How it works
Our methodology
No black boxes. Here is exactly how Slate models your financial future — the data we use, the assumptions we make, and the limitations you should know.
The five-step process
Portfolio mapping
Your holdings are classified into asset classes (UK equities, global equities, bonds, gilts, property, cash, alternatives) using ticker-level lookups. Each class carries its own return, volatility, and correlation assumptions drawn from long-run historical data.
Monte Carlo simulation
We generate 1,000 independent market paths by sampling from a multivariate normal distribution calibrated to historical asset-class parameters. Each path applies compounding, contributions, withdrawals, and inflation year by year to produce a full trajectory for your portfolio.
Goal probability
For each simulated path, we check whether your portfolio can fund every planned expense — retirement spending, lump sums, property purchases — through the end of your time horizon. The percentage of paths that succeed is your goal probability.
Stress testing
We apply deterministic shocks modelled on historical crises: the magnitude, volatility, and duration of each event are applied to your current allocation. Custom scenarios let you define your own drop, recovery time, and volatility multiplier.
Goal-aligned allocation
The engine searches across thousands of allocation combinations to find the weights that maximise your goal probability, subject to constraints you set (minimum cash, maximum equity, income floor). This is numerical optimisation, not a risk-questionnaire lookup.
Key assumptions and limitations
Return assumptions
Based on 50+ years of UK and global asset-class data. Equities: 7–9% nominal; bonds: 3–5%; gilts: 3–4%; cash: 2–3%. Updated periodically. All projections are in nominal terms with a separate inflation adjustment.
Inflation
Default assumption: 2.5% annual CPI, based on the Bank of England's long-run target. You can override this in your profile settings.
Correlations
Asset-class correlations are estimated from historical monthly returns. In stress scenarios, we increase equity-equity correlations to reflect the well-documented rise in co-movement during market crises.
Limitations
Past performance is not a reliable indicator of future returns. Models assume normal-distribution returns, which understate the probability of extreme events. Tax is not modelled. Projections are illustrative, not predictions.