The Bullish, Neutral, or Bearish badge hides a fair amount of math. VAMS blends four momentum windows, realized volatility, five trend checks, and a confirmation delay.
All of that machinery decides how much of an asset’s top-down target should be active. VAMS produces an exposure multiplier, never a price target.
Start with four momentum horizons
The raw momentum calculation uses log returns over four windows:
| Horizon | Weight |
|---|---|
| 1 month | 20% |
| 3 months | 35% |
| 6 months | 30% |
| 12 months | 15% |
The three-month window receives the largest weight, but no single horizon controls the result. A one-month bounce can help without erasing a weak six-month trend. A strong twelve-month return can remain visible without overpowering a recent breakdown.
Research on time-series momentum and a century of trend following provides the broader evidence for using trends across more than one horizon.
Adjust momentum for volatility
The weighted momentum score is divided by 63-day annualized realized volatility. A 10% move in a calm asset is treated differently from a 10% move in a highly volatile asset.
The adjustment is especially noticeable with bitcoin. Fidelity Digital Assets’ discussion of bitcoin volatility shows why a raw return alone says little about portfolio risk.
The current raw score is then compared with roughly three years of its own history. The resulting z-score measures how unusual the reading is for that asset. A z-score above +0.25 creates bullish pressure. A z-score below -0.25 creates bearish pressure. The middle zone is neutral.
Check five trend votes
VAMS also evaluates five yes-or-no conditions:
- price is above the 50-day moving average;
- price is above the 200-day moving average;
- the 50-day average is above the 200-day average;
- the three-month return is positive;
- the 50-day average is higher than it was ten trading days earlier.
These votes overlap. Price above the 200-day average and the 50-day average above the 200-day average often agree. The model does not describe them as five independent economic facts. They are a compact trend checklist.
The exact state rules
VAMS becomes Bullish when either condition is true:
- z-score is above +0.25 and at least three trend votes are bullish; or
- price is above the 200-day average, the three-month return is positive, and at least four votes are bullish.
VAMS becomes Bearish when either condition is true:
- z-score is below -0.25 and no more than two trend votes are bullish; or
- price is below the 200-day average, the three-month return is negative, and no more than two votes are bullish.
If neither side triggers, the state is Neutral.
Bullish applies 100% of the top-down target. Neutral applies 50%. Bearish applies 0%. RSI 14 is displayed on the dashboard, but it does not affect the VAMS state, multiplier, or allocation.
Confirmation slows the model down
A new VAMS state must persist for two consecutive closes before it becomes confirmed. Allocation confirmation is tracked independently for each sleeve. A lower sleeve target requires two closes and a higher sleeve target requires five, so a change in one asset does not reset another asset’s count. The model simulation executes a confirmed sleeve change on the next trading session.
That delay can look late around a sharp turn. It is the price of refusing to trade every one-day reversal. Quantpedia’s asset-class trend-following summary covers the general strategy, while Vanguard’s rebalancing guidance explains the practical cost of changing a portfolio too often.
Reading a VAMS card
Read VAMS in layers: z-score, five votes, triggered rule, confirmation status, then multiplier.
The badge gives the answer. The calculation details show whether it rests on broad trend confirmation or one unusually strong momentum reading.