Vose SoftwareModelRisk example · Oil and gas
Exploration Portfolio with Shared Play Risk
Which wells should we drill next year? Four drilling programmes compared on value and on the chance of a dry year.
From the model’s saved run
Which programme?
Expected cost, expected value (EMV), chance of a dry year and chance of losing money, from the simulation.
| Programme | Cost US$m | EMV US$m | Finds nothing chance | Loses money chance | Acceptable? |
|---|---|---|---|---|---|
| 1 Proven play focus | 202.5 | 31.4 | 13.3% | 51.0% | Yes |
| 2 Balanced | 245.8 | 58.4 | 17.0% | 53.9% | Yes |
| 3 Frontier upside | 237.2 | 69.0 | 28.7% | 63.8% | No: risk |
| 4 Your programme | 263.8 | 51.5 | 23.5% | 56.5% | No: risk |
| Recommended programme | 2 Balanced |
The highest exact EMV among the programmes within the budget and the risk limit (Programmes sheet).
S-curve (cumulative probability) of 1 Proven play focus, 2 Balanced, 3 Frontier upside, 4 Your programme from the saved simulation.
The business question
The exploration team has eight drill-ready prospects in three plays and a budget for about five wells next year. Prospects in the same play share the play's risk: if the play does not work — no charge from the source rock, say, or no regional seal — every prospect in it is dry, however good its trap.
A proven play almost always works, so its prospects succeed or fail on local risk. A frontier play may not work at all, but its prospects are large. Four drilling programmes are on the table: concentrate on the proven play, balance across the plays, chase the frontier upside, or the team's own mix.
The model evaluates all four on the same simulated outcomes, so the differences between them come from the programmes, not from the luck of the draw. It reports each programme's expected value, its chance of finding nothing at all and its chance of losing money, and it recommends the most valuable programme that stays within the budget and the company's appetite for a dry year.
It also shows how much a simple calculation that treats the prospects as independent understates the chance of a dry year.
- Who it is for
- Exploration managers, portfolio and planning teams, and geoscientists who risk prospects.
- Main outputs
- For each programme: expected value (EMV), chance of finding nothing, chance of losing money and the range of value · the programme recommended within the budget and the risk limit · how shared play risk changes the chance of a dry year · discoveries and value drivers of one programme in detail.
How the model works
Four sheets after this one, read from left to right. Each one feeds the next.
Inputs sheet
The exploration budget, the largest acceptable chance that a programme finds nothing, the programme to analyse in detail, and for each play the chance that it works at all.
Prospects sheet
Each prospect's chance of success if its play works, its well cost (PERT) and its value if it finds oil (Lognormal from P90 and P10). The 1s and 0s on the right define which prospects each programme drills. In every iteration one Bernoulli trial per play decides whether the play works and one per prospect decides the local outcome.
A prospect is a discovery only if both succeed.
Programmes sheet
For each programme: its expected cost against the budget, its expected value and its chance of finding nothing, calculated exactly from the inputs. And, for this iteration, its discoveries, its cost and its value. All four programmes use the same simulated outcomes.
Results sheet
Expected value, chance of finding nothing and chance of losing money for each programme; the recommended programme; the effect of shared play risk. And the discoveries and value drivers of the programme chosen for detail.
ModelRisk techniques
Find similar models
- Product
- Type
- Application
- Uses
The results in detail
Shared play risk and the chance of a dry year
Simulated, calculated exactly, and as the independent shortcut would have it.
| Programme | Simulated chance | Exact, shared chance | If independent chance | Understated by points |
|---|---|---|---|---|
| 1 Proven play focus | 13.3% | 13.4% | 11.3% | +2.1% |
| 2 Balanced | 17.0% | 17.1% | 15.4% | +1.7% |
| 3 Frontier upside | 28.7% | 28.2% | 25.7% | +2.4% |
| 4 Your programme | 23.5% | 23.4% | 18.1% | +5.3% |
The programme analysed in detail
The programme chosen on the Inputs sheet (simulate again after changing it).
| Measure | Unit | Value |
|---|---|---|
| Programme | 2 Balanced | |
| Expected value, EMV | US$m | 58.4 |
| Programme NPV, 10th percentile | US$m | −243.1 |
| Programme NPV, 90th percentile | US$m | 464.0 |
| Mean number of discoveries | wells | 1.54 |
| Number of discoveries | Probability |
|---|---|
| No discovery | 17.0% |
| One discovery | 34.6% |
| Two discoveries | 30.3% |
| Three discoveries | 14.1% |
| Four or more | 4.0% |
| Total | 100.0% |
What drives the value of the programme in detail
Lowest and highest mean programme NPV over ten equal groups of each input (VoseSimSensitivityMean), drawn from the overall mean.
Tornado chart: the inputs ranked by how far the mean output moves across each input’s range.
The ranking as a table 13 inputs
| Input (ranked by influence) | Lowest mean | Highest mean | Rank |
|---|---|---|---|
| Play C works | −35.8 | 239.9 | 1 |
| Play B works | −64.8 | 123.0 | 2 |
| Ginkgo value if discovered | 1.9 | 175.7 | 3 |
| Ginkgo succeeds | −35.7 | 134.9 | 4 |
| Fir succeeds | −1.8 | 148.7 | 5 |
| Fir value if discovered | 17.9 | 149.0 | 6 |
| Dogwood succeeds | −2.7 | 120.1 | 7 |
| Cedar succeeds | 16.1 | 135.6 | 8 |
| Dogwood value if discovered | 25.8 | 138.4 | 9 |
| Alder succeeds | 12.7 | 115.7 | 10 |
| Cedar value if discovered | 27.1 | 126.9 | 11 |
| Alder value if discovered | 31.5 | 102.6 | 12 |
| Fir well cost | 34.5 | 81.7 | 13 |
How to read these results
- EMV ranks the programmes by value, but a programme can have the highest EMV and still find nothing in one year out of four. The risk limit on the Inputs sheet turns the company's appetite for a dry year into a rule, and the recommendation follows it.
