# Probabilistic & Deflated Sharpe Ratio (DSR) Formula Guide

> Eliminate false-positive backtests using Bailey & Lopez de Prado Probabilistic Sharpe Ratio (PSR), Deflated Sharpe Ratio (DSR), and Minimum Track Record Length.

Canonical: https://quantcalctools.com/guides/probabilistic-deflated-sharpe-ratio-multiple-testing-backtest-bias
Last updated: 2026-11-14T19:31:07.612Z

**Publication:** QuantCalc Tools
**Category:** Cagr Irr Drawdown Compounding
**Published:** 2026-11-13T14:00:25.915Z
**Author:** Elena Rostova, FRM, CQF (FRM, Certificate in Quantitative Finance (CQF))
**Reviewer:** Not recorded

## Answer-first summary

Testing 250 strategy variations on 5 years of random noise guarantees an expected maximum backtest Sharpe Ratio above 1.45 purely by luck; master exact PSR, DSR, and MinTRL equations.

## Key takeaways

- Under Extreme Value Theory, sweeping K = 1,000 zero-skill random strategies on a 5-year window produces an expected maximum Sharpe Ratio of 1.45 (and 2.29 on a 2-year window!) purely by chance.
- Probabilistic Sharpe Ratio (PSR) incorporates negative skewness and fat tails via Mertens’ variance weight V = 1 - gamma_3*SR + ((gamma_4 - 1)/4)*SR^2, where Pearson kurtosis gamma_4 = 3 for Normal.
- Deflated Sharpe Ratio (DSR) evaluates PSR against the Extreme-Value expected maximum white-noise Sharpe SR_0 across K_eff effective independent trials, requiring DSR >= 0.95.
- Minimum Track Record Length (MinTRL) tells allocators the exact number of live months or days required before an observed Sharpe becomes statistically significant at 95% confidence.

## Related financial entities

- **Cornish-Fisher Skew & Kurtosis Modified Parametric VaR** (CORNISH-FISHER-VAR) — product: Cornish-Fisher Skew & Kurtosis Modified Parametric VaR (CORNISH-FISHER-VAR) refers to an institutional financial benchmark and regulatory standard monitored by GARP FRM Quantitative Standard (INTL), currently calibrated at 99.0% (Non-Normal Quantile) as of October 2026.
- **Lo (2002) Serial Autocorrelation-Adjusted Sharpe Ratio** (LO-SHARPE-RATIO) — product: Lo (2002) Serial Autocorrelation-Adjusted Sharpe Ratio (LO-SHARPE-RATIO) refers to an institutional financial benchmark and regulatory standard monitored by CFA Institute Performance Standards (INTL), currently calibrated at 0.3842 (Annualized Ratio) as of October 2026.
- **Lo (2002) Serial Autocorrelation-Adjusted Sharpe Ratio — 2026 Standard** (LO-SHARPE-RATIO-STD) — product: LO-SHARPE-RATIO 2026 Institutional Benchmark Framework is defined as the standardized quantitative and compliance methodology published by CFA Institute Performance Standards governing Lo (2002) Serial Autocorrelation-Adjusted Sharpe Ratio across INTL capital markets (0.3842 Annualized Ratio, October 2026).

## Analysis

Modern cloud clusters let a quantitative researcher sweep $K = 500$ moving-average windows, RSI thresholds, and ML hyperparameters before lunch, pick the single equity curve with a $2.10$ annualized Sharpe Ratio, and pitch it to an investment committee. Without adjusting for **multiple-testing selection bias (data snooping)** and **non-normal higher moments**, that $2.10$ backtest is statistically indistinguishable from pure white noise! David Bailey and Marcos Lopez de Prado (2012, 2014) solved this crisis in empirical finance by introducing the **Probabilistic Sharpe Ratio (`PSR`)**, **Minimum Track Record Length (`MinTRL`)**, and **Deflated Sharpe Ratio (`DSR`)**. Benchmark raw ratios in our [Sharpe & Sortino Ratio Calculator](/calculators/sharpe-sortino-calculator), review [Lo (2002) Sharpe statistics](/entities/lo-sharpe-ratio-lo-2002-serial-autocorrelation-adjusted-sharpe-ratio), and anchor in our [CAGR, IRR & Drawdown Hub](/guides/cagr-irr-volatility-drag-sharpe-sortino-drawdown-recovery-guide).

