Position Sizing: How Big Is Too Big?
If the previous pieces in this series built out the layers of a post-exit portfolio, concentrated equity treated as a portfolio question, the active-versus-passive framework, the choice of an active manager, and the fixed-income layer most often underweighted, this one is about the operating discipline that determines whether the portfolio holds together over time. Position sizing sounds technical and inside-baseball. In practice it is the question that decides, over a long enough arc, whether a thoughtfully-constructed portfolio remains thoughtfully-constructed five years on, or whether it silently drifts into concentration risk no one consciously took.
The question is deceptively simple. For any single holding in the portfolio, a fund, a stock, a private investment, a structured product, how large should the position be, and at what point does it become too large? The honest answer is that most investors and most advisers do not have a written answer, and so the answer that ends up implemented is whatever the price action and the inflow timing happen to produce. That is not sizing. That is drift.
Two failure modes.
Sizing produces two characteristic failure modes, and both are expensive in different ways.
The first is the position that is too small to matter. An investor identifies a high-conviction opportunity, a fund manager whose work they admire, a private investment that has been independently diligenced, an individual holding they have followed for years, and allocates one percent, or two, to the portfolio. The investment performs well over the next decade. It contributes essentially nothing to the household's wealth. The work of identifying the opportunity, doing the diligence, sitting through the volatility, and tax-managing the position has been done; the result, in absolute dollars, is rounding error. A portfolio composed entirely of high-conviction positions sized at one percent each is, in effect, an expensive index fund.
The second is the position that has become too big without anyone having decided that it should be. It enters the portfolio at five percent because the conviction is high. It performs well over five years and is now eleven percent. No one has explicitly sized it at eleven percent; the market sized it at eleven percent. If the position falls fifty percent next year, the household's wealth feels the consequence in a way no one signed up for. The investor was running a portfolio in which they were comfortable with a five-percent maximum loss on any single position; they are now running a portfolio in which they are not.
Both errors come from the same root cause: an absence of a written sizing policy, applied consistently, that defines what an appropriate weight is and how the portfolio gets rebalanced back toward it.
A framework for sizing.
A written sizing discipline answers four questions, and answers them before the position is taken rather than after.
What is the conviction level? A position the investor or their adviser has high conviction in, meaning a clear, articulable thesis, documented diligence, and an understanding of what would falsify the thesis, can carry meaningfully more weight than a position that is held because it seems reasonable or because the adviser had to put the money somewhere. Conviction is not optimism. It is the result of work done before the position is sized.
What is the liquidity profile? A position in a daily-liquid public security can be sized differently than a position in a fund with quarterly liquidity, which can be sized differently again than a position in a private fund with a multi-year lockup. The sizing question becomes harder, not easier, as liquidity decreases, because the cost of being wrong about the position size is no longer "trim it next quarter." It is "live with the position until the lockup expires."
What role does this position play in the portfolio? A core fixed-income holding does work that is different in kind from a satellite equity position; the appropriate size for each is governed by what each is trying to accomplish. A position that is sized as if it were a core holding but actually behaves like a satellite, concentrated, volatile, narrowly-thesised, produces unpleasant surprises during difficult periods.
What is the household-scale risk? This is the question most often skipped, and it is the most important of the four. The same position sized identically as a percentage of the portfolio carries radically different consequences for two different households: a household with a fully-deployed portfolio funding annual distributions, versus a household with a portfolio that is supplemental to other income. The first cannot absorb a fifty-percent loss on a fifteen-percent position; the second can. The sizing policy needs to be calibrated to the household, not to the portfolio in isolation.
When to add, when to trim.
A sizing framework that exists only at entry is incomplete. Most of the actual sizing work happens after the position is established, and the discipline that matters is mechanical, not heroic.
