Glossary

Confidence tells you how much evidence supports the currently published classification. It is not a prediction score and it is not the probability that the regime is 'true'. A higher value means the current label is backed by more complete data and a clearer internal signal structure.

In the current backend, confidence_score is the geometric mean of data_quality_score and label_confidence_score: sqrt(data_quality_score × label_confidence_score). That means confidence only stays high when both inputs are strong. It should be read as evidence sufficiency for the present classification, not as forecast skill, expected return, or directional conviction.

This score asks a simpler question than full confidence: 'Do we have enough complete and recent data to evaluate the chain properly right now?' It is the data sufficiency side of confidence, before the model asks whether the regime itself is internally clear.

The backend computes data_quality_score from five weighted components: current_row_coverage (30%), recent_metric_coverage (20%), recent_density (20%), history_depth (15%), and freshness_asof (15%). The score is clipped to 0..1. This is about data completeness and freshness only; it does not yet judge whether the regime label is sharp or ambiguous.

How much historical depth is available for the current computation. More history usually makes baselines, percentiles, and unusualness estimates more trustworthy.

In the current backend this is capped at 1.0 once roughly 90 distinct days are available. The score is not trying to reward infinite history forever; it is trying to avoid giving full confidence to a regime that was inferred from a very short local sample.

This shows where today's value sits relative to roughly the last 90 days. For example, a value near 95 means the metric is higher than most days in that recent history; a value near 5 means it is lower than most days.

pct_90d is useful because it complements z_robust. z_robust measures standardized unusualness, while percentile gives a direct rank-based location in the recent distribution. Together they make it easier to see whether a metric is merely above average or genuinely near an edge of the recent sample.

Which high-level dimension the driver belongs to: demand, friction, or capacity. This tells you what kind of pressure the metric is describing.

The axis field is part of the published explanation payload and should be interpreted as the metric's role inside the scorecard/regime model, not just a cosmetic tag. It lets advanced users trace local signals back to the dimension-level classification logic.

The raw current value of the driver metric. This is the actual metric reading before it is translated into z-scores, percentiles, or scorecard scores.

Raw current values are crucial for traceability because they let advanced users move from explanation back to the underlying observed number. Correct interpretation depends on the metric's own unit definition, not on the driver wrapper itself.

The metric currently standing out enough to be listed as a driver of the published regime. Drivers are the pieces of evidence the model thinks are most relevant right now.

Drivers are filtered and ranked from candidate signals, with axis-specific weighting and label-consistency filtering. The result is a compact, published explanation layer showing which metrics most strongly support the classification.

This compares a shorter recent average with a slower longer average. Positive values usually mean the metric has been accelerating recently. Negative values usually mean it has been cooling or fading.

Momentum 7d vs 30d is the product's compact short-versus-long directional signal. It matters because many descriptive states should care not only about level but also about whether pressure is still building or easing. This is one of the core ingredients that separates a persistent trend from a one-day spike.

This tells you how unusual the metric currently looks relative to its own history. The larger the absolute value, the more exceptional the reading is. 'Robust' means the method tries to be less sensitive to outliers than a naive standard deviation approach.

z_robust is one of the main driver-sorting signals in the UI and in backend support logic. It is especially important because label confidence uses driver signal support. Very small absolute z-scores mean the metric is not standing far from its own baseline; large absolute z-scores mean the metric is contributing unusually strong evidence.

The directional reading attached to a driver, such as heating, cooling, or flat. It tells you whether the metric has recently been building, fading, or staying roughly unchanged.

Driver trend is part of the regime-engine signal summary and is used both for explanation and for label-consistency checks. For example, HEATING labels prefer drivers that are either high or still directionally heating rather than merely elevated in isolation.

CHEAP means the chain currently looks easy to use relative to its own history. That usually means friction is low and capacity pressure is low at the same time, so the network appears to have room to spare.

The ruleset assigns CHEAP when both Friction and Capacity are in low bands. This is important: the label is not 'fees are low' in isolation. It is a joint state where the chain looks inexpensive and unconstrained relative to its own historical behavior.

CONGESTED means the chain appears to be operating under real capacity pressure. Usage is high enough relative to available throughput that users are more likely to feel the network becoming crowded through higher fees, fuller blocks, slower execution, or more failures.

The ruleset assigns CONGESTED when Capacity is EXTREME_HIGH, or when Capacity is HIGH and Friction is also HIGH. This is intentionally stricter than 'demand is high'. The label is meant to describe a chain where demand is pressing against execution capacity, not merely a chain that is busy.

HEATING means demand looks stronger than usual and the recent direction still points upward. In plain language, activity appears to be building rather than merely producing a single isolated spike.

The regime engine assigns HEATING when Demand is in a high band and at least one relevant axis trend is also HEATING. That means the model is looking for both elevated level and positive short-versus-long momentum. It is therefore stronger than 'high today' but weaker than a fully congested state.

This is the published color hint for the regime badge. It helps users separate states visually, but the label itself is what matters.

The UI prefers the published status.color when present and only falls back to local color mapping if needed. Color is presentation, not methodology. It should never be interpreted as extra model output beyond the published regime label.

The regime label is the product's compact description of the chain's current on-chain state. It is descriptive only. It does not predict what happens next and it does not tell the user what to do. Its job is to summarize whether the latest published evidence looks more like stable conditions, heating demand, congestion pressure, cheap conditions, or a degraded / low-confidence state.

The frontend treats status.label as the canonical published regime label and only falls back to regime.label if status.label is unavailable. In the backend, the label is produced by deterministic rules over Demand, Friction, and Capacity evidence, with a confidence gate that can force UNKNOWN/DEGRADED. The UI does not recompute the label. The correct interpretation is therefore 'published classification result', not 'UI opinion' or 'forecast'.

The one-liner is a short human-readable summary of the published regime. It is there to make the page readable at a glance before the user dives into the detail.

This text is pipeline-authored descriptive copy published alongside the regime label. The UI renders it directly and should not be treated as an independent inference layer. It compresses regime, confidence, and chain context into one short sentence.

STABLE means the chain does not currently show a strong enough combination of demand pressure, friction pressure, or cheap-capacity conditions to justify a more extreme label. It does not mean 'nothing is happening'. It means the chain still looks broadly within its normal historical operating range.

In the ruleset, STABLE is the default label when the evidence does not meet CONGESTED, CHEAP, or HEATING conditions and the confidence gate does not force UNKNOWN/DEGRADED. In practice this usually means scorecard dimensions are not far enough from neutral, or the directional evidence is not persistent enough, to support a stronger regime label.

UNKNOWN/DEGRADED means the product does not have enough trustworthy evidence to publish a stronger regime label confidently. The latest data may still be visible for traceability, but the classification itself should be treated as insufficiently supported.

This state is usually triggered by the confidence gate rather than by a separate market condition. In the current model, the published regime becomes UNKNOWN/DEGRADED when combined publish confidence falls below the configured threshold. It is therefore an evidence-quality state, not a fifth economic regime in the same sense as STABLE, HEATING, CONGESTED, or CHEAP.