Last updated:
A ratio that rewards smooth, persistent compounding. The quant's favorite.
K-ratio measures how smoothly a portfolio compounds upward over time. Calculated by regressing the log of the cumulative equity curve on time and dividing the slope by the standard error of that slope, scaled by the square root of the number of observations. Higher K-ratio means a straighter, more reliable upward equity curve.
K = slope · sqrt(n) / std_error(slope)K = (slope · sqrt(n)) / std_error(slope)You fit a straight line through your log-equity curve. The K-ratio is essentially a t-statistic for whether the slope is real, multiplied by sqrt(n) so it scales with sample size.
K-ratio rewards what most retail metrics miss: smooth compounding. Two strategies with the same Sharpe can have very different K-ratios if one's equity curve is a near-straight line and the other's is a sawtooth that happens to average out the same.
Created by Lars Kestner specifically for evaluating trading systems. A K-ratio above 0.5 is considered competent; above 1.0 is rare and excellent.
For a strategy with 252 daily log-equity observations, a slope of 0.0004 per day (~10% annualized) and standard error of 0.00015:
K = (0.0004 · sqrt(252)) / 0.00015 = (0.0004 · 15.87) / 0.00015 = 42.3
The scaling varies by convention; some software divides by sqrt(n) rather than multiplying. Foliolytic uses the Kestner 2003 formulation consistently across portfolios.
K-ratio interpretation depends heavily on the formulation (Kestner 1996 vs. 2003 vs. 2013). Foliolytic uses Kestner 2003. Higher is better; relative comparisons between portfolios are more meaningful than absolute thresholds.
Evaluating the smoothness of an equity curve. Most useful when comparing two strategies with similar Sharpe — the one with higher K-ratio has a steadier compounding path.
Sharpe is mean / std of returns. K-ratio is the slope / std_error of the log-equity curve. They capture different aspects of "smoothness."
Depends on the formulation. Under Kestner 2003, above 0.5 is competent and above 1.0 is excellent.
Yes — using the Kestner 2003 formulation. The metric appears in the Advanced section of the metrics panel.
Upload your brokerage CSV — Foliolytic computes K-Ratio plus 70+ other metrics using real historical prices, real Treasury yields, and real CPI data. Free, no signup, your data stays in your browser.
Analyze your portfolio free →