The Speed Trap Most COD Programs Fall Into
Most coaches will tell you that change of direction is a speed and agility problem. Run faster cone drills, sharpen your footwork, get quicker off the line. Here's where I disagree: approach speed is not the primary limiter of COD performance. Deceleration capacity is. And almost nobody trains it directly.
A 2025 systematic review published in Sports Medicine (Singh et al., James Cook University) pulled 13 studies covering 374 participants performing 45° to 180° COD tasks, and the findings back this up hard. Change of direction biomechanics are dominated by braking kinetics and centripetal force output, not raw velocity. If you can't control your momentum into the cut, your exit velocity doesn't matter because you're already compromised before you plant your foot.
Let me show you exactly what the research identified and what it means for how you train.
What Actually Determines COD Completion Time
The 2025 systematic review identifying biomechanical determinants of COD performance found several kinematic and kinetic variables consistently correlated with faster completion times across COD angles. These aren't soft associations. They're the structural variables that explain why one athlete cuts faster than another at the same approach speed.
The key contributors the review identified include:
- Shorter ground contact time during the final foot contact before the cut
- Higher approach and exit velocities (yes, approach matters, but it's downstream of braking capacity)
- Increased braking and propulsive ground reaction forces
- Greater trunk inclination angle at contact
- Lower center-of-mass height during the penultimate and final contacts
- Increased moments and power at the hip, knee, and ankle
Notice what's at the top of that list. Braking forces and ground contact time. Not 40-yard dash speed. Not hip flexor flexibility. The ability to apply force quickly into the ground to decelerate, and then redirect, is the mechanical engine of every cut you'll ever make.
The Angle Changes Everything
One detail the review makes clear that most agility programs completely ignore: the biomechanical demands shift substantially depending on COD angle, and a program that doesn't account for that is training the wrong thing for your sport.
At 45°, faster performance correlated with greater average sagittal hip power generation and peak hip flexor and plantar flexor moments. You're still moving mostly forward, so the demand is hip drive and push-off power. At 90°, the review found that greater average frontal plane hip power generation and peak knee extensor moment became the differentiators. The lateral component is now dominant. At 180° and sharper cuts, braking demand surges because approach momentum has to be almost entirely absorbed before you can redirect. That's a force absorption problem first and a propulsion problem second.
The principle of specificity applies here directly. If your athletes mostly perform 90° cuts (soccer, basketball, rugby), training their 180° deceleration mechanics through repeated shuttle runs without loaded eccentric work at that angle is leaving a gap in their preparation.
Sex Differences and What They Mean for Programming
The Singh et al. review on COD biomechanical determinants noted that 79.4% of study participants were male, which is a methodological limitation worth naming. But the data that does exist on sex differences is important, particularly for ACL injury risk profiling.
Male athletes demonstrated significantly greater maximum velocity, acceleration, deceleration magnitude, and centripetal force output across all COD angles tested. That sounds like a pure performance advantage. But here's the injury-side implication: unplanned COD tasks (reactive agility) produce greater knee abduction moments than pre-planned cuts. And knee abduction moment is one of the most consistently identified biomechanical predictors of ACL injury risk.
Female athletes, on average, show higher knee abduction moments during cutting tasks, which tracks with the well-documented 2-8x higher ACL injury rate in female athletes in sports like soccer and basketball per existing epidemiological data. The review's distinction between pre-planned COD and reactive agility matters here. If you're only testing athletes on pre-planned cuts, you're not capturing the true knee load they experience in competition. That's a screening blind spot.
Why Your Deceleration Mechanics Are Probably Undertrained
I've worked with athletes at the high-school and collegiate level who come in able to squat 1.5-2x bodyweight but can't control a single-leg deceleration at game speed without their knee caving. The strength is there. The rate of force development on the braking side isn't, and neither is the motor pattern to express it in a cutting context.
The force-velocity curve tells you why. Deceleration is a force absorption task. The muscles have to generate high eccentric force rapidly during the braking phase. If you've only trained the concentric side of that curve (squat up, jump up, accelerate), your eccentric braking capacity is the weak link. The review's finding that increased braking forces correlate with faster COD completion times is a direct argument for eccentric-dominant lower-body work as a COD training tool, not just cone drills.
If you hand an athlete a cone drill and call it COD training without also addressing their deceleration kinetics and ground contact strategy, you've given them a test, not a training stimulus. The drill reveals the problem. The strength work fixes it.
Supercompensation doesn't happen from drills alone. It happens when you expose the neuromuscular system to progressive loading, let it recover, and adapt. That means structured plyometric and resistance work targeting the eccentric phase of cutting movements, not just repeated sprint-and-cut exposure.
How to Actually Train COD Biomechanics
Based on what the systematic review on COD biomechanics identified, here's where your training emphasis should go if COD performance is a priority:
- Eccentric strength development at the knee and hip: Nordic hamstring curls, single-leg RDLs with controlled descent (3-4 seconds), Bulgarian split squats with emphasis on the lowering phase
- Plyometric progressions that include deceleration as the primary objective: broad jump to stick landing, lateral bound to controlled hold, progressing to multi-directional bounds at sport-relevant angles
- Ground contact time reduction: reactive jumps and hops, but only after the athlete has shown they can absorb force cleanly under slower tempos first
- Trunk inclination coaching: athletes who don't lean into the cut lose the mechanical advantage at the hip. This is a technical cue, not just a strength fix
- Angle-specific drill design: if your sport requires 90° cuts, train 90° cuts under load. Don't assume 45° cone work transfers directly
Run this kind of block for 8-12 weeks before reassessing COD completion time. Expect measurable change in ground contact time and exit velocity, not in week two but in weeks 8-12 when the neuromuscular adaptations have had time to consolidate.
The Reactive Agility Gap You're Missing
Here's the finding from the review that I think deserves more attention than it's getting: every single study in this systematic review used pre-planned COD tasks. Not one looked at reactive agility in a correlational design linking biomechanics to performance time.
That's a massive gap. Because in sport, most cuts are reactive. A defender shifts, a ball changes direction, a teammate opens a lane. Your athlete doesn't know the cut is coming. And as the review notes, unplanned COD produces greater knee abduction moments than pre-planned movement. That means the injury risk is highest exactly when your athlete is least prepared mechanically, and the research base on that scenario is almost empty.
Practically, this means your training program should include reactive agility work alongside structured COD drills. Don't assume pre-planned COD training transfers completely to decision-driven cutting. The motor patterns have meaningful differences, and the force outputs at the knee are not the same.
If you're working with athletes returning from lower-limb injury or coming back after a long layoff, the Comeback Code program addresses this exact progression: building braking mechanics and neuromuscular control before reintroducing sport-speed reactive cutting, which is the sequence the evidence supports.
What to Take Into the Gym
Change of direction biomechanics come down to a few non-negotiable mechanical qualities: deceleration magnitude, centripetal force output, ground contact time, and the ability to maintain trunk position through the cut. Those are the variables the 2025 Singh et al. systematic review identified as primary determinants of COD completion time across 45°, 90°, and 180° tasks.
Approach speed matters. But if your athletes can't brake efficiently, faster approach speed just means harder crashes into the cut. Train the brakes first. Then build the engine on top of a foundation that can actually control it.
References
Singh U, Leicht AS, Connor JD, Brice SM, Alves A, Doma K. Biomechanical determinants of change of direction performance: a systematic review identifying braking forces, centripetal output, and ground contact time as primary COD performance variables. Sports Medicine. 2025;55(9):2207-2224. doi: 10.1007/s40279-025-02278-3