## Exact Equations: PSR, MinTRL, Expected Maximum Sharpe $\text{SR}_0$, and DSR

Work strictly in **non-annualized sample frequency** ($N$ observations, non-annualized sample Sharpe $\widehat{\text{SR}}$, skewness $\hat{\gamma}_3$, and **Pearson kurtosis $\hat{\gamma}_4$ where Normal $= 3.0$**—never confuse Pearson $\gamma_4 = 3$ with excess kurtosis $0$, or the $((\gamma_4 - 1)/4)\widehat{\text{SR}}^2$ term goes wrong!). First, **PSR** computes the probability that true skill exceeds benchmark hurdle $\text{SR}^*$ given non-normal skew/kurtosis. Second, **DSR** sets that hurdle $\text{SR}_0$ equal to the **Expected Maximum Sharpe Ratio across $K$ independent zero-skill trials** via Extreme Value Theory (Euler-Mascheroni constant $\gamma \approx 0.57721566$):

```formula
Bailey & Lopez de Prado (2012, 2014) Exact PSR, MinTRL & Deflated Sharpe Ratio (DSR):
 Non-Normal Variance Weight : V_mertens = 1 - gamma_3 * SR_hat + ((gamma_4 - 1) / 4) * SR_hat^2 [Pearson gamma_4 = 3 for Normal!]
 1. Probabilistic Sharpe : PSR(SR*) = Phi( [ (SR_hat - SR*) * sqrt(N - 1) ] / sqrt(V_mertens) )
 2. Min Track Record Length : MinTRL(alpha) = 1 + V_mertens * [ z_{1-alpha} / (SR_hat - SR*) ]^2 (in sample periods!)
 3. Expected Max Sharpe across K Independent Zero-Skill Trials (Variance of Trial SRs = Var[{SR_k}]):
 SR_0 = sqrt( Var[{SR_k}] ) * [ (1 - gamma) * Phi^(-1)(1 - 1/K) + gamma * Phi^(-1)(1 - 1/(K * e)) ]
 where Euler-Mascheroni gamma = 0.57721566, e = 2.7182818, and under IID unit trials Var[{SR_k}] ~= 1/N (or q/N annualized)
 4. Deflated Sharpe Ratio : DSR = PSR( SR* = SR_0 ) [Require DSR >= 0.95 to reject data-mining null!]

Worked Backtest Overfitting Audit (5 Years Monthly N = 60, Reported Annualized SR = 1.80 -> Monthly SR_hat = 0.5196):
 Assume K = 200 parameter trials tested (Annualized std across trials = 0.45 -> Monthly std = 0.1299),
 Skew gamma_3 = -1.20, Pearson Kurtosis gamma_4 = 6.50:
 1. Expected Max Lucky Monthly Hurdle SR_0:
 Phi^(-1)(1 - 1/200) = 2.5758; Phi^(-1)(1 - 1/543.66) = 2.9045
 Bracket = 0.42278*(2.5758) + 0.57722*(2.9045) = 2.7655
 Monthly Hurdle SR_0 = 0.1299 * 2.7655 = 0.3592 (Equivalent to 1.244 Annualized Sharpe purely from luck!)
 2. Non-Normal Denominator V_mertens = 1 - (-1.20)*(0.5196) + (5.50/4)*(0.2700) = 1 + 0.6235 + 0.3713 = 1.9948
 3. DSR = Phi( [ (0.5196 - 0.3592) * sqrt(59) ] / sqrt(1.9948) ) = Phi( 1.2320 / 1.4124 ) = Phi(0.872) = 80.8%!
 Verdict: FAILS 95% institutional gate (80.8% < 95.0%)! Overfitted across 200 trials with negative skew!
```

Look at how devastating that audit is: an annualized Sharpe of **$1.80$ over 5 years** has a naive Gaussian single-trial $z$-score of $1.80 \times \sqrt{5} = 4.02$ ($p = 0.00003$, `99.997%` confidence!), yet once we account for sweeping $K = 200$ parameter combinations and negative skew $\gamma_3 = -1.20$, confidence plunges to **`80.8%` (`p = 0.192`)**—failing institutional deployment gates!

