CMJ Season Intelligence Dashboard

1 · Takeaways: what to do

Five game-plan priorities. Swipe or scroll sideways to read all five cards on smaller screens.

TEAM PERFORMANCE

1 · The team gets quicker, not higher, and peaks at the right time

Jump height barely moves all year (+0.7%, not a real change). But by Post-Season players leave the ground ~5% faster and explosiveness (mRSI) is up 5% (both p<0.001).

Do this: Keep the current in-season plan, because it's working. Judge readiness by mRSI / time to takeoff, not jump height alone. A change only counts if it's bigger than that player's normal day-to-day noise: about 2–4% for jump height and 3–7% for mRSI.

ATHLETE WATCH LIST

2 · Check in with Athlete 9 and Athlete 5. Copy Athlete 4.

A9 (Guard, 5'9"): jump height down 4.8% (real), body weight up 2.3%, and the team's worst landing imbalance (~20–32% vs ~9% for everyone else). → Check body composition, single-leg strength and landing mechanics.

A5 (Big, 6'11"): jump height drops 3% from Conference to Post-Season, and braking force drops too. Classic late-season fatigue. → Manage minutes and practice jump volume.

A4 (Big): best responder, with jump height up 11%. Whatever he's doing (training, recovery), protect it.

COACH FOOTWEAR

3 · Bigs land harder as the season goes on. Guards land softer.

Guards' landing force drops 11% over the season. Bigs' landing force rises, and by Post-Season Bigs absorb ~33% more force (relative to body weight) per landing than Guards. This position difference is statistically real (p<0.001).

Coach: drill soft landings and cap high-impact jump volume for Bigs late in the season. Footwear: test cushioning / impact protection for Bigs and responsiveness / quick energy return for Guards. Their gains come from speed, not height.

FOOTWEAR INNOVATION

4 · Left-right imbalance creeps up late in the season, mostly in smaller players

The left-right difference when braking (absorbing force before the jump) grows +66% from Non-Conference to Post-Season (p<0.001). Shorter players are hit hardest. Biggest jumps: A3 (+230%), A9 (+140%), A1 (+64%) (all Guards) and Big A6 (+85%). Jump height won't reveal this (r ≈ 0), so it hides.

Do this: make L/R asymmetry a primary outcome in shoe tests. Test stability, heel lockdown and traction on these athletes during Conference and Post-Season, when the problem actually shows up, not only in preseason.

FOOTWEAR TEST DESIGN

5 · The season itself changes the jump, so don't let it fool your shoe test

Athletes naturally get ~5% quicker by Post-Season. If Shoe A is tested in November and Shoe B in March, Shoe B looks better even if it isn't.

Do this: compare shoes within the same phase, ideally the same week (A → B → A swap). Compare each athlete to their own baseline and group Guards with Guards and Bigs with Bigs. Call a shoe a "win" only if the gain beats that athlete's noise band (see the stats file, sheet athlete_noise_CV).

Metric

Compare

Athlete

2 · Team scoreboard

Team-wide change for the selected comparison. Green means the change helps, red means it hurts. Statistics use a mixed model that accounts for each athlete's own baseline.

How the Coil-Cushion Index works (exploratory footwear-testing tool)

CCI v1 combines three equally weighted signals: greater stiffness divided by body weight, shallower squat depth, and shorter braking time. Each is converted to a 0–100 percentile against this roster's Non-Conference tests, with each athlete weighted equally. Depth and braking-time percentiles are reversed, then all three are averaged. The same reference is used throughout the season; missing or non-positive inputs produce no score.

Cushion ≤40: deeper / longer-loading strategy. Hybrid >40 to 60: middle range. Coil >60: compact / quick-loading strategy. Labels use each athlete's average score in the selected destination phase, not a permanent athlete identity. These cutoffs and equal weights are provisional, not validated. The inputs overlap biomechanically; they are not three independent proofs of elasticity.

Use profiles to stratify matched shoe crossover tests. Coil: compare forefoot responsiveness and stability. Cushion: compare controlled compression and platform support. Hybrid: compare balanced setups. Measure jump performance, comfort, landing force and asymmetry separately in every profile. The index does not measure tendon elasticity, shoe energy return, injury risk, or prove that a shoe works. Higher is not better; “Cushion” describes takeoff strategy, not softer landings.

Score changes are shown in points, not percent. CCI tests have their own exploratory FDR family, preserving the original metrics' statistical results. Correlations with its ingredients partly reflect the formula. The spreadsheet includes the reference distribution and athlete-phase score means.

3 · Season report card: who changed, and how much?

Each cell shows the % change for one athlete on one metric (Coil-Cushion Index uses points). Green = better, red = worse, gray = context (neither good nor bad). ★/★★/★★★ = significant (p<.05/.01/.001, corrected). Click any cell to open that athlete and metric below.

4 · Athlete explorer

Left: every test across the season, with a 3-test rolling average and the average for each phase (dashed). Right: spread of tests in each phase, with significance brackets.

5 · Season in motion ▶

Press Play to watch each athlete move through the season (3-test rolling average). The title follows the session: 1–13 Non-Conference, 14–31 Conference, 32–38 Post-Season. Bubble size = body weight. Moving up and left means jumping higher and taking off faster.

X Y

6 · All 9 athletes side by side

Average ± 95% confidence interval for each phase. The leftmost Y-axis in each row labels the metric selected above. The title shows the % change for the selected comparison (Coil-Cushion Index uses points).

7 · Does position or height change the season story?

Left: Guards vs Bigs (thick lines = group average ± SE; thin lines = individual athletes). Right: athlete height vs. change (%; Coil-Cushion Index uses points). Note: in this roster, every Guard is shorter than every Big, so position and height are closely linked.

8 · What moves together?

Correlation: +1 = rise together, −1 = one rises while the other falls, 0 = unrelated. Use Within-athlete to answer: “when this player has a good day on X, is Y also better?”

In plain English: explosiveness (mRSI) moves almost one-for-one with how fast a player leaves the ground (r = −0.80) and with power (r = +0.82). A deeper squat means a slower takeoff (r = +0.80). Jump height is only loosely tied to explosiveness (r = +0.46), so a player can jump the same height while getting quicker. Asymmetry and landing force are unrelated to performance (r ≈ 0), so they won't show up in jump results and must be watched on their own.

9 · Next opportunity: see how they move

MARKERLESS MOTION CAPTURE

Force plates show the output. Video adds the movement story.

CMJ force-plate data quantify jump performance, force and timing. Markerless kinematics can add how the athlete moves: hip, knee and ankle motion, trunk position, and joint-level movement strategy during takeoff and landing. The same jump result can come from different movement patterns.

Keep it simple: explore single-camera tools such as Essex EZCap markerless motion capture, using standard video without reflective markers. With a prepared camera station, aim for setup in <60 seconds (a workflow target, not a guaranteed tool specification). Record alongside routine CMJ testing so movement can be checked frequently throughout the season, rather than only in an occasional lab visit.

Coach: investigate whether a change in jump performance accompanies a change in movement strategy. Footwear innovation: compare joint motion and landing strategy between shoes alongside force, performance, comfort and asymmetry. Use matched, same-phase tests; these measurements do not by themselves prove a shoe benefit.

Make comparisons trustworthy: keep camera position, lighting and the test protocol consistent; collect 2–3 clean trials and obtain consent for video recording. Single-camera estimates depend on viewing angle and visibility and do not replace high-precision 3D laboratory motion capture. This is a future opportunity; no markerless measurements were collected in the current dataset.