AAPL ▲ +1.24% MSFT ▲ +0.83% NVDA ▲ +2.41% TLT ▼ −0.37% GLD ▲ +0.62% SPY ▲ +0.91% QQQ ▲ +1.15% IWM ▼ −0.18% BRK.B ▲ +0.44% TSLA ▼ −1.72% DBC ▲ +0.29% IEF ▼ −0.11% AAPL ▲ +1.24% MSFT ▲ +0.83% NVDA ▲ +2.41% TLT ▼ −0.37% GLD ▲ +0.62% SPY ▲ +0.91% QQQ ▲ +1.15% IWM ▼ −0.18% BRK.B ▲ +0.44% TSLA ▼ −1.72% DBC ▲ +0.29% IEF ▼ −0.11%
Quantitative Portfolio Intelligence

Find the
optimal
portfolio.

Given any set of stocks, PortOpt finds the allocation that maximises return per unit of risk — then proves it holds up in real market conditions.

5+
Algorithms
20yr
Data history
Backtests
Optimized Portfolio · 5yr backtest
+247%
+2.31 Sharpe ratio
vs S&P 500
+139%
1.08 Sharpe
Optimized Portfolio
SPY Benchmark
2.3× Average Sharpe ratio improvement over equal-weight
5 Optimization algorithms: Markowitz, HRP, CVaR, RP, Robust
3 Crisis period stress tests: 2008, COVID, 2022 rate shock
95% Bootstrap Sharpe confidence intervals on all strategies

Three questions
worth answering

Instead of staring at an empty text box, start with one of these pre-built scenarios. Each one demonstrates a real portfolio decision and shows you the math behind it.

01 / 03

Should I just buy NVDA and hold?

Compare a concentrated NVDA bet against a Markowitz-optimized tech portfolio. See where concentration risk shows up in the drawdowns.

NVDAAAPLMSFTAMZNQQQ
Max Drawdown (NVDA-only) −65.8%
02 / 03

How would this hold up in 2008?

Run HRP optimization on an All Weather basket, then stress-test it against GFC, COVID, and the 2022 rate shock back to back.

SPYTLTIEFGLDDBC
HRP Sharpe vs Equal Weight +0.41
03 / 03

Is SPY actually hard to beat?

Find the max-Sharpe frontier portfolio from the S&P 500 universe, backtest it walk-forward, and see if it survives transaction costs.

SPYBRK.BJPMJNJTLT
Walk-forward alpha vs SPY +1.8% ann.

From tickers
to conviction.

Four steps. No finance PhD required. The math is under the hood — you get the answer.

01 📥
Enter your universe

Type any stock tickers. PortOpt pulls 5 years of adjusted prices automatically via Yahoo Finance.

02 ⚙️
Choose an algorithm

Pick from Markowitz mean-variance, HRP, CVaR minimization, risk parity, or robust optimization.

03 📈
See the frontier

The efficient frontier shows every risk/return tradeoff. The star marks the max-Sharpe portfolio.

04 🔬
Stress-test it

Run walk-forward backtests, Monte Carlo simulations, bootstrap Sharpe CIs, and crisis period replays.

The optimizer,
in action.

The efficient frontier maps every possible portfolio in risk/return space. Each dot is a different allocation — the ★ is the one with the best Sharpe ratio.

Efficient Frontier — AAPL · MSFT · TLT · GLD · SPY
Markowitz MVO
Volatility (Annual) Return (Annual) GLD NVDA AAPL TLT SPY ★ Max Sharpe SR = 2.31
Max Sharpe Weights
★ Optimal
AAPL
28.4%
GLD
24.1%
TLT
21.8%
MSFT
15.9%
SPY
9.8%
2.31Sharpe
18.4%Ann. Return
7.9%Ann. Vol

Five algorithms.
One decision.

📐
Markowitz MVO

Classic mean-variance optimization with covariance shrinkage. Traces the full efficient frontier.

Classic · 1952
🌳
HRP

Hierarchical Risk Parity uses clustering to build diversified portfolios that don't invert the covariance matrix.

Robust · 2016
⚖️
Risk Parity

Allocates capital so each asset contributes equally to total portfolio risk. Naturally defensive.

Balanced · 2005
🛡️
CVaR Min

Minimizes Conditional Value-at-Risk — the expected loss in the worst 5% of scenarios.

Tail Risk · 2000
🔩
Robust MVO

Ellipsoidal uncertainty sets on expected returns. Finds portfolios that hold up even when your return estimates are wrong.

Conservative · 2004

Built for
the curious.

PortOpt is an open-source quantitative portfolio research tool. It's not financial advice — it's a rigorous framework for asking better questions about risk and return.