Skip to content
AthleticPeak Pro

Language

Text size

MenuMenu
Language

Text size

Start FreeSign in
Start Free
Back to journal
Methodology

Buying Launch Angles With Speed You Don't Have

Elite jumpers take off at almost the same angle as average ones — the difference is how little speed they lose.

David Brannan Published Nov 4, 2026 10 min read
Buying Launch Angles With Speed You Don't Have

Athens, August 2004. The bar sits at 2.36 meters, and the man who clears it to win Olympic gold stands 1.81.

Stefan Holm heard the verdict his whole career: too short. He grew up in the shadow of Patrik Sjöberg, Sweden's 2.00-meter world-record holder, and next to the jumpers in any warm-up area he looked like he'd wandered in from the 400. Shorter legs. Shorter levers. A lower center of mass to lift.

What Holm had instead was a takeoff. A penultimate step driven into the track like he meant to break it. A plant so short and violent that the whole jump was decided in less time than it takes to blink. Indoors, he went on to clear 2.40 — 59 centimeters over his own head, tying the height-differential record Franklin Jacobs set at the Millrose Games in 1978.

So here is the obvious lesson: launch angle is trainable. You can build it.

And here is the less obvious one, the one that should change how you coach jumpers tomorrow: most athletes are trying to build the wrong thing.

The half-degree that doesn't matter

A 2026 study filmed twenty-four male long jumpers in competition at 100 frames per second and split them into two groups: elites jumping 8.00 to 8.20 meters, and sub-elites jumping 7.80 to 7.99.

If launch angle were the secret, this is where it would show up. It didn't.



Elite (8.00–8.20 m)

Sub-elite (7.80–7.99 m)

Takeoff angle

20.15°

19.59°

Contact time

0.110 s

0.116 s

Horizontal velocity at takeoff

8.63 m/s

7.77 m/s

Speed lost at the board

10.4%

12.1%

Half a degree apart on angle — statistically nothing. Six thousandths of a second apart on contact. But nearly a full meter per second apart on the speed they carried off the board, because the elites arrived faster and gave less of it away.

Read that table again, because it rewrites the job:

The angle isn't the thing you chase. It's the receipt for how much speed you kept.

Two jumpers can leave the board at the same angle and land three-quarters of a meter apart. The difference wasn't in the sky. It was in the last few steps on the ground.

Every degree costs something

At takeoff your body leaves the ground with a single velocity vector. The horizontal part is everything your run-up built. The vertical part is created in the contact — in about a tenth of a second. Launch angle is just the direction of the arrow those two make together.

Which means angle is never free. To leave steeper, you redirect force upward, and you pay for every degree in horizontal speed. Takeoff velocity is a function of touchdown velocity; vertical rises at the direct expense of horizontal.

Think of it as a budget. The run-up deposits the money. The takeoff spends it. And every jumping event has a different idea of what's worth buying.

Long jump: about 20°, and speed is the whole bank account

Forty meters of near-maximal sprinting supply most of the result. Angle is a small correction on a very large number. Over-rotate and you get a beautiful parabola that lands short.

Mike Powell spent years finishing behind Carl Lewis. His fix was never more angle. It was a run-up he could reproduce, so his speed hit the board in the same shape every time. On 30 August 1991 in Tokyo — on a runway he later described as "really, really fast" — it all arrived at once: 8.95 meters, five centimeters past Bob Beamon's record, which had stood for nearly 23 years. Thirty-five years later, it still stands.

Triple jump: the tightest budget in the sport

One run-up has to pay for three takeoffs, and the hop leaves the ground even flatter than a long jump. Elite phase ratios sit near 33% hop, 29% step, 37% jump, with contacts of 0.12 to 0.14 seconds. Overspend early and the step and jump go bankrupt.

Christian Taylor already had an Olympic title when he decided to rebuild. His hop was so big it was eating the rest of the jump. He'd also switched his takeoff leg — a near-unthinkable change at that level — and he chose to take shorter hops and steps than his rivals, deliberately giving away distance he could obviously produce. At the 2015 World Championships in Beijing he jumped 18.21 meters, the second-longest triple jump in history at the time, and took off from roughly 11 centimeters behind the board to do it. He gave away the thing he was best at, and got better.

High jump: 40–50°, and the budget runs the other way

Now you need height over a bar, so the approach curves and shrinks to 9–11 steps. Contact is longer because the job is redirection, not transmission.

This is where Holm's genius was. A shorter athlete has a smaller budget to begin with, so he made the vertical purchase cheaper: huge eccentric strength in the penultimate so the leg didn't buckle, and a contact quick enough that less speed was burned to get the lift. Same physics as everyone else. Better exchange rate.

Pole vault: 17–20°, and the pole does the redirecting

This is the one people get backwards. You don't redirect for the pole; the pole redirects you. Take off too steeply and you plant the pole under yourself, stall it, and drain the energy the whole vault runs on.

Sergey Bubka, working with coach Vitali Petrov, pushed the idea to its limit with the free takeoff: leaving the ground just before the pole tip hits the back of the box, so the braking shock is minimized and the pole is loaded by a body still running at full speed. He broke the world record 35 times — 17 outdoors and 18 indoors — usually by a single centimeter. Almost nobody since has managed his model consistently. But every vaulter can borrow its principle: keep the speed, let the pole spend it.

