V2 Ratio
A performance measure built on the relative curve, the portfolio divided by its benchmark. It divides the annualized growth rate of that relative curve by the quadratic mean of the drawdowns of the same curve. A portfolio that stays close to its own relative peak while it outgrows the benchmark scores highly.
Overview
The V2 Ratio is a drawdown-based performance metric that goes beyond the Calmar and Sterling Ratios in two ways. It works on the relative curve against a benchmark rather than on the portfolio equity curve, and it uses every drawdown observation rather than one worst episode or a set of annual worst episodes.
The intuition is that an investor holds the portfolio instead of the benchmark, so the relevant underwater periods are the ones where the portfolio trails its own best prior standing against that benchmark. The V2 Ratio rewards a strategy that pulls ahead of the benchmark and then holds the lead. It penalizes a strategy that gives back its lead repeatedly, even if the two end at the same point.
The denominator is a quadratic mean, not a centred standard deviation. Every relative drawdown is at or below zero, so the series mean is far from zero, and the two forms give materially different numbers. The section below states which one FolioLab computes.
This metric is related to the broader class of drawdown-based performance measures studied by Eling and Schuhmacher (2007), who showed that for normally distributed returns, many risk-adjusted performance measures produce similar rankings, but for non-normal distributions (common in hedge funds), drawdown-based measures can provide meaningfully different insights.
A note on provenance. The V2 Ratio has no peer-reviewed paper and no institutional methodology document behind it. It comes from a 2011 SeekingAlpha post by Emmanuel Marot of Zenvestment, and from the Wikipedia article that cites that post. Read it as one view of relative drawdown, and prefer a documented measure where a documented one answers the same question.
Mathematical Formulation
Core Formula
The published formula divides the annualized growth rate of the relative curve by the quadratic mean of that curve's drawdowns:
Both halves read the same series . The numerator is its growth rate, annualized by the periods-per-year factor . The denominator is the root mean square of its drawdowns.
The Relative Curve
is not the portfolio equity curve. It is the relative curve, the portfolio compound curve divided by the benchmark compound curve at the same date:
A rise in means the portfolio gained on the benchmark over that step. A fall means it gave ground back. The running peak is therefore the best standing the portfolio ever held against the benchmark, and a relative drawdown measures the distance below that standing.
This is why the measure needs a benchmark. Without one it is a different statistic. FolioLab computes it against the benchmark selected for the run, and returns zero when the aligned benchmark series holds 10 observations or fewer. The numerator annualises with over the count of aligned daily observations.
Benchmark basis. A price-return index omits the dividend income of its constituents, while the portfolio returns include it. On a price-return benchmark the relative curve therefore rises faster, by approximately the index dividend yield each year, so the numerator reads high. A Total Return Index removes that gap. Each run records the series it used in its benchmark_basis block, and a fund-manager run that names no benchmark measures against the NIFTY 500 Total Return Index.
The Denominator is Uncentred
The denominator is a quadratic mean of the relative drawdowns. It never subtracts the series mean:
Every is at or below zero, so the series mean sits well below zero. A centred sample standard deviation would subtract that mean and measure the spread of the drawdowns around it. That is a different quantity, it is not a recognized V2 denominator, and it gives a materially different number.
FolioLab divides the sum of squares by the number of observations . The published formula divides by . The gap is one observation in a daily series and moves the result very little.
Intuitive Interpretation
Read the V2 Ratio as the growth of the lead over the benchmark, divided by the typical size of the giveback. A strategy that pulls ahead and then holds its relative peak has small values and scores highly. A strategy that wins and loses the same lead repeatedly carries large values and scores poorly, even at the same end point.
Because the denominator is uncentred, it charges for depth and for time spent underwater together. A curve that sits 10% below its relative peak for most of the sample is charged for every one of those dates, not only for the worst one.
Advantages & Limitations
Advantages
- Benchmark-relative: Both halves read the relative curve, so a drawdown the portfolio shared with its benchmark does not count against it.
- Depth and duration together: The uncentred quadratic mean charges for every date spent below the relative peak, not only for the worst one.
- Robust to outliers: Less dominated by a single extreme event compared to the Calmar Ratio, since it averages the squares of the entire drawdown series.
- Non-parametric: Makes no assumptions about the distribution of returns.
Limitations
- No formal source: There is no peer-reviewed paper and no institutional methodology document for this measure. It is far less adopted than the Sharpe, Sortino, or Calmar Ratios, which limits comparability.
- Benchmark dependent: A different benchmark gives a different relative curve and a different number. Two V2 Ratios are comparable only when both use the same benchmark.
- Data frequency sensitivity: The quadratic mean of the drawdowns depends on the frequency of observation (daily, weekly, monthly).
- Interpretation complexity: Less intuitive than simpler metrics; requires understanding of both drawdown dynamics and volatility concepts.
- Path dependency: Like all drawdown metrics, the V2 Ratio is path-dependent and can differ for portfolios with identical return distributions but different return orderings.
References
- Marot, E. (2011). "The V2 Ratio: A New Way to Measure Investment Performance." SeekingAlpha. The original source for this measure. It is a blog post, not a peer-reviewed paper.
- Caporin, M., & Lisi, F. (2011). "Comparing and Selecting Performance Measures Using Rank Correlations." Economics: The Open-Access, Open-Assessment E-Journal, 5, 1-34.
- Eling, M., & Schuhmacher, F. (2007). "Does the Choice of Performance Measure Influence the Evaluation of Hedge Funds?" Journal of Banking & Finance, 31(9), 2632-2647.
Not investment advice. Past performance is not indicative of future results.