Monte Carlo simulation is a computational technique that uses random sampling from probability distributions to model the uncertainty inherent in oil and gas investment decisions. Rather than producing a single deterministic answer, Monte Carlo generates thousands to millions of possible outcomes, revealing the full range of risk and upside for a project. The method is named after the Monte Carlo Casino in Monaco and was formalized in the 1940s by scientists working on nuclear weapons research. Today, it is the standard tool for probabilistic reserves estimation and risk assessment in the petroleum industry, endorsed by the SPE/PRMS guidelines and used by virtually every major E&P company and financial institution.
How It Works
A Monte Carlo simulation follows a structured process of defining inputs, running iterations, and analyzing outputs:
- Input Distributions — Each uncertain variable is assigned a probability distribution instead of a single value. Common inputs include initial production rate (qi), decline parameters (Di, b-factor), oil/gas price, well cost, operating expenses, and recovery factor. Distributions may be normal, lognormal, triangular, or uniform depending on the data available.
- Random Sampling — The simulation randomly selects a value from each input distribution and calculates the output (e.g., EUR, NPV, or IRR) for that combination. This process is repeated 10,000 to 100,000 times (iterations), with each iteration representing one possible outcome.
- Output Distribution — The collected results form a probability distribution of the output variable. The P10 value (90th percentile — only a 10% chance of exceeding) represents the optimistic case, P50 (50th percentile) represents the most likely outcome, and P90 (10th percentile — 90% chance of exceeding) represents the conservative case.
- S-Curve — The cumulative probability distribution of the output is plotted as an S-curve (or ogive), providing a visual representation of the full range of outcomes and their probabilities.
- Correlation — Inputs often have dependencies (e.g., wells with higher qi tend to have higher EUR). Monte Carlo models can incorporate correlation coefficients between input variables to avoid unrealistic combinations.
Why It Matters
Deterministic economics — running a single case with best-guess inputs — hides the true risk of an investment. A well with a deterministic NPV of $2 million may have a Monte Carlo P90 NPV of negative $500,000, meaning there is a 10% chance of losing money. This information is critical for capital allocation, portfolio management, and risk budgeting. In acquisitions, the difference between P50 and P90 reserves determines the range of acceptable bid prices — a $500 million acquisition evaluated at P50 reserves may have a P90 value of only $350 million, representing $150 million of downside risk that must be priced into the transaction. Banks lending against reserves typically use P90 values, extending credit at 50 to 65% of PV-10 of proved reserves.
How Netora Handles Monte Carlo Simulation
Netora Upstream Platform includes a built-in Monte Carlo engine that runs probabilistic simulations across production, economics, and reserves. Users define input distributions for key variables, and the platform generates P10/P50/P90 outputs for EUR, NPV, IRR, and cash flow. Results are displayed as S-curves, tornado charts, and probability tables, giving decision-makers a complete risk profile for every well, project, and portfolio. Learn more about Netora Upstream Platform.