07
FORCE-TIME CMJ SERIES · READINESS SIGNAL
OUTPUT HELD.
RATE COLLAPSED.
Jump height stayed flat or rose. Braking RFD fell 20–40% from baseline. The curve already told a different story.
L·1 · L·2 · OPP·3
07
THE FINDING
THREE ATHLETES.
JUMP HEIGHT HELD.
BRAKING RFD
COLLAPSED.
JUMP HEIGHT HELD.
BRAKING RFD
COLLAPSED.
Six monitoring sessions across 10 weeks. Output stayed flat or rose. Braking RFD fell up to 40.2% from individual baseline. The decoupling is systematic. It is not noise.
−40.2%
Braking RFD drop · L·1 · session 1 → session 3 (peak drop)
7,665 N/s → 4,583 N/s across 6 sessions
+6.2%
Jump Height · L·1 · same 6-session window
37.01 cm → 39.30 cm. The output went up. The signal collapsed underneath it.
−43.3%
Peak rate-output divergence · L·1 · session 4 (Mar 20)
BrRFD −37.0% while JH was +6.3% from baseline. Same athlete. Same session.
N1 PERFORMANCE LAB
ISSUE 07
The Architecture
NOT ALL RFD IS THE SAME
THE WINDOW DEFINES THE METRIC. CHANGE THE WINDOW AND YOU CHANGE WHAT YOU ARE MEASURING.
THE EVIDENCE
MERRIGAN
ET AL.
2022
ET AL.
2022
Journal of Strength and Conditioning Research. Ohio State · West Virginia University. Compared Hawkin Dynamics, ForceDecks, and custom MATLAB scripts across CMJ, SJ, DJ, and IMTP. 24 subjects. 6 trials per test.
"It is best to ignore propulsive RFD... only braking phase RFD shows the most promise for comparisons across software."
— MERRIGAN ET AL. (2022), JSCR
RFD METRIC RELIABILITY · CMJ
| RFD Type | R² vs. MATLAB | Avg. Error | ODS |
|---|---|---|---|
| Propulsive RFD | 36% | 723% | Fluffy |
| IMTP 0–50ms RFD | 75% | 118% | Fluffy |
| IMTP 0–250ms RFD | 81% | 11.8% | Fluffy |
| Braking RFD | 99.9% | <1% | Driver |
HAWKIN DYNAMICS OVERALL · ALL TESTS
| Test | Error vs. MATLAB | Verdict |
|---|---|---|
| Squat Jump | <3% | Validated |
| CMJ | <3% | Validated |
| Drop Jump | <3% | Validated |
| IMTP | <3% | Validated |
ODS classification: Lake & Bishop (2021). Error = % difference vs. custom MATLAB analysis. Source: Merrigan JJ, Stone JD, Galster SM, Hagen JA. JSCR 2022.
L·1 · 6-SESSION MONITORING
BRAKING
RFD
COLLAPSED.
RFD
COLLAPSED.
Jump Height rose +6.2% across the monitoring window. Braking RFD fell −40.2% by session 3. The output went up while the rate collapsed underneath it.
7,665
Baseline Braking RFD (N/s)
4,583
Session 3 Braking RFD (N/s) · peak drop
SESSION LOG · L·1
| Session | Braking RFD (N/s) | RFD Δ% | JH (cm) | JH Δ% |
|---|---|---|---|---|
| Feb 13 | 7,665 | baseline | 37.01 | baseline |
| Feb 17 | 6,141 | −19.9% | 38.66 | +4.5% |
| Mar 13 | 4,583 | −40.2% | 36.39 | −1.7% |
| Mar 20 | 4,832 | −37.0% | 39.33 | +6.3% |
| Apr 03 | 5,250 | −31.5% | 37.97 | +2.6% |
| Apr 17 | 6,219 | −18.9% | 39.30 | +6.2% |
Sessions 3–5: JH at +6.3% to −1.7% while BrRFD was −40.2% to −31.5%. Output masked a substantial rate deficit. BrRFD CV avg 13.03%; S5 spike 28.77% — flagged, does not reverse the trend direction.
CONTEXT — L·1 OVER SIX SESSIONS
THE OUTPUT
WENT UP.
THE RATE
COLLAPSED.
WENT UP.
THE RATE
COLLAPSED.
L·1's Braking RFD fell to −40.2% by Session 3 (Mar 13). Jump Height rose to +6.3% by Session 4 (Mar 20). The output went up while the rate collapsed. Motor strategy compensation preserved the jump while the rate capacity degraded.
SOURCE: HAWKIN DYNAMICS CMJ · L·1 · n=6 SESSIONS · FEB–APR 2026
PROOF — THE PATTERN HOLDS ACROSS THE COHORT
THREE
ATHLETES.
ONE
PATTERN.
ATHLETES.
ONE
PATTERN.
Three athletes. Three peak divergence moments. L·1 −43.3% at Session 4. L·2 −38.2% at Session 6. OPP·3 −28.9% at Session 4. The cleanest CV in the cohort (OPP·3 at 5.75%) still showed a 29-point gap. The decoupling is not noise.
SOURCE: HAWKIN DYNAMICS CMJ · n=3 ATHLETES · 6 SESSIONS · FEB–APR 2026
RISK — THE RECOVERY WINDOW
DAY ONE
FALLS
HARDEST.
FALLS
HARDEST.
Vertical jump decrement peaks at 24h post-session (ES = −0.42; 95% CI −0.57 to −0.26; k=78). Day 2 still significant (ES = −0.20; p = 0.015). Returns to baseline by Day 3. Peak Height detects the decrement more sensitively than Peak Power (ES −0.47 vs −0.14; p = 0.04).
SOURCE: WILLIAMS ET AL. 2025 · J FUNCT MORPHOL KINESIOL · META-ANALYSIS k=184 EFFECTS · 22 STUDIES · n=296
DECISION — WHAT TO DO MONDAY
WHAT TO DO
MONDAY.
MONDAY.
0–5% drop: load normally. 5–10%: amber, watch RPE. 10–20%: deload trigger. >20%: medical screen. Trigger is BrRFD % from individual baseline — not a population norm. n=18 sessions, 3 athletes, 6 sessions each.
NEXT TEST ANSWERS
Whether Braking RFD recovery tracks within the 72-hour window after a deload — and whether Yielding RFD leads or lags the Braking RFD recovery signal.
YOSHIDA 2023 · WILLIAMS 2025 · BERBERET 2021 · n=18 SESSIONS · 3 ATHLETES · 6 SESSIONS EACH
07
THE SIGNAL
IS IN THE
CURVE.
JUMP HEIGHT TELLS YOU WHAT HAPPENED. BRAKING RFD TELLS YOU THAT IT WAS ALREADY HAPPENING.