While the asset management industry expends vast amounts of time and energy discussing this activity surprisingly little is known about which managers size positions effectively, which don’t, and how to tell them apart. Fortunately, this situation is changing for the better. The improvement is due to the growing number of firms providing decision-based skill analytics.[1]
POSITION SIZING
One such firm is Alpha Theory. They are laser focused on position sizing. This includes quantification of a manager’s sizing skill, helping managers improve their sizing processes for greater alpha capture, and supporting asset owners/investors in assessing a manager’s sizing acumen. Alpha Theory has analyzed well over 200 equity funds involving more than 14 years of historical data. Their research indicates far more sizing opportunity than success currently, as they relate in their 2025 Year In Review: “Active sizing, the activity managers devote enormous amounts of energy to, reduces returns on average in our dataset.” [2] Specifically, they have uncovered: “Across the past 14 years, the Optimal portfolio has outperformed Actual by an average of +3.9% annualized.” Alpha Theory attributes the lost opportunity to process shortfalls: “The result is a persistent gap between research conviction and capital allocation.”
The way Alpha Theory computes its results is to compare a fund’s actual returns to a counterfactual portfolio in which positions are sized optimally based on Alpha Theory’s proprietary analytics. The latter approach is referred to as “optimum position sizing.” In constructing the counterfactual portfolio Alpha Theory uses all of the manager’s actual decisions of when to buy (open a position) and sell (close a position). Alpha Theory then adjusts how positions are sized to take full advantage of all the information available to the manager. As they describe it: “The Optimal portfolio is not an outside model or black box optimizer. Instead, it’s the manager’s own price targets, probabilities, conviction levels, and risk constraints applied consistently and without the behavioral noise that creeps into day-to-day sizing decisions.” This method assures that the counterfactual constructed reflects realistic and achievable sizing levels bespoke to each fund.
The Alpha Theory research indicates that the results are persistent over time as well. Their report states: “On average, the Optimal portfolio has outperformed Actual in 13 of 14 years — a 93%-win rate.” Interestingly the research shows that higher success is mostly about doing better at the margins: “By Position: Optimal sizing wins 57% of the time — a modest edge per position, but powerful at scale.” This finding underscores the importance of both having an effective process and actually adhering to it.
UPSHOT
Achieving benchmark-beating results is difficult. Doing so requires that each skill is well understood, contributing positively to alpha, and is likely to continue doing so going forward. The work done by Alpha Theory enables asset owners and allocators to formulate a deeper understanding of how well the manager is capitalizing on their best buys. With 4% hanging in the wings this type of analysis is invaluable.
CONCLUSION
Position sizing can be either a source of incremental alpha or a risk factor. Developing and then relying upon a sound sizing process is what makes the difference, according to research from Alpha Theory. Their findings are confirmed by other decision-based analytics investigations.[3] Interestingly, Alpha Theory found that on average the funds studied would have performed better even if the positions were equally weighted. However, managers can do much better than this they encourage: “The solution is not to abandon sizing, but to structure it. In our dataset, when those same research insights are applied through a disciplined sizing framework, they outperform equal weight by an additional 2.2%. That is the real advantage: not better stock selection, but better alignment between capital and conviction.”
Decision-based analytics such as those provided by Alpha Theory are substantially improving the industry’s understanding of manager skill. These newer analytics can rigorously quantify individual skills like sizing and compute its consistency. This information supports stronger allocation decisions and enables fund managers to become more self-aware and improve.
ENDNOTES
- Michael A. Ervolini, Skill Versus Luck: Taking The Guessing Out Of Equity Fund Selection, MIT Press, February 2026, Chapter Three.
- Cameron Hight and Justin Olson, “Alpha Theory 2025 Year in Review: Position Sizing as a Persistent Edge.” The full text is available at: https://www.alphatheory.com/blog/alpha-theory-2025-year-in-review-position-sizing-as-a-persistent-edge
- Michael A. Ervolini, Skill Versus Luck: Taking The Guessing Out Of Equity Fund Selection, MIT Press, February 2026, Chapter Thirteen.
The ideas expressed on this website are developed and/or curated by Michael Ervolini. Mike has spent his entire 35 year + career leading efforts to improve and strengthen active management.