- The exact and simulated chances of finding nothing agree to within the simulation's precision (about ±1 percentage point with 10,000 samples): a check that the model is wired correctly. The verdicts and the recommendation use the exact figures, so they do not move with the sampling.
- Treating prospects as independent understates the chance of a dry year, most for programmes that drill several prospects in one risky play: if the play fails, all of them fail together.
- Expected values add up whatever the dependence, so shared risk does not change a programme's EMV. It changes its spread and its chance of a dry year.
- All four programmes are evaluated on the same simulated outcomes, so the differences between them are not sampling noise.
See it move. Open the workbook, change an input on the Inputs sheet and run the simulation: every table and chart on this page updates in Excel.
Inputs and assumptions
Every assumption in the model, with its uncertainty and where it comes from. The figures are illustrative.
Budget and risk appetite 3 inputs
The constraints a programme must meet, and which programme the Results sheet shows in detail.
| Assumption | Unit | Value |
|---|---|---|
| Exploration budgetExpected cost of all the wells in a programme may not exceed this. | US$m | 270.0 |
| Largest acceptable chance of finding nothingA programme whose chance of a dry year is higher is not recommended, whatever its value. | probability | 20% |
| Programme analysed in detail1 Proven play focus, 2 Balanced, 3 Frontier upside, 4 Your programme. Simulate again after a change. | 1 – 4 | 2 |
Plays 3 inputs
The chance that each play works at all: source rock charge, regional reservoir and seal. Shared by every prospect in the play.
| Assumption | Unit | Value |
|---|---|---|
| Chance that play A worksUpper Jurassic sandstones (proven). Six discoveries in the play to date: the play works; the risk left is local. | probability | 95% |
| Chance that play B worksLower Cretaceous turbidites (emerging). One discovery so far; whether the reservoir extends away from it is uncertain. | probability | 65% |
| Chance that play C worksPre-salt carbonates (frontier). No wells yet; charge and seal inferred from analogues on the conjugate margin. | probability | 35% |
Units and conventions
- Money
- US$ million (US$m), present values at the time of the drilling decision.
- Value if discovered
- NPV of developing a discovery, after development costs and tax, before the cost of the exploration well.
- P90 / P10
- Petroleum convention for prospect values: P90 is the low estimate (90% chance of being exceeded), P10 the high one.
- EMV
- Expected monetary value: the average programme NPV over all simulated outcomes, dry holes included.
Key assumptions and limitations
- Dependence
- Prospects in the same play share the play's risk. Once the play works, their local risks are independent. The plays are independent of each other.
- Values
- A discovery's value does not depend on other discoveries: no shared facilities and no capacity limits.
- Costs
- A well costs the same whether it finds oil or not; costs are independent of each other.
- Data
- Prospects, chances and values are invented for this example.
How to use it
- Review the plays on the Inputs sheet and the prospect inventory on the Prospects sheet. Change the 1s and 0s to design your own programme (column 4).
- Set the budget and the acceptable chance of a dry programme on the Inputs sheet.
- Run a simulation from the ModelRisk ribbon (10,000 samples with a fixed seed are set) and read the recommendation on the Results sheet.
- To see another programme in detail, change its number on the Inputs sheet and simulate again.
The tornado chart fills in after a simulation with ModelRisk’s classic engine.
ModelRisk functions used
- VoseBernoulli
- Whether each play works and whether each prospect succeeds locally: one trial per play and one per prospect.
- VosePERT, VoseLognormalAlt
- Well cost from low, most likely and high; value if discovered from its P90 and P10.
- VoseInput, VoseOutput
- Name the uncertain quantities and the programme outputs for the Results Viewer and sensitivity analysis.
- VoseSimMean, VoseSimProbability, VoseSimPercentile
- Statistics of the simulation for the tables and charts on the Results sheet.
- VoseSimSensitivityMean
- The tornado chart: the thirteen VoseInputs that move the mean programme NPV most across their range, ranked.
Further reading
- Sensitivity analysis in Excel: a tornado chart step by step (Vose guide)
- Rose, P. R. (2001). Risk Analysis and Management of Petroleum Exploration Ventures. AAPG Methods in Exploration 12.
- Vose, D. (2008). Risk Analysis: A Quantitative Guide, 3rd edition. Wiley — chapters on correlation and dependency.
Download the workbook and run it yourself. It opens in Excel with ModelRisk. The free trial runs every example model.
Related example models
- ModelRisk★★★★★Field Development Concept SelectionTieback, FPSO, or start small and expand?Three 25-year development cash-flow models on common random numbers, with a decision rule inside the simulation.
- ModelRiskOptimizer★★★★★Exploration Portfolio Chosen by the OptimizerWhich wells should we drill next year?The programme with the highest expected value within the budget and the risk limit, chosen by the Optimizer.
- ModelChoice★★★★★Mature Oil Field: Enhanced Oil Recovery DecisionFor a mature oil field: CO₂ injection, water alternating gas, a polymer pilot, primary recovery, or decommissioning?Three decisions and 11 chance events. Best first decision: Continue Primary Recovery, expected value $48.0m.
An illustrative example: the figures are invented for teaching and do not describe a real company, project or asset. Last updated 9 October 2026.