HOSTUVO LAB
HOSTuvo PRE-MOVE: event warnings, direction and honest validation
Understand HOSTuvo PRE-MOVE, its 5–90 minute diagnostic horizons and the difference between a research prediction and an authorised entry.
PRE-MOVE is the name of an early-warning layer in HOSTuvo. Its purpose is to highlight changes worth investigating, not to guarantee that the market will move first in the predicted direction. The product separates broad discovery, short-horizon timing and the primary accumulation-swing assessment.
Keep detection and direction as separate questions
A model can assess whether a meaningful move is likely and still have insufficient evidence about direction. It can also miss a move or generate a false alarm. The research workflow includes 5, 15, 30 and 90-minute horizons, but a horizon label does not certify that every component is production-ready. Read the model maturity and data status rather than assuming all displayed analytics have the same role.
A research score is not a trading instruction
SHADOW means an output is being evaluated without automatic production influence. A diagnostic BUY or UP in such a panel must not be mistaken for an authorised entry in the main decision panel. A confidence-like number is not a verified win rate unless its calibration and evaluation are demonstrated. In particular, a short-term observation does not replace the separate 4h/1d structure used by the main swing assessment.
Evaluate forecasts only after their outcome window
A forecast record should be saved before the outcome becomes known and evaluated only after the relevant window has matured. Precision asks how many predicted events occurred; recall asks how many actual events were detected. Direction accuracy needs its own clearly defined subset and sample count. None of those metrics alone reports profit after costs. Repeated observations of the same move should not be mistaken for independent successful trades.
Promote a model because of evidence, not activity
A training job finishing successfully shows that a computation ran; it does not show that a strategy improved. Compare candidates on data not used for fitting, document acceptance and rejection, and keep a rollback path. HOSTuvo currently uses research-only learning and separate production checks. The interface should show what has actually been admitted to production. No part of this description is an assurance of future returns or automatic real-money execution.
Keep the prediction contract explicit
An event warning asks whether a defined amount of movement may occur within a horizon. A direction view asks which side is favoured under its own target definition. A trading decision additionally needs an executable price, costs and an invalidation condition. These outputs should not be collapsed into one percentage. A strong direction score can coexist with a low expected movement or an unavailable entry. In HOSTuvo, read the horizon, evidence status and output semantics together; a diagnostic or SHADOW result must not be mistaken for an executed trade.
Hypothetical correct direction with an unusable entry
A 90-minute view may favour an upward path while expected remaining movement is 0.25% and estimated round-trip cost is 0.30%. Directional correctness would not establish economic usefulness. In another example, a target is reached before the scanner publishes its result: the chart later looks supportive, but the warning was not early. A proper record keeps the actual emission timestamp and the price reference separate from the source candle's timestamp. The examples are explanatory and do not represent reported HOSTuvo returns.
Inspect forward evidence rather than activity counters
Look for predictions frozen before their outcomes, a fully defined horizon and an explicit treatment of missing or ambiguous data. Adjacent forecasts may share most of the same price path, so a large count can contain less independent evidence than it suggests. Compare the governed direction with its recorded raw baseline and report uncertainty alongside the result. Model updates, worker cycles and the number of analysed markets describe activity; they do not establish improved accuracy. Live influence should remain distinguishable from experiments collecting evidence.
Sources and example scope
The references explain the underlying definitions and mechanisms. Numerical scenarios are hypothetical teaching examples, not forecasts or reported HOSTuvo returns.
IN HOSTuvo
See this context in the live terminal.
Check the source, horizon, maturity and missing confirmations. A scanner can identify a candidate while the main decision correctly remains unconfirmed.