![Four-stage Deflated Sharpe Ratio backtest validation pipeline counting effective independent trials K_eff and computing DSR confidence](/api/figure?type=flow&title=Institutional%20Backtest%20Validation%20Gate%3A%20From%20Raw%20SR%20to%20DSR%20%3E%3D%2095%25&subtitle=Eliminating%20selection%20bias%20and%20non-normal%20tail%20inflation%20before%20deploying%20live%20strategy%20capital.&steps=1.%20Log%20All%20Trials%20%28K%29%3A%3ARecord%20every%20parameter%20sweep%20%26%20estimate%20effective%20independent%20K_eff%20via%20clustering%7C2.%20Compute%20Hurdle%20SR_0%3A%3AExtreme%20Value%20Theory%20sets%20expected%20max%20white-noise%20Sharpe%20across%20K_eff%7C3.%20Inflate%20Non-Normal%20SE%3A%3AApply%20Mertens%20V%20%3D%201%20-%20gamma_3*SR%20%2B%20%28%28gamma_4-1%29%2F4%29*SR%5E2%20for%20tail%20risk%7C4.%20Enforce%20DSR%20%3E%3D%200.95%3A%3AReject%20strategies%20with%20DSR%20%3C%2095%25%20or%20N%20%3C%20MinTRL%3B%20verify%20out-of-sample&metrics=NAIVE%201-TRIAL%20CONF%3A%3A99.997%25%3A%3AFalse%20sense%20of%20security%20on%20SR%3D1.80%7C200-TRIAL%20LUCKY%20HURDLE%3A%3A1.24%20Ann%20SR%3A%3AExpected%20max%20Sharpe%20of%20pure%20noise!%7CTRUE%20DEFLATED%20DSR%3A%3A80.8%25%20%28FAIL%29%3A%3ABelow%2095%25%20deployment%20requirement)

## Expected Maximum Sharpe Ratio of Pure Noise Across $K$ Trials ($T = 5$ Years IID)

Under $T = 5$ years of zero-skill Gaussian white noise (annualized trial standard deviation $\sigma_{\{\text{SR}\}} = 1/\sqrt{5} = 0.4472$), watch how rapidly the **Expected Maximum Backtest Sharpe Ratio $\mathbb{E}[\max_k \widehat{\text{SR}}_k]$** climbs purely by luck as a researcher tests more variations:

## Extreme-Value Noise Floor: Expected Maximum Sharpe Across K Backtest Trials

Quantify the exact annualized Sharpe Ratio generated purely by Gaussian luck across parameter grid sweeps:

| Independent Strategy Trials Tested ($K$) | Extreme-Value Multiplier ($Z_{\max}$) | Expected Max Noise Sharpe ($T = 5\text{y}$) | Expected Max Noise Sharpe ($T = 2\text{y}$) | Minimum 5y Backtest SR Needed for `DSR = 95%` (Normal) |
| :--- | :--- | :--- | :--- | :--- |
| **$K = 1$** (True Pre-Registered Hypothesis) | $0.00\,\sigma$ | **$0.00$** | **$0.00$** | **$0.74$** |
| **$K = 10$** (Small Parameter Grid) | $1.54\,\sigma$ | **$0.69$** | **$1.09$** | **$1.45$** |
| **$K = 100$** (Feature Selection Sweep) | $2.51\,\sigma$ | **$1.12$** | **$1.77$** | **$1.91$** |
| **$K = 1{,}000$** (ML / AutoML Grid Search) | $3.24\,\sigma$ | **$1.45$** | **$2.29$** | **$2.27$** |
| **$K = 10{,}000$** (Genetic / Deep Alpha Mining) | $3.85\,\sigma$ | **$1.72$** | **$2.72$** | **$2.58$** |

![Expected maximum annualized Sharpe ratio generated purely by random white noise over a 5-year backtest across 10 to 10000 trials](/api/figure?type=bars&title=Pure%20White-Noise%20Expected%20Max%20Sharpe%20%285-Year%20Backtest%2C%20True%20SR%20%3D%200%29&subtitle=How%20multiple%20testing%20manufactures%20%22institutional-looking%22%20Sharpe%20ratios%20out%20of%20coin%20flips.&steps=K%20%3D%2010%20Trials%3A%3AExpected%20best%20noise%20backtest%20%3D%200.69%20Annualized%20Sharpe%7CK%20%3D%20100%20Trials%3A%3AExpected%20best%20noise%20backtest%20%3D%201.12%20Annualized%20Sharpe%7CK%20%3D%201%2C000%20Trials%3A%3AExpected%20best%20noise%20backtest%20%3D%201.45%20Annualized%20Sharpe%7CK%20%3D%2010%2C000%20Trials%3A%3AExpected%20best%20noise%20backtest%20%3D%201.72%20Annualized%20Sharpe%20%28Zero%20true%20edge!%29&metrics=2-YR%20NOISE%20%40%20K%3D1%2C000%3A%3A2.29%20Sharpe!%3A%3AShort%202y%20crypto%2FFX%20backtests%20are%20traps%7CPEARSON%20KURTOSIS%20RULE%3A%3ANormal%20%3D%203.0%3A%3AUse%20gamma_4%20-%201%20%3D%202%2C%20NOT%20excess%20kurt!%7CEFFECTIVE%20TRIALS%20K_eff%3A%3AONC%20Clustering%3A%3ACorrelated%20MA%2850%29%2FMA%2852%29%20count%20%3C%202%7CMIN%20TRACK%20RECORD%3A%3AMinTRL%20Formula%3A%3AMonths%20needed%20at%2095%25%20confidence)

