CMJ Depth Inconsistency Is Quietly Wrecking Your Power and RSI Comparisons

An athlete posts a season-best peak power reading in Tuesday's testing block. The coach flags it in the weekly report, then reverses course three days later when the same athlete's power number drops 18% with no change in bar speed, sleep quality, or reported soreness. Nobody touched the training load between the two sessions. What changed was about five centimeters of dip that nobody was watching.

Countermovement depth is the one input variable in a CMJ trial that quietly self-selects session to session unless someone actively controls it, and it drives peak power, net impulse, RSImod, and even jump height in ways that have nothing to do with an athlete getting stronger, more explosive, or more fatigued. A rushed warm-up produces a shallower dip; a slower, more deliberate approach on a fresh day produces a deeper one. Both can happen on the same athlete, a week apart. The dashboard shows two different numbers. Only one of them reflects a real change in the athlete.

Depth Sets the Time and Distance Available to Produce Force

A countermovement jump runs on impulse-momentum: the net force an athlete applies above bodyweight, multiplied by how long they apply it, determines the velocity they carry into takeoff. Countermovement depth sets the ceiling on that available time and distance. A deeper dip generally buys more time to build net impulse before the concentric phase has to fire, which tends to raise jump height and peak force up to a point. Push depth past an athlete's own efficient range and the concentric phase runs into the force-velocity trade-off — more distance to cover, but less force available at the joint angles and contraction speeds required to cover it quickly, so movement time balloons faster than output improves.

The Same Height Can Come From Two Different Movements

Because depth pulls several metrics in different directions at once, a coach watching only one number gets fooled in both directions. A shallower dip shortens movement time, which pushes RSImod up even when jump height and peak power both come in lower — the athlete spent less time in the well, not more power at the bottom of it. A deeper dip does the opposite: movement time stretches out, RSImod drops, and yet net impulse and jump height can be equal to or better than the shallow trial. Read any one of these metrics in isolation and depth inconsistency can make an athlete look like they regressed on their best week or improved on their worst one.

Domire and Challis (2007): Depth Has an Individual Optimum, Not a Universal One

Domire and Challis (2007, Journal of Sports Sciences) built a torque-driven computer simulation of the lower extremity, calibrated to individual subjects, to test how varying squat depth changes maximum achievable jump height. Their model consistently found a depth window, specific to each simulated athlete's own segment lengths and strength profile, where predicted jump height peaked — and jump height fell off on both sides of that window. Go shallower than an individual's optimum and there isn't enough time or range to build net impulse before takeoff; go deeper than it and the concentric phase runs into the force-velocity trade-off, spending extra time without a matching payoff in propulsive force. The practical read for testing: there's no single correct countermovement depth that applies across an entire roster, which is exactly why comparing one athlete's raw depth number to another's, or to a fixed team-wide target, tells you less than comparing each athlete's number to their own established range. The limitation is real: this is a simulation calibrated on a small number of subjects rather than a large empirical field trial, so the precise width of the optimal window shouldn't be read as a fixed figure to apply to every athlete on a roster.

McMahon, Suchomel, Lake, and Comfort (2018): Metrics Only Mean the Same Thing If the Phases Do

McMahon, Suchomel, Lake, and Comfort (2018, Strength and Conditioning Journal) laid out standardized definitions for the phases of the CMJ force-time curve — unweighting, braking, and propulsion — specifically because inconsistent depth changes how long each phase runs and how it should even be identified on the trace. Their central argument: a metric like RSImod, calculated as jump height divided by total movement time, isn't measuring a fixed quality of the athlete when the movement time itself is inflated or compressed by how deep the countermovement went. A shorter movement time from a shallower dip can push RSImod up even when the underlying propulsive output is unchanged or worse, and a longer movement time from a deeper dip can drag RSImod down even when jump height improved. The limitation to flag: this is a methodological and narrative synthesis of existing force-time curve research rather than a single new dataset with its own effect size, so it's best read as the applied consensus on why phase and depth consistency matters, not as a number to cite directly.

Depth Control Is Rarely Zero-Effort, and Rarely Free

Left to self-select, athletes rarely repeat the same countermovement depth from one testing day to the next, especially without a visual or physical target to aim for. Warm-up length, verbal cueing, fatigue, and even how rushed a session feels all nudge depth in one direction or another before the athlete notices anything is different. The table below reflects the general pattern seen in applied field testing as depth control tightens step by step — treat the ranges as practical planning figures rather than a number to cite from a single trial.

Five Steps to Lock Depth Down Without Losing Natural Movement

1. Establish each athlete's own baseline first. Run 4-5 unconstrained maximal CMJ trials and record the depth on the best two jumps by height. That average becomes the athlete's target, not a team-wide number pulled from a chart.
2. Give the athlete something physical to aim for. A dowel or length of PVC pipe held at hip height, or a rope stretched across the rig at the target depth, works better than a verbal cue alone — athletes chase a visual target far more consistently than a spoken instruction.
3. Set a tolerance band, not a single number. Roughly plus or minus 3-5 centimeters, or about 10% of the athlete's baseline depth, is a workable field threshold. Anything tighter starts to interfere with a natural stretch-shortening rhythm.
4. Record depth on every trial, not just height. Most force plate software already reports a dip depth or countermovement depth field alongside jump height — pull it into the same table you use for power and RSI, and treat it as a gating column before you read anything else.
5. Flag, don't discard, out-of-band trials. A trial outside the tolerance band isn't automatically bad data — it might be genuinely useful information about a rushed warm-up or a fatigued approach — but it shouldn't be compared directly against a session where depth sat inside the band without that context attached.

The Numbers

Reading It

Glance at RSImod alone and Wednesday looks like the better session — a higher ratio, which usually reads as sharper reactive strength. Jump height and peak power tell the opposite story: both came in meaningfully lower on Wednesday. The athlete rushed the dip, spent less time producing force, and the shortened movement time inflated the ratio even though the actual output dropped. Friday's trial, run against the athlete's own standardized depth target, shows what their reactive strength quality actually looks like once the input variable that drives movement time stops drifting underneath the number. Report Wednesday's RSImod on its own in a weekly summary and the conclusion runs backwards from what actually happened in the session.

Where Depth Control Efforts Usually Break Down

• Setting one depth target for the whole roster. Depth scales with leg length, ankle mobility, and an athlete's own force-velocity profile. A target built for a 175 cm athlete misapplies to a 195 cm one, and both end up jumping outside their efficient range.
• Cueing height instead of depth. Telling an athlete to jump as high as possible says nothing about how they get there, and under that instruction most athletes drift toward whatever depth feels fastest on a given day rather than the one that reflects their trained capacity.
• Comparing RSImod across sessions without checking the depth column first. RSImod moves for two completely different reasons — a real change in reactive strength, or a change in how deep the athlete dipped — and the metric alone can't tell you which one happened.
• Treating a tighter depth constraint as automatically better. Locking depth down to within a centimeter or two can start to override an athlete's natural stretch-shortening timing, producing a movement that's consistent but no longer representative of how they actually jump in competition.
• Losing the depth data entirely. A lot of testing setups log jump height and power but never export the depth field, which makes it impossible to go back and check whether a flagged trend was real or just a drifting dip.