You’ve probably heard that swimmers need threshold work, cyclists need pyramids and runners need polarized weeks. The best pooled evidence doesn’t back that up. What moves results more is your level and how long your block lasts.
Here’s what the data actually says, and how to build a sane week across all three sports.
Intensity Distribution for Triathletes: What the Labels Mean
Your intensity distribution is how your training time splits between easy, moderate and hard. Coaches use three buckets.
- Polarized: mostly easy, a small slice hard, very little in the middle.
- Pyramidal: lots of easy, a medium chunk moderate, a small hard tip.
- Threshold: a big share of time near your threshold pace.
Here’s the catch. The same week can wear two labels. A 2025 narrative review in Frontiers in Physiology noted that counting sessions makes a week look polarized, while counting minutes makes it look pyramidal. In short: the label depends on how you count.
Picture a week with eight sessions. Seven are easy and one is a hard interval day. Counted by session, that’s mostly easy with a hard slice, so it looks polarized. Now count minutes. The easy days are long and the hard day is short, but a few moderate minutes hide inside warm-ups and long rides. Counted by time, the middle bucket grows and the week looks pyramidal. Same week, two labels.
Your zones matter too. Heart rate, power and pace each split the effort a little differently. So does a three-zone model versus a five-zone one. The review’s authors said that measurement method changes classification.
So when someone says “runners are polarized,” ask who counted what. One marathon study showed 78/4/18% (low, moderate, high). Another running study showed 71/21/8%. Same sport, two shapes.
Does Any Meta-Analysis Show a Sport-by-Sport Winner?
No. The 2024 meta-analysis in Sports Medicine pooled 14 studies and 437 participants. It reported no pooled effect by sport.
What it did find: polarized training beat other models for VO2peak (your engine’s peak oxygen use), with an effect size of 0.24. For time-trial performance, the difference was -0.01. Translation: a bigger engine on paper, no faster race on the clock.
Look at who was studied. Runners made up 6 studies, cyclists 4, triathletes 2, and swimmers just 1. The swim and triathlon evidence base is thin.
A 2025 network meta-analysis pooled individual data from 348 athletes. Polarized versus pyramidal showed no real gap in VO2max (-0.06) or time trials (-0.05). The 2025 Frontiers review reached the same place: no single model is universally superior. (It’s a narrative review, not a meta-analysis, and it has no swim-specific data.)
What Moves the Result: Block Length and Athlete Level
Two things changed the answer more than sport did.
The polarized edge was clear in short blocks and gone in long ones. It was also bigger in highly trained athletes than in developing ones. In short: the fancy model helps most when you’re already fit and the block is brief.
The 2025 analysis pointed the same way. Competitive athletes may benefit from polarized work. Recreational athletes may do better pyramidal. The gap between those groups was -0.63. That should matter to you if you’re an age-grouper.
Think of it like a kitchen. Swim, bike and run are three different pans. The recipe is mostly the same: a lot of easy base, a little spice. Level and block length change how much spice you need.
Hard truth: your level beats your sport.
What Swimming Actually Shows
Swimmers often train threshold-heavy or pyramidal. That’s a description of what they do. It isn’t proof of what works.
The best direct trial points the other way. Pla and colleagues put 22 elite junior swimmers through a crossover with two 6-week blocks. Polarized training was 81/4/15%. Threshold training was 65/25/10%.
Polarized swimmers also reported less fatigue and better recovery. Translation: the harder-middle approach cost more and paid less. One small trial can’t settle it, but it breaks the “swimming loves threshold” story.
Triathlon Data: Useful, Small, Observational
Munoz and colleagues tracked 9 Ironman athletes for 18 weeks. Training was about 68% easy time. More easy time linked to faster races (r = -0.92). More time in the middle zone during the race linked to slower ones. Nine athletes is a tiny group, so treat it as a hint.
In 18 recreational half-Ironman athletes, polarized (84.5/4.2/11.3%) and pyramidal (77.9/18.8/3.3%) groups both improved in all three segments. Both worked. See our Zone 2 science breakdown for why that easy base does the heavy lifting.
| Sport | Typical pattern | Best-supported finding | Evidence quality |
|---|---|---|---|
| Swim | Often threshold or pyramidal | Polarized beat threshold for 100 m (22 swimmers) | One small crossover trial |
| Bike | Mostly easy, 2-8% hard in one cited summary | No clear sport-specific winner | Descriptive plus mixed trials |
| Run | Polarized or pyramidal, by how you count | Mixed results | Descriptive plus mixed trials |
| Triathlon | Mostly easy time | More easy time linked to faster races | Two small observational studies |
A Practical Triathlon Week (Coaching Practice, Not Proof)
Nothing above proves this layout is best. It’s a reasonable way to apply the evidence. The core rule: keep roughly 75-80% of your time easy in every sport. Where you place the hard work is a coaching call, mostly about injury load.
| Day | Session | Intensity |
|---|---|---|
| Mon | Swim: technique and easy aerobic | Easy |
| Tue | Bike: threshold or hard intervals | Hard |
| Wed | Run: easy | Easy |
| Thu | Swim: short fast sets | Hard |
| Fri | Run: easy, or rest | Easy |
| Sat | Long ride | Easy |
| Sun | Long run, or a short run after the ride | Easy |
Why the bike gets more hard work: it’s gentler on your legs than running. Pounding out hard run miles costs more tissue. That’s a coaching rationale, not a trial result.
Let’s do the math for a 70.3 athlete with 9 hours a week. That’s 540 minutes. At 80% easy, about 432 minutes are easy and 108 are hard. Roughly 60 of those hard minutes could go on the bike, 30 in the pool and 18 on the run. That split is a coaching choice, not a finding. It keeps the run, where the legs take the most pounding, as the lightest hard slice.
Notice the same 80% holds in every sport. Your swim can be 80% easy. So can your ride and your run. You don’t need three different recipes. You need one honest count.
Take a fictional athlete, Dana, 41, training for a 70.3 on 9 hours a week. She did two hard runs, two hard rides and a hard swim. She felt flat by week six. She cut the hard sessions to one ride and one swim a week, and kept the rest easy. Her easy share rose from about 60% to about 80%. These numbers are made up to show the idea, not research results.
Check your own split with a long, steady session. A rising drift ratio on easy days means they’re not easy enough.
AthleteOS tracks your training load in swim, bike and run and adapts your training plan when a week goes sideways. Pair that with how training load builds fitness and fatigue to see where your minutes really went. You can start a free trial here.
What About Swapping One Sport for Another?
A 2026 meta-analysis in Frontiers in Sports and Active Living looked at run versus cycle cross-training across 7 trials. Running performance showed almost no difference (Hedges g = 0.02), but the authors said it does not prove the sports are interchangeable. And that study covered only running and cycling, so it says nothing about trading swim time for bike time.
Pick a mostly easy base. Add a little hard work. Repeat.