Value at Risk is the loss a normal bad day should not exceed, at a chosen confidence level: the line the market should not cross more than, say, one day in twenty. Sentinel draws that line three ways, from volatility (the bell curve), from your holdings’ actual history, and from thousands of simulated days, then shows expected shortfall: how bad the days beyond the line tend to be.
Three estimates of the same loss, each with its own verdict. Where they disagree is the part worth reading.
Volatility (parametric)
History (historical)
Simulation (Monte Carlo)
Expected shortfall
Diversification benefit
Backtest verdict
Stress
What a past crisis would have cost today’s holdings.
A stress test applies a shock to your portfolio and reports the damage, the way engineers load a bridge before the traffic does. Sentinel ships seven of them, from a ten percent index fall to the 2008 and 2020 crashes, and replays each one on the holdings you own. The Stress page is on the Analyst plan.
The scenario library, each past crisis ranked by the size of the market fall, mildest first.
2008 (Analyst)
COVID crash (Analyst)
Rate spike (Analyst)
Your own scenario (Analyst)
Per-holding override (Analyst)
Concurrent shocks (Analyst)
Regime
Which market you are in, and what it means for your numbers.
Markets move between calm and stressed states, and the same portfolio carries different risk in each. Sentinel reads the current regime from prices and cross-checks it against credit markets, where stress often shows first, then reports what changed.
The regime reading, with the volatility, yield-curve and credit signals behind it.
VIX
Yield curve
Credit spreads
Twelve months of history
Updated daily
Portfolio
What you paid, what it is worth, and how it compares.
Cost basis and profit or loss per holding, performance against the S and P 500, and the standard return measures (Sharpe, Sortino, drawdown), each with a one-line explanation of what it tells you.
How your current mix would have done over the past year against SPY, indexed to 100, with the risk and return metrics beneath.
Cost basis
Profit and loss
Sharpe and Sortino
Worst drawdown
Beta and alpha on Analyst
Concentration
Watchlist
Named lists, with the one thing other watchlists leave out.
Keep as many lists as you like. Each shows a correlation strip: a quick read of whether the names you are watching would all fall on the same bad day, which is the test of diversification.
A named list with a one-day loss estimate per name and the diversification strip above it.
Named lists
One-day loss per name
Correlation strip
52-week range
CSV export
Alerts and digest
Told when it matters, with the number it saw.
An alert fires when a holding grows past the share of your portfolio you set, or when the market turns jumpy, and each one records the number it saw at that moment. A weekly digest lands on Sunday.
Creating an alert. Fired alerts land in the list beneath it, each with the value it saw when it fired.
Price alerts
Position-size alerts
Volatility alerts
Observed value on every fire
Sunday digest
Brokerage sync
Read-only connections through regulated services. The broker login is entered with the broker, not with Sentinel.
Link accounts through Tink in the EU and UK or SnapTrade in the US and Canada. Sentinel receives positions and prices; it cannot trade, move money or see your brokerage password. Prefer not to connect? Add holdings by hand on any plan.
Connecting a broker. Read-only: Sentinel sees positions and cannot move money.
Turn on two-factor sign-in in Settings: a six-digit code from your phone on top of your password. Holdings are encrypted at rest and in transit, sessions expire when idle, and you can export or delete everything you have given us.
Two-factor setup in Settings, with the codes from your phone.
Two-factor codes
Encrypted at rest
Encrypted in transit
Idle session expiry
Export and delete
Start with Basic. Upgrade when you hit a feature you need.