Everything below happens in roughly a tenth of a second. Each link exists to produce vertical impulse without bleeding horizontal speed.

  • Glutes drive. Hip extension is the main source of vertical velocity. This is the engine.
  • Quads hold. On the plant, the knee wants to fold. The quads absorb that load eccentrically, then extend. They're the strut. A weak strut means longer contact and lost speed.
  • Soleus stiffens. The ankle is the last joint in the chain and gets the least time. It doesn't need to move much; it needs not to give. A soft ankle leaks the force the hips just made.
  • Free leg swings. The opposite leg drives forward and up. It's a real contributor to vertical impulse, not decoration.

Launch angle is a hip-and-knee problem with an ankle constraint. And here's the bargain hiding in plain sight: most athletes train the ankle least, yet it's the cheapest link to fix and the one most likely to be leaking speed.

Training the budget: speed, snap, stiffness, preservation

Reaction time doesn't matter at takeoff — nobody is reacting to a gun. What matters is reactivity: how much force you can produce inside a 0.11-second window.

Speed — raise the budget. Full-recovery sprints at 95–100%, 30–60 meters, never tired. This is the highest-value work in the plan, and it's the one variable the long-jump data says actually separates good from great.

Snap — spend it fast. Rate of force development: weighted jumps, trap-bar jumps, short heavy sled pushes. Loaded, fast, low volume.

Stiffness — stop the leaks. Depth jumps with a contact-time target. Stiff-ankled pogo hops. Heavy seated calf raises, 4–5 sets of 4–6. Tempo squats with a three-second lowering. Heavy isometrics for the Achilles, which have some of the best evidence for stiffening tendon — expect it to take a full training block of 8–12 weeks, not a few sessions.

Preservation — the actual skill. Timed penultimate steps. The cue tall–fast–tall: tall on the penultimate, fast into the board, tall at takeoff. Film the last three steps at 240 fps and count contact frames. Contact getting longer as you approach the board? You're sitting. Foot landing well ahead of the hips? You're reaching. Both cost speed, and both are right there on video.

Measure it, or you're guessing

Two numbers tell you whether any of this is working.

Contact time at takeoff, from 240 fps video. As a working rule, anything above about 0.13 s in the long jump usually means the leg is collapsing.

Velocity loss from penultimate to takeoff, from timing gates or a phone app with splits. The elite group in the study still lost about 10%. My rule of thumb: if you're losing more than 14%, you're buying angle with speed you can't spare.

Check your angle monthly with video and an overlay. If the angle went up and the distance went down, you over-bought.

A three-week block

Speed on fresh legs. Reactive work after speed. Heavy strength after that. Never the reverse.

Day 1 — Speed. 4 × 40 m at 95%, full recovery. Then 3 × 3 penultimate-to-takeoff drills from a five-step approach, filmed.

Day 2 — Reactive. Depth jumps 4 × 4, contact under 0.20 s. Pogo hops 3 × 15. Heavy sled push 4 × 10 m.

Day 3 — Strength. Tempo back squat 4 × 4 at 3-1-X. Heavy seated calf raise 4 × 6. Nordics 3 × 5.

Day 4 — Speed. 3 × 30 m at 100%. Four full approaches through the gates.

Run it for three weeks. Film weeks one and three and compare velocity-loss percentage. If the number hasn't moved, the problem is contact mechanics, not strength: go back to the penultimate drill and cut the volume.

What the best of them knew

Holm never got taller. He built a takeoff that spent less and bought more, and it was worth Olympic gold against men built for the event.

Taylor gave away hop distance and jumped farther. Bubka left the ground before the pole could slow him. Powell stopped chasing a bigger jump and built a run he could repeat.

None of them turned the angle dial up. All four renegotiated the budget.

So fix speed first and angle last. Because adding angle to a slow approach is buying something you can't afford — and the best jumpers in the world don't overpay.


Sources

Research on jump mechanics

  • Wei, Yuan & Xie (2026). Kinematic determinants of take-off performance in elite and sub-elite male long jumpers.
  • Linthorne. Biomechanics of the long jump.
  • Linthorne. Optimum take-off angle in the long jump.
  • McGinnis. Mechanics of the pole vault take-off.
  • Kinematics of the final approach and take-off phases in world-class pole vaulters.
  • Biomechanical evaluation of the phases of the triple jump take-off in a top female athlete.
  • Effects of phase ratio and velocity conversion coefficient on triple jump performance.
  • Biomechanical aspects of technique of high jump.
  • Bohm et al. (2023). Speed-specific contractile conditions of the soleus from slow to maximum running speed.

Coaching practice

  • Track & Field News: The biomechanics of eccentric and concentric actions in the high jump.
  • Track & Field News: Horizontal jumps training and design.
  • Track & Field News: In defense of the free takeoff.
  • Jacoby. Pole vault concepts — speed, plant and takeoff.
  • SimpliFaster: The forgotten plantar flexor — training the soleus (Brandon Pigg).
  • SimpliFaster: Repairing and reloading the running athlete (Colin Griffin).

Athletes

About the author
David Brannan

Keep building your training record.

Start with a free athlete account. Train independently or connect with a coach, and keep your completed history.

Explore your athlete account

Get new articles by email

Occasional notes on training, coaching and athletic performance. No daily noise.

Share this article