For positions that drift upward, the rebalancing rule should be written before the drift happens. A position whose target weight is six percent, with a tolerance band of plus-or-minus two percent, gets trimmed back when it crosses eight; it is not allowed to silently become twelve percent before someone notices and asks. The execution of the trim is tax-aware, preference for harvesting losses elsewhere in the portfolio, attention to short-term versus long-term gains, coordination with the household's CPA on year-end tax planning, but the decision to trim is mechanical. The execution mechanics belong with your CPA. The size discipline belongs with the portfolio manager.
For positions that drift downward, the same logic applies in reverse, with one important caveat. A position that has fallen below its target weight may be a good candidate for adding to, if the original thesis still holds. It may also be a candidate for full liquidation, if the thesis has materially changed. The decision between "add" and "exit" is not a sizing decision; it is a thesis decision. Sizing assumes the thesis still applies. When it does not, the conversation moves to a different question.
Mechanical rebalancing has a quiet additional benefit: it produces a small but persistent contribution to long-term return through the discipline of trimming winners and adding to laggards. The effect is undramatic. Over decades, in many regimes, it is meaningful.
Discretionary sizing exists, and is harder than it looks.
None of the above precludes discretionary sizing, adding to a position when conviction has increased, trimming a position when concern has increased, deviating from the target weight because something in the underlying situation has changed. Discretionary sizing is a legitimate practice. It is also a practice that requires more discipline than mechanical sizing, not less. The reason is that human judgment is more susceptible to recency bias, narrative drift, and price-driven inference than a written rule is.
The test of a discretionary sizing decision is the same test that applies to any active investment decision: can it be articulated, in writing, in terms that would survive being read again two years later? "I added to the position because the price had fallen significantly while the underlying fundamentals were unchanged" is articulable. "I added to the position because the market was overreacting" is not, at least not on its own. The second formulation is the kind of explanation that feels self-evident at the time and embarrassing in retrospect.
An adviser who manages positions discretionarily should be able to walk through their sizing decisions for any holding in the portfolio and explain, in specific terms, why each position is sized the way it is. If the answer is consistently "that's just where it ended up," the practice is drift, not sizing.
Concentration creep.
The most common silent failure of a post-exit portfolio is concentration creep. The investor pays attention to sizing at the moment of construction, signs off on a portfolio that feels well-balanced, and then turns to other matters for a year. By year-end, the portfolio's largest position has compounded into a position three times the size of the smallest holdings; the second-largest has become twice the size of the average; a single private investment has had a substantial markup that no one has trimmed because there is nothing to trim. The portfolio is now meaningfully more concentrated than the IPS sanctioned. Nothing went wrong. Everything went right. And yet the household's risk profile has shifted in a way no one consciously chose.
Concentration creep is the operating-discipline reason for the quarterly portfolio review. The review is not about identifying tactical trades; it is about checking whether the portfolio remains the portfolio the IPS describes. When it does not, the discipline is to bring it back, not because the drift is necessarily wrong, but because the drift should be an explicit decision rather than an emergent one.
The principle.
A portfolio that has been thought through at construction can still drift over time into a portfolio no one would have signed off on at construction. Position sizing is the operating discipline that prevents this. It is mechanical where it can be, discretionary where it has to be, written down in advance in both cases, and reviewed often enough that the portfolio remains the portfolio you actually agreed to.
The next piece looks at the other operating-discipline question that determines long-term outcomes: what happens to the portfolio when the market falls thirty percent.
Alphyn Capital Management, LLC is an investment adviser registered with the U.S. Securities and Exchange Commission. This post is for informational and educational purposes only and does not constitute personalized investment, tax, or legal advice or a recommendation to buy or sell any security. Past performance is no guarantee of future returns. Investing involves risk and possible loss of principal. No portion of this article is to be interpreted as a testimonial or endorsement of the firm's investment advisory services. Please consult qualified professionals before making any investment decisions. This post may contain forward-looking statements and represents the author's opinions as of the date of publication. Such statements are inherently uncertain, and actual outcomes may differ materially. ACML does not undertake to update any forward-looking statement.