Adjunct Analysis (Analysis)
Source: (inline in performance/page.tsx)
What You See
Cards for tracked extras like supplements, shoes, or nutrition. Each card shows: number of sessions analyzed, median projection delta (seconds gained or lost), HR drift change, volatility change, and a confidence bar.
How the Numbers Work
PIRX compares sessions where you used a particular adjunct (like a specific shoe) against sessions without it. The median projection delta shows the typical difference in projected time. HR drift measures whether your heart rate climbed faster or slower during those sessions.
Step by Step
The backend splits activities into two groups: those tagged with an adjunct (a shoe, supplement, or nutrition item) and those without. For each group it computes: median projection delta (the change in projected time on the day after each session), mean HR drift (how much heart rate climbed from the first half to the second half of each run), and projection volatility (standard deviation of recent projection changes). The reported values are the difference between tagged and untagged groups. Confidence = min(tagged_session_count / 20, 1.0), so you need at least 20 tagged sessions to reach full confidence. The confidence bar width is proportional to this value.