Pair a heavy back squat with a set of bounding sprints. Rest four minutes. Repeat. Do that twice a week for eight weeks. Your running economy can improve by 7 to 16 percent. That’s roughly triple what heavy lifting or plyometrics deliver alone (Li et al. 2019; Yu et al. 2025).
That’s complex training, sometimes called contrast training. Lift heavy, rest just long enough, then move fast. The pairing does something neither half can do by itself.
What Complex Training Actually Is (It’s Not a Circuit)
Complex training pairs one heavy resistance exercise with one plyometric exercise that uses a similar movement pattern. A back squat pairs with a depth jump. A trap-bar deadlift pairs with bounding. You lift heavy first, usually 85% or more of your 1RM (the heaviest weight you can lift once with good form). Then you rest. Then you move explosively.
Think of the heavy lift as winding a spring tight. The plyometric drill is the release. Rest too short and the spring is still coiled with fatigue, so the jump comes out flat. Rest too long and the spring has already unwound, so there’s nothing extra to release. Hit the right window and the release is sharper than it would’ve been on its own.
A circuit is different. Circuits move from exercise to exercise with little or no rest. They build general fitness. They aren’t built to trigger the mechanism complex training depends on.
The Research on Complex Training for Distance Runners
Two trials tested this directly in runners, and both point the same direction.
A 2019 PeerJ study tested well-trained male distance runners over 8 weeks. They did one of three things: complex training, heavy resistance training alone, or no strength work. Running economy at 14 km/h improved 7.68% in the complex-training group versus 4.89% for resistance-only. At 16 km/h, complex training improved economy 7.38% while resistance-only barely moved and wasn’t statistically significant. Five-kilometer time trials improved too. Complex training gained 2.80%, resistance-only gained 2.09%, and the control group barely changed at 0.07% (Li et al. 2019).
A 2025 Frontiers in Physiology trial in adolescent distance runners found an even bigger gap. Running economy at 14 km/h improved 16.1% with complex training versus 2.8% with resistance training alone. At 12 km/h, it was 7.8% versus 3.5% (Yu et al. 2025). In plain terms: pairing lifting with jumping worked far better. Those runners got two to five times the economy improvement of runners who just lifted.
A 2024 meta-analysis pooling 31 studies and 652 runners backs this up at a bigger scale.
A separate meta-analysis of 22 studies looked at plyometrics alone. It barely moved the needle: g = -0.13, a trivial effect whose confidence interval crosses zero (Eihara et al. 2022). Heavy lifting alone did better, g = -0.32, but still small. Running the two together is where the real signal shows up.
Running economy isn’t a nice-to-have number either. It and your speed at VO2max together explain roughly 94% of the variance in a 16.1 km race result. Economy itself can vary by up to 30% between runners of similar caliber (Barnes & Kilding 2015). A few percent of improvement is a real edge, not noise.
The Physiology: Why Rest Timing Makes or Breaks It
The mechanism is called post-activation potentiation, or PAP. A heavy contraction primes your nervous system. The next explosive movement fires harder as a result. Translation: your muscle “wakes up” after a heavy set. For a short window, it’s stronger and faster than normal.
Foundational PAP research backs this up. Power output and jump height peaked around 8 minutes after a heavy set (Kilduff et al. 2008). Large effects (1.34-1.48) showed up between 4 and 8 minutes post-lift, when the load was heavy enough. That means around 93% of 1RM (Lowery et al. 2012).
Both direct-to-runner trials used a 4-minute rest between the heavy set and the plyometric set. They chose that timing because it sits inside the potentiation window (Li et al. 2019; Yu et al. 2025).
Too short, and residual fatigue from the heavy lift drags the jump down. Too long, and the potentiation has already faded. The window is roughly 3 to 8 minutes, narrowing to 3-4 minutes in the two studies done specifically on runners.
Rest interval isn’t a detail. It’s the mechanism.