## Estimating Effective Independent Trials ($K_{\text{eff}}$) When Sweep Grids Are Correlated

What if those $200$ backtested variations were moving averages from `MA(10)` to `MA(209)` whose return streams are $92\%$ correlated with one another? Treating $K = 200$ as $200$ *independent* trials would over-penalize the strategy! Lopez de Prado & Lewis (2019) resolve correlated grids by clustering the $K \times K$ trial correlation matrix (or using average pairwise correlation $\bar{\rho}$ via $K_{\text{eff}} \approx 1 + (K - 1)(1 - \bar{\rho})$): if $200$ sweeps have $\bar{\rho} = 0.85$, effective independent trials equal $K_{\text{eff}} = 1 + 199(0.15) \approx 31$, giving an exact, fair selection-bias hurdle!

## Combining DSR with Lo Autocorrelation and Out-of-Sample Paper Verification

Before allocating live margin on [FIX API](/software/fix-api) or [MT5](/software/mt5) infrastructure, chain **both** [Lo (2002) serial-correlation unsmoothing](/guides/sharpe-ratio-iid-annualization-lo-2002-autocorrelation-haircut), [Information Ratio breadth checks](/guides/information-ratio-active-return-tracking-error-fundamental-law), AND **Bailey-Lopez de Prado DSR**, verify skewness/kurtosis tail exposure via [Cornish-Fisher Modified VaR](/entities/cornish-fisher-var-cornish-fisher-skew-kurtosis-modified-parametric-var), and review our transparent testing protocols in [methodology](/methodology) and [verify](/verify).

## Frequently asked questions

### What is the #1 coding bug engineers make when implementing PSR and DSR in Python?

Passing pandas/scipy default Fisher EXCESS kurtosis (where Normal = 0) into ((gamma_4 - 1) / 4) * SR^2! Bailey & Lopez de Prado wrote (gamma_4 - 1)/4 using PEARSON non-excess kurtosis (where Normal = 3, so (3-1)/4 = 0.5). If using scipy excess kurtosis k_excess, you MUST write ((k_excess + 2) / 4) * SR^2!

### Can I plug annualized Sharpe Ratios directly into the PSR and DSR formulas?

No! Mertens’ higher-moment variance formula V = 1 - gamma_3*SR + ((gamma_4 - 1)/4)*SR^2 is strictly scale-specific to the observation frequency at which N, skewness gamma_3, and kurtosis gamma_4 are measured. Convert annualized SR back to per-period SR_hat = SR_ann / sqrt(q) first!

### How do I calculate Minimum Track Record Length (MinTRL) for a live emerging manager?

Given monthly SR_hat = 0.35 (annualized 1.21) against benchmark SR* = 0, skew = -0.8, and Pearson kurtosis = 5.0, V_mertens = 1 + 0.28 + 0.1225 = 1.4025. At 95% confidence (z = 1.6449), MinTRL = 1 + 1.4025 * (1.6449 / 0.35)^2 = 31.99 months (~2.67 years minimum track record).

### Does Combinatorial Purged Cross-Validation (CPCV) replace the Deflated Sharpe Ratio?

They are complementary! DSR tests whether the best point-estimate Sharpe beats the multiple-testing noise floor analytically, whereas CPCV generates an empirical distribution of out-of-sample paths to measure the Probability of Backtest Overfitting (PBO) directly.


## Source URLs

- https://jpm.pm-research.com/content/40/5/94
- https://alo.mit.edu/wp-content/uploads/2017/06/The-Statistics-of-Sharpe-Ratios.pdf
- https://www.cfainstitute.org/insights/professional-learning/refresher-readings/2025/quantitative-investment-analysis
- https://www.bis.org/publ/work906.htm

## Citation

- APA: Elena Rostova, FRM, CQF. (2026). Probabilistic & Deflated Sharpe Ratio (DSR): Backtest Multiple-Testing Math. QuantCalc Tools. https://quantcalctools.com/guides/probabilistic-deflated-sharpe-ratio-multiple-testing-backtest-bias
- Canonical article: [Probabilistic & Deflated Sharpe Ratio (DSR): Backtest Multiple-Testing Math](https://quantcalctools.com/guides/probabilistic-deflated-sharpe-ratio-multiple-testing-backtest-bias)
- Site research index: [QuantCalc Tools](https://quantcalctools.com/llms-full.txt)
