The COD Myth That's Costing You Time
Ask most S&C coaches what drives change-of-direction speed and you'll hear the same answer: explosiveness out of the cut. Build a more powerful push-off, accelerate harder after the turn, win. It's intuitive. It's also incomplete, and for sharper cuts, it's flat-out wrong.
A 2025 systematic review published in Sports Medicine (Singh et al., James Cook University) examined 13 studies covering 374 participants across COD tasks ranging from 45 to 180 degrees. The consistent finding: horizontal braking impulse change of direction performance relationships show that braking force, not peak propulsive force, is the dominant mechanical driver, especially at angles of 90 degrees and above.
That changes how you should be programming. Not slightly. Fundamentally.
What Horizontal Braking Impulse Actually Means
Impulse is force multiplied by time. When an athlete enters a cut, they're carrying horizontal momentum in the original direction of travel. To redirect that momentum, they have to kill it first. The question isn't just "how hard can they push off?" It's "how quickly can they absorb that incoming force over the shortest possible ground contact?"
The Singh et al. review makes this concrete: athletes who generate greater braking forces over shorter ground contact times complete COD tasks faster. Shorter contact time with adequate braking force means higher impulse per unit time. That's the mechanical signature of a clean, fast cut.
The practical implication is direct. If you can brake hard and fast, you can approach the entry point at a higher velocity. You don't have to start slowing down 3 strides out. You slow down in the cut, not before it, and that shrinks total COD time more than anything you can do on the propulsive side.
Where Angle Changes Everything
This isn't a blanket rule across all COD tasks. The review is careful here, and you should be too. At shallower angles (45 degrees), sagittal hip power and plantar flexor moment correlate strongly with faster times. The cut is more of a redirect than a full deceleration, so propulsive capacity matters more.
At 90 degrees and beyond, the calculus shifts. Sharper cuts require substantially greater braking to manage the higher momentum vectors, and the athletes who complete them faster are the ones generating higher braking force, not higher push-off force. A 505 test, a T-test, a 180-degree soccer cut, a basketball closeout step. These are the tasks where deceleration is the rate-limiter.
The review also flags that shorter ground contact time at the final foot plant (r = -0.48 in a 75-degree task per Marshall et al., cited within the review) correlates with faster completion. Shorter contact means the athlete is both stiff enough and strong enough to absorb and redirect without settling into a prolonged brake. That's a trainable quality. Most programs don't train it directly.
The Pre-Planned vs. Unplanned Problem
Here's something the review surfaces that deserves more attention: every single one of the 13 included studies used a pre-planned COD task. Zero studies examined reactive agility under the same biomechanical lens.
That matters because the knee abduction moment profile, the trunk lean angle, and the ground contact timing all change when an athlete has to react to a stimulus rather than execute a memorized pattern. Pre-planned cuts allow the athlete to prepare their deceleration strategy. Unplanned cuts don't. The braking demand doesn't disappear under reactive conditions; it arguably increases because entry velocity is less controlled.
You can't directly extrapolate the braking impulse correlations from pre-planned research to unplanned agility without a caveat. But the training intervention still applies: an athlete who can't generate adequate horizontal braking force under controlled conditions won't generate it under chaotic ones.
Most Coaches Will Tell You to Add More Acceleration Work
Here's my contrarian take: the typical team-sport COD block is 70 percent acceleration-dominant. Resisted sprints, band-assisted cuts, pro-agility variations with a focus on the push-off. Coaches feel good about this because it's measurable and athletes look explosive.
The 2025 systematic review evidence pushes back hard. Braking forces, alongside propulsive forces, were identified as key contributors to faster COD completion times, with the braking component being the under-programmed half of that equation. I've seen this at the high-school level constantly: athletes who test well on a 10-meter acceleration but fall apart on a 5-10-5 because they can't absorb force quickly enough to shorten their deceleration distance.
The fix isn't sexy. It's deceleration-specific loading: drop landings with horizontal velocity, heavy eccentric single-leg work, Nordic curl progressions, and loaded lateral bounds that emphasize the stop, not the jump. Plyometric and resistance training modalities targeting rapid force absorption are exactly what the review recommends to practitioners.
If your athlete looks great in a straight line but bleeds 0.3 seconds every time they hit a 90-degree cut, you don't have a power problem. You have a braking problem. Those are different prescriptions.
How to Screen and Program for Horizontal Braking Impulse
The review recommends braking impulse metrics as priority COD screening variables. Here's what that looks like practically:
- Force plate COD assessments: bilateral and unilateral hop-to-stick protocols measuring peak braking force and contact time at entry. You want both values, not just peak force, because impulse depends on the ratio.
- GPS-derived deceleration events: modern GPS units (10 Hz and above) can flag deceleration intensity and frequency in training. An athlete with low high-deceleration event counts is undertrained in this quality regardless of their sprint speed.
- Modified 505 with split timing: place a timing gate at the turn point, not just at the finish. A slow split to the turn against a fast total time tells you the athlete is compensating with exit acceleration. That's a braking deficit masked by propulsive ability.
For programming, the sequence matters. Build single-leg eccentric strength first (12 weeks of progressive overload minimum), introduce deceleration-dominant plyometrics in week 5 or 6, then add reactive agility only after the athlete can consistently shorten ground contact time under load. Skipping the eccentric base and jumping to agility ladders is the most common programming error I see.
If you're working with athletes returning from lower-limb injury, this framework is the backbone of how I structure the deceleration phase in Comeback Code. Restoring braking capacity before reintroducing full-speed COD isn't optional; it's the whole point.
The Bottom Line on Horizontal Braking Impulse and COD Speed
The Singh et al. systematic review is clear: horizontal braking impulse change of direction performance relationships are strong, consistent, and currently underweighted in most programming. Faster athletes entering a cut at higher velocity, braking harder, and spending less time on the ground are faster out of the cut, not because they push off harder, but because they brake smarter.
Stop building COD programs that look like acceleration programs with a cone at the end. Measure braking capacity, train it with eccentric and plyometric loading, and screen for it with the tools you already have. The research is settled enough to act on.