Everything a client needs on one page: what you hold, why you hold it, what it has actually done, and the trades that get you there — rebuilt the moment you change profile, horizon or regime.
€10,000 grown over 10 years across 160 paths drawn from the historical volatility of this mix. Illustrative, not a forecast — the median path is not an expectation.
The Explainer is off. Turn it on and every number on this screen — the meter, the sleeves, the projected range, the drawdown halt and the record — is written out in ordinary language, in euro, with the downside stated as plainly as the upside.
—
Model output on an illustrative profile over one sample period, costs included. Past behaviour is never a promise about the future, and nothing here is personal investment advice.
Model output for an illustrative profile — not a recommendation for any individual investor.
Gold: liquidity index. Green: regime score. Dashed verticals: events that moved a component by more than one standard deviation.
Put in what you actually hold and what it is worth. Everything below is then priced in euro, not percentages.
Deltas compare the model with what you entered; the live model caps changes at 5 points per sleeve per week, so a large gap is closed in steps. Age it three months drifts your sleeves with the last quarter of the sample so you can see what an un-rebalanced portfolio looks like.
Not a chart of the gap — the list of trades that closes it, in the order you would place them, priced in euro.
Costs are an expected-slippage estimate that grows with the size of the trade; a single trade over the 20 bps cap is skipped and carried to next week. Nothing here is an instruction to trade — it is arithmetic on the weights you entered.
Keep this portfolio in this browser and the portal opens on it, with the drift and the guard headroom already measured.
Stored in this browser only — it never leaves this device and clearing your browser data removes it. Accounts open soon.
The honest part of a projection is the bottom of it. Here is the loss you would have to sit through.
Four limits sit above every weighting on the previous page — they are the reason a client can close the laptop and go to sleep.
Every tilt away from the strategic mix is sized at a quarter of the mathematically optimal bet. Full-size bets are optimal only if the estimates are perfect; they never are. A quarter gives up a little of the good years and removes most of the paths that end badly.
If the portfolio is 20 % below its own high-water mark, the model stops adding risk and moves the sleeves to their defensive floor until the reading recovers. Recovery comes from patience and time, never from doubling the position.
No single company carries more than 7.5 % of the portfolio. Sleeves are baskets, never one name — and the Bitcoin sleeve keeps its own hard cap of 15 %, never added into contracting liquidity.
A rebalancing trade is skipped when expected slippage passes 20 basis points. Drifting one week longer costs less than paying a wide spread, and the 5-point weekly step limit means nothing has to be done in a hurry.
Guards are model settings, not promises about markets. Everything here is educational and illustrative — never personal investment advice.
Every regime call this model has made over the sample period — the ones it got wrong printed as large as the ones it got right — and what the whole thing did against a static 60/40.
€100 invested at the start of the sample, both mixes rebalanced quarterly with costs taken out. The shaded stretches are the ones where holding this model was the worse decision.
A wealth model that never shows a lagging stretch is showing you a selected picture. These are measured from the relative curve: each one starts at a high-water mark of model-over-benchmark and ends when that mark is regained.
Each row is one flip of the label, what the sleeves did in the three months after it, and whether the call was right.
Scoring rule, fixed before the numbers were read: a risk-off call is right if equities returned less than cash over the three months after the flip; a risk-on call is right if equities returned more. A neutral reading is not a directional call, so it is not scored. The call still running is shown but not counted.
How far below its own high-water mark each mix went. This, not the return line, is what the guards are for.
Tell it who you are rather than which risk profile you want, and the model reshapes around you — then says, in plain words and in euro, what changed, why, and what the choice costs you.
The Explainer is off. Turn it on and every number on this screen — the meter, the sleeves, the projected range, the drawdown halt and the record — is written out in ordinary language, in euro, with the downside stated as plainly as the upside.
Nothing here leaves this browser. The answers only change which numbers are put in front of you and which of the four guards is argued hardest.
Pick one. The model sets the profile and the horizon, argues the guard that matters most to you, and prints what it changed, what it now holds and what it gives up.
A persona is an illustrative profile, not a judgement about you and never personal investment advice. Sizing stays at quarter-Kelly (0.25) inside the 7.5 % per-company ceiling, the −20 % drawdown halt and the 20 bps execution cap in every one of them.
One editorial letter a week that explains what changed in the data and what the model did about it — in plain language, with the series named.
Two-year yields led the print by eleven days. What the curve knew, and what the model did with it.
Official-sector demand has changed who sets the gold price. Why a 10–20 % sleeve is not a bet.
Four liquidity cycles, four Bitcoin cycles, one uncomfortable overlap. How we size a volatile sleeve.
Hysteresis in the meter, explained with the 2019 and 2023 false starts.
A room of long-horizon investors reading the same brief — free to read, and moderated so it stays that way.
Accounts open soon. One Mast account will keep your profile, horizon and holdings between visits.
Say what you want once. The banker then works on its own — watching the drift, checking the guards, digesting the briefs — and every single thing it does is written down here, with the reason and the number it produced.
Each line is what it did, why it did it, and the number that came out — computed from the same functions the panels on the other views use, so the log and the screen can never disagree.
Model output on an illustrative profile, never personal investment advice. The banker is unleveraged, it rebalances weekly at most, and it sizes every tilt at quarter-Kelly (0.25) inside the 7.5 % per-company ceiling, the −20 % drawdown halt and the 20 bps execution cap.
Model output for an illustrative profile — never personal investment advice.
Demo · simulated data · Part of the Mast family