Investment performance is not only a question of picking the right securities. It is also a question of how much weight to give them. Sizing is one of the quiet but decisive levers in active management, and measuring its quality is less straightforward than tallying stock-picking hit rates.
In practice, an analyst might show a high hit rate by making many small positive contributions, while another may be right less often but earn more when they are right. Distinguishing between these patterns requires a lens that looks beyond raw returns.
The notion of slugging ratio comes from baseball but adapts well to portfolio analysis. Instead of counting how often an analyst’s picks outperform (hit rate), slugging captures the payoff per unit of risked or allocated capital. It reflects whether big bets deliver proportionally bigger results.
By evaluating both hit rate and slugging across position sizes and holding periods, we can see whether an analyst’s edge comes from consistent singles or from occasional home runs. Neither style is inherently superior; what matters is whether the sizing decision is justified by the payoff profile.
This distinction is important because sizing is not purely an expression of conviction. It is also constrained by portfolio construction rules, liquidity, and the surrounding opportunity set. An analyst who habitually takes small positions may be playing within limits rather than showing low conviction. Conversely, an analyst who concentrates in a handful of names may appear bold but, if the slugging ratio is weak, is in fact wasting scarce portfolio risk.
To move from intuition to measurement, we built a normalized attribution framework that breaks down alpha contributions into cohorts by size and duration. This allows for comparison across analysts, market backdrops, and opportunity sets. Importantly, the framework scales alpha to the sizing decision itself, not just the security selection.
The model looks at cohorts such as “large positions held for 12+ months” versus “small tactical positions held for less than a quarter.” In each case, it computes whether the sizing amplified or diluted the analyst’s selection skill. By controlling for the market environment, we avoid mistaking a period of rising tide for skill in position scaling.
This attribution highlights when conviction-weighted positions carry their weight and when they do not. It also shows whether small-sized positions meaningfully add value or simply serve as filler in the book. Over time, the data builds a profile of each analyst’s true sizing effectiveness.
For teams, the framework turns a vague question—“is this analyst good at sizing?”—into a structured review. It enables managers to see whether analysts deliver consistent alpha per unit of sizing or if performance depends on a handful of large swings. That, in turn, feeds into coaching, compensation, and how capital is allocated across analysts.
There are also broader risk implications. Concentrated positions with low slugging ratios are not only inefficient but can pose outsized downside risk to the book. On the other hand, a strategy dominated by small, low-conviction positions may not justify its complexity if the incremental alpha is marginal. By mapping outcomes to sizing, portfolio managers can calibrate exposure in a way that is both more evidence-based and more aligned with the analyst’s proven strengths.
This perspective has already started reshaping internal review discussions. Books with long-only carveouts, where position sizing has different constraints than in market-neutral portfolios, are exploring how to adapt the method. The larger point is that attribution should not only trace what was bought but also how it was sized.
