True diversification is not about the number of positions or the number of asset class labels in your portfolio. It is about how independently your holdings behave under stress. A portfolio of 20 technology stocks is less diversified than one with 8 positions across uncorrelated sectors — and the difference becomes most apparent precisely when diversification matters most: during drawdowns.
Why Asset Class Labels Mislead
Standard portfolio tools categorise holdings by sector or asset class and report the count. This approach has a fundamental limitation: it treats all positions within a category as identical and ignores the correlation structure that determines actual risk reduction.
Two portfolios can have identical sector breakdowns yet dramatically different risk profiles if one holds highly correlated positions within each sector and the other holds positions with low intra-sector correlation.
Correlation-Adjusted Diversification
The Herfindahl–Hirschman Index (HHI) provides a starting point — summing the squares of each position's weight to produce a concentration metric. But raw HHI still ignores correlations.
A correlation-aware view of diversification treats positions that move together under stress as providing less diversification benefit than positions that behave independently. This is the lens KlirInvest applies in its analytics — for example, the Health Score's correlation-stress penalty captures regime-dependent co-movement that static concentration metrics miss.
Example: A portfolio with 10 positions and raw HHI of 0.10 (appearing well-diversified) may behave like an HHI of 0.22 once correlations are accounted for, because most positions co-move under stress — revealing exposure that raw metrics miss.
An Illustrative Diversification Rubric
The rubric below is illustrative — it is one way to think about the dimensions of diversification, not a calibrated or authoritative scoring scheme. Real composite scores (including KlirInvest's Health Score) use different inputs, different weights, and different reference distributions.
Structural Balance (illustrative bands)
- Single asset class: lowest band
- Two asset classes with meaningful allocation: middle band
- Three or more asset classes with intentional weighting: highest band
Geographic Distribution (illustrative bands)
- Domestic holdings only: lowest band
- Domestic plus one international region: middle band
- Domestic plus two or more international regions: highest band
Sector Concentration (illustrative bands)
- Largest sector exceeds 50%: lowest band
- Largest sector between 35–50%: middle band
- No single sector exceeds 35%: highest band
Correlation Quality (illustrative bands)
- Top 5 holdings all declined together in the last correction: lowest band
- 3–4 of top 5 declined together: middle band
- 2 or fewer declined together: highest band
Interpreting Results
- 75–100: Genuinely diversified — risk is spread across independent return drivers
- 50–74: Moderate diversification with identifiable blind spots
- 25–49: Concentrated portfolio — vulnerable to sector or asset class shocks
- 0–24: Highly concentrated — significant single-event exposure
The Health Score Advantage
KlirInvest calculates structural risk automatically using your actual holdings and current market data. The Health Score (portfolio_health_score@2.1.0) is a single 0–100 composite that combines six weighted convex penalties: volatility, tail risk (CVaR), concentration (largest-position and top-5 weight), correlation stress, drawdown, and data confidence (a data-quality proxy). The rubric above (asset class, geography, sector, correlation quality) is an educational framework — it is not the formula behind the Health Score.
The objective is not maximum diversification — which dilutes returns — but intentional diversification: understanding precisely where your risks are concentrated and ensuring that concentration reflects a deliberate choice rather than an accidental drift.
See how your portfolio scores — run your free analysis at [KlirInvest](/free-tools?tab=analysis).