Definitions for the activity signals, replay windows, and market-performance terms used throughout the terminal.
CAS Score measures how much core protocol activity is accelerating right now compared with the project's recent baseline. Quantmetric computes it from core commits over the past 7 days divided by the project's normalized 30-day activity rate. A higher score means the developer surface is heating up, but it is not a price prediction by itself.
Surge Multiplier is the multiple captured when a signal is created. It freezes the activity spike at T0 so later settlement can replay what the system knew at the time. This differs from Current CAS Score, which keeps moving as new commits arrive and older commits leave the rolling window.
T0 Baseline is the token and BTC price snapshot taken when a signal is created. Quantmetric uses this point as the starting line for 7-day and 30-day settlement. If T0 is missing, the signal cannot be fairly evaluated because future price information would leak into the result.
Settled (7d) means the signal has reached its 7-day evaluation window. Quantmetric compares the token's return from T0 against BTC over the same period, then records absolute return and alpha. The signal may still remain open for the 30-day replay if the longer window has not closed.
Settled (30d) means the full replay window has completed. At this point the signal has both short-window and full-window performance data, including token return and alpha versus BTC. These records are used in the track record to judge whether developer activity surges have historically preceded outperformance.
Invalidated means a signal was removed from fair evaluation because its original assumptions no longer held. Common reasons include missing T0 price data, look-ahead contamination, or new core commits changing the event boundary before the replay could settle. Invalidated signals are preserved for transparency but excluded from performance claims.
Alpha vs BTC is the token's return minus BTC's return over the same window. Quantmetric uses BTC as the crypto market benchmark because many tokens move with broad market beta. Positive alpha means the token beat BTC during the replay window; negative alpha means it lagged BTC.
Core Change is a commit touching protocol-critical paths rather than docs, tests, examples, release notes, or peripheral tooling. Each project has monitored directories that define its core surface. This keeps CAS focused on engineering work that is more likely to matter for protocol behavior, security, execution, or market perception.
Look-Ahead Bias happens when future information changes how a past event is interpreted. Quantmetric avoids this by freezing signal inputs at creation time, using T0 price snapshots, and invalidating signals when the event boundary is contaminated. The goal is to make every replay answer the same question a trader could have asked then.
Monitoring means a signal has been created and is waiting for its settlement windows to close. The signal has a valid T0 baseline, but the 7-day or 30-day outcome is not ready yet. During this period, the record stays live while Quantmetric waits for enough time to measure the replay honestly.