Risk metrics

Return is only half of the picture. Two portfolios can deliver identical returns over a given period while exposing their owners to vastly different levels of risk. Monitoring risk alongside performance is what allows an investor to assess whether the returns they are earning are commensurate with the risks they are taking, and to identify fragilities before they become losses.

IBviz computes risk metrics at the portfolio level and, where relevant, at the strategy and asset level. This article covers each metric in turn, explains what it measures, and why it matters for ongoing portfolio monitoring.

Drawdown

A drawdown measures the decline from a portfolio's peak value to its subsequent trough, expressed as a percentage. IBviz tracks drawdowns continuously, showing the current drawdown if the portfolio is not at an all-time high, and the full history of past drawdown episodes with their depth and duration.

Drawdown is one of the most intuitive risk metrics because it measures what investors actually experience: how much of their gains they gave back, and for how long. A strategy that returns 15% per year but regularly draws down 40% is a very different proposition from one returning 12% with a maximum drawdown of 10%, even if the average return looks similar. The Calmar ratio (covered in the Performance article) formalises this trade-off by dividing annualised return by maximum drawdown.

Leverage

Leverage is the ratio of a portfolio's gross exposure to its net asset value. A leverage ratio above 1 indicates the portfolio is using borrowed capital or derivatives to hold more economic exposure than the equity it owns. IBviz computes leverage at the portfolio and strategy level, updated daily.

Monitoring leverage matters for two reasons. First, leverage amplifies both gains and losses, so a leveraged portfolio can experience drawdowns that exceed its initial capital in adverse scenarios. Second, leverage levels can drift upward unintentionally as winning positions grow and margin availability increases, taking the portfolio to a risk level the investor did not consciously choose.

Asset correlation

Correlation measures how closely two assets move together. A correlation of 1 means they move in lockstep; a correlation of -1 means they move in opposite directions; a correlation of 0 means their movements are unrelated. IBviz displays a full correlation matrix across all positions in the portfolio.

For portfolio construction and monitoring, correlation is critical because diversification only works when assets are not perfectly correlated. A portfolio of 20 positions that are all highly correlated with each other offers far less diversification than its apparent breadth suggests. It is also worth noting that correlations are not stable: assets that appear uncorrelated during normal markets often converge toward 1 during periods of stress, precisely when diversification is most needed.

Value at Risk (VaR)

Value at Risk estimates the maximum loss the portfolio is likely to suffer over a given time horizon at a given confidence level. For example, a 1-day VaR of 2% at the 95% confidence level means that, based on historical return patterns, the portfolio should not lose more than 2% in a single day on 95% of trading days.

IBviz computes VaR using a historical simulation approach, which makes no assumptions about the shape of the return distribution and naturally captures fat tails and skewness present in the actual data. VaR is a useful daily risk budget: if the portfolio's VaR is rising without a corresponding increase in expected return, that is a signal worth investigating.

Expected Shortfall (CVaR)

Expected Shortfall, also called Conditional VaR or CVaR, answers the question VaR deliberately leaves open: on the days when the loss does exceed the VaR threshold, how bad does it get on average? CVaR is the mean of all losses that fall in the worst tail of the distribution.

CVaR is generally considered a more complete risk measure than VaR because it is sensitive to the severity of tail losses, not just their probability. A portfolio with a modest VaR but a very high CVaR is one where losses, when they occur, tend to be severe. That profile may be acceptable for some strategies but alarming for others.

Marginal Contribution to Risk (MCTR)

MCTR decomposes the portfolio's total volatility into the contribution of each individual position. Specifically, it measures how much overall portfolio volatility would decrease if a position were slightly reduced. The MCTR of each position sums to the portfolio's total volatility, making it a complete and additive decomposition.

MCTR is a powerful tool for risk-aware position sizing. A position with a large MCTR is contributing disproportionately to overall portfolio risk relative to its size, either because it is large, highly volatile, or highly correlated with the rest of the portfolio. Reducing it would lower total risk more efficiently than reducing a position with a low MCTR. Conversely, a position with very low or negative MCTR is providing diversification and may be worth holding even if its standalone return is modest.

Concentration metrics (HHI)

IBviz measures portfolio concentration using the Herfindahl-Hirschman Index (HHI), a standard measure of how evenly distributed the portfolio's exposure is across its positions. An HHI of 1 indicates a single-position portfolio; an HHI approaching 0 indicates perfectly equal weights across a large number of positions. IBviz displays HHI both at the position level and at the asset-class level.

Concentration risk is closely related to attribution: a concentrated portfolio may show excellent returns driven by one or two positions, but it is also one bad position away from a severe drawdown. Tracking concentration over time reveals whether the portfolio is becoming more or less diversified as holdings evolve.

Risk-adjusted returns (Sharpe, Sortino, Calmar)

These ratios are covered in detail in the Performance article. In a risk monitoring context, their most useful function is comparative: tracking how risk-adjusted returns evolve over time reveals whether the portfolio is maintaining its efficiency or whether recent gains are being achieved by taking on more risk rather than through better decisions.

Risk metrics not covered by IBviz

IBviz focuses on the metrics most relevant to individual investors and discretionary portfolio managers. Several additional risk metrics are widely used in institutional settings but fall outside the current scope of the platform:

Beta: the sensitivity of the portfolio's returns to a specific market index. A beta above 1 means the portfolio amplifies market moves; below 1 means it dampens them.
Tracking error: the standard deviation of the difference between the portfolio's returns and a benchmark's returns. Relevant primarily for strategies managed against an index.
Factor exposures: decomposing returns into systematic factors such as market, size, value, momentum, and quality (Fama-French, Barra-style models). Requires a factor model and is primarily used by quantitative and institutional managers.
Greeks (options): delta, gamma, vega, theta, and rho measure the sensitivity of options positions to changes in the underlying price, volatility, time, and interest rates. Essential for options-heavy portfolios.
Liquidity risk: the risk that a position cannot be exited at a fair price due to thin markets. Particularly relevant for small-cap equities, OTC instruments, and private assets.
Stress testing and scenario analysis: simulating portfolio performance under historical stress events (2008 financial crisis, March 2020) or hypothetical scenarios (a 30% equity sell-off, a 200bps rate rise). Provides a forward-looking complement to the backward-looking metrics above.