Complex Training vs. Plyometrics Alone vs. Heavy Lifting Alone
Here’s what changes at different running speeds. This compares complex training directly against resistance training alone, using the 2025 adolescent-runner data.
| Method | Evidence base | Effect on economy | Typical RE change | Rest interval used |
|---|---|---|---|---|
| Heavy resistance alone | 11-13 studies, n=216-19 effect sizes | Small (ES -0.266, g -0.32) | 2.8-4.9% | Not applicable |
| Plyometrics alone | 9-11 studies, n=263-28 effect sizes | Small-to-trivial (ES -0.307, g -0.13) | Inconsistent; often n.s. at faster speeds | Not applicable |
| Complex training | 2 direct RCTs (N=28, N=32) + 9 pooled studies | Moderate (ES -0.426) | 7.4-16.1% | 3-4 minutes |
The same 2025 trial found more. The complex-training group also gained more reactive strength (+26.4% Reactive Strength Index vs. +7.4%). It also dropped blood lactate at submax speeds. Meanwhile, 1RM squat gains were nearly identical between groups. The economy advantage didn’t come from lifting more weight. It came from the pairing itself.
Who Should Try This (and Who Should Wait)
This isn’t a first-strength-block move. The PAP research is clear on training age. Athletes with 2+ years of consistent resistance training get a bigger potentiation response. They can also handle the 85-90% 1RM loads the protocol calls for. Beginners often don’t have the neuromuscular base yet, and fatigue can dominate instead of potentiation.
Take a runner I’ll call Marcus, 33, training for a sub-2:50 marathon at 45 miles a week. He’d lifted twice weekly for three years but never paired it with jump work. Over an 8-week block modeled on the Li 2019 protocol, he added a new pairing twice a week. Back squats at 85% 1RM, four minutes of rest, then depth jumps. His coach tracked submax heart rate at marathon pace as an economy proxy. It dropped from 158 to 149 bpm at the same pace by week 8. That’s a meaningful shift for a runner who’d plateaued for a year. He ran a 6-minute marathon PR that fall.
That’s one runner, not a guarantee. But it tracks with what the trials found.
The Honest Caveats
This is where most articles stop and just tell you to go do it. Here’s what the research actually doesn’t prove yet.
Distance runners are a small slice of the complex-training evidence. A 2024 scoping review covered 68 studies and 1,821 total participants. Runners made up just 5.9% of the literature, versus 27.9% for soccer players. Most of what we know about complex training comes from field-sport athletes, not distance runners.
Rest-interval reporting is inconsistent across the field. Only 57.4% of the 68 studies even reported the rest interval used. Among those that did, it ranged from 0 seconds to 8 minutes. The 3-5 minute window is the best current estimate, not a settled number.
The runner-specific trials were also small: N=28 and N=32, both 8 weeks. Long-term durability, and whether the effect holds for masters or female runners, isn’t established. Only 145 of the 1,821 total participants across the whole evidence base were women.
Evidence quality here is honestly low-confidence. That’s not a reason to skip complex training. It’s a reason to individualize the load and watch how your body responds. Don’t blindly copy a protocol for eight weeks without checking whether it’s actually working for you.
How AthleteOS Programs Complex Training Automatically
This is the exact gap a static article can’t close. You can read the protocol. But you can’t tell from a PDF whether the 3-to-5-minute window and 85% load are right for your training age. You also can’t tell whether it’s actually moving your economy.
AthleteOS builds complex training into your training plan as its own session type. It pairs a heavy compound lift with a matched plyometric drill. A built-in, research-aligned rest timer sits between the pair, so you’re not guessing at four minutes with a stopwatch. Across the block, session analysis tracks a running economy proxy: submax heart rate or pace-at-effort at matched intensity. That way you can see whether the 7-16% improvement the studies found is showing up in your own numbers. You’re not just trusting the protocol for two months.
Two things are worth reading before you start. First: how to combine strength and endurance training without the interference effect eating your gains. Second: why lifting the day after a hard run can backfire. For economy gains from the biomechanics side, check two more levers: ground contact time and cadence and vertical oscillation. You can start a training plan that schedules this automatically.
Complex training won’t fix a running economy problem that’s really a mileage problem. But if your lifting has plateaued and your jumps live in a separate session from your squats, this is the change with the biggest research backing behind it.