Two coaches can look at the same week of your training data. One says you nailed 80/20. The other says you missed it by 18 percentage points. Neither coach is wrong. The number moved because the counting method moved. Your training didn’t.
That’s intensity blindness, and it’s why 80/20 polarized training breaks down for most recreational athletes. It isn’t bad discipline. It’s bad accounting, stacked on a target that may be wrong for your training volume. Fix both, and your training is usually closer to right than your watch admits.
How Intensity Blindness Hides in Your 80/20 Split
Picture the same week run through three translators. One says “mostly easy.” Another says “moderate-heavy.” Only the translator changed.
That’s a measured effect, not sloppy logging. A 2023 review by Sperlich, Matzka, and Holmberg pooled 437 elite athletes and 175 training files. It compared two ways of counting the same weeks: by session goal, or by actual time-in-zone. Switching methods alone moved the easy-zone share by 17.6 percentage points. Swap heart rate for power and the numbers move another 12 points.
In short: your “80/20” score depends on which ruler someone hands you. Your training doesn’t have to change at all.
| Accounting method | Zone 1 (easy) | Zone 2 (moderate) | Zone 3 (hard) |
|---|---|---|---|
| Session-goal count | ~75-80% of sessions | ~0-5% of sessions | ~20-25% of sessions |
| Time-in-zone, heart rate | ~17.6 points lower than session-goal | ~4.6 points higher | ~12.6 points higher |
| Time-in-zone, power | ~11.8 points higher than HR time-in-zone | ~11.7 points lower | about the same |
| Real elite median (437 athletes, pooled) | 85% (range 77-90%) | 7% (range 4-11%) | 6% (range 4-11%) |
Even the elite median isn’t 80/20. It’s closer to 85/7/6. Pyramidal training, with more moderate-zone work, was the most common shape in the real data, used by 51% of athletes. Polarized came second at 37%.
Nobody trains a perfect number. The number is a snapshot, not a law.
Even the Landmark Polarized Study Didn’t Hit Its Own Target
The Stöggl and Sperlich 2014 study is the one most 80/20 advice leans on. It split 48 highly trained endurance athletes into four groups for nine weeks. The polarized group gained most: VO2peak up 11.7%, time to exhaustion up 17.4%.
Here’s the part that gets left out. The polarized group’s measured distribution wasn’t 80/20. It was 68% easy, 6% moderate, 26% hard. A supervised lab study couldn’t hit its own target. It still beat a high-volume group at 83/16/1 and a threshold group at 46/54/0.
The lesson isn’t that 80/20 is fake. Mostly easy plus real hard beat the exact ratio. If a supervised lab can’t land its own number, you won’t either.
Execution drift isn’t a discipline problem. It’s the norm.
The Evidence at Low Volume: What Happens Under 10 Hours a Week
Most 80/20 evidence comes from athletes training 10 to 15 hours a week. Most recreational athletes train nowhere near that.
A 2020 trial tested this directly. Thirty-eight recreational runners trained about 3.2 hours a week. A strict polarized group and a much less polarized group both improved their 2km time by about the same amount. The gain was 3.5% versus 3.0%. The less-polarized group used 17% less training time to get there.
A 2019 study of age-group half-Ironman triathletes found the same pattern. Polarized and pyramidal groups finished 2 seconds apart over a 5-hour race.
At population scale, the story holds. A 2024 study by Muniz-Pumares and colleagues pulled data from 119,452 marathon runners. Faster amateurs weren’t hitting a purer 80/20 split. They were running more easy miles. Pyramidal was the dominant shape among the fastest.
| Volume tier | Study | What was tested | What happened |
|---|---|---|---|
| ~3 h/week, recreational | Festa et al. 2020 (n=38) | Polarized (78/3/19) vs. less-polarized (39/49/13) | No real difference; less-polarized group trained 17% less for equal gains |
| ~5-8 h/week, age-group triathlon | Sellés-Pérez et al. 2019 | Polarized (84/4/11) vs. pyramidal (78/19/3) | No real difference; 2 seconds apart over a 5-hour race |
| Well-trained, ~6 sessions/week | Filipas et al. 2022 (n=60) | Pyramidal vs. polarized vs. sequenced blocks | Switching pyramidal into polarized mid-block beat holding either one constant |
| Highly trained, multi-sport | Stöggl & Sperlich 2014 (n=48) | Polarized (achieved 68/6/26) vs. three other distributions | Polarized won clearly despite missing its own 80/20 target |
| Real-world marathoners, all levels | Muniz-Pumares et al. 2024 (n=119,452) | Naturally occurring distribution vs. finish time | Faster runners simply added easy volume; pyramidal was the dominant real-world shape |
Why Your Zone 2 Isn’t Anyone Else’s Zone 2
Even honest time-in-zone data hides a second problem. Two athletes can measure the same way and train different things.
A 2025 study by Meixner and colleagues tested 50 cyclists against different lab markers for the top of Zone 2. The methods disagreed by as much as 29%. One marker put the boundary at 194 watts. Another put it at 224 watts, in the same riders. Consumer zone systems disagree too. Percent-of-max-heart-rate puts easy’s ceiling near 60-70% of max. Percent-of-threshold puts it near 76-85% of threshold. For where that line sits for you, see Zone 2 vs. LT1.
Translation: Zone 2 isn’t one fixed point. It’s wherever a given test draws the line.
80/20 comes from a 3-zone model. Most watches show 5 or 7 zones instead, and they don’t map cleanly onto it. That mismatch is why athletes ask why they always seem to be in Zone 4.
There’s a second problem inside a single session. Heart rate climbs over time at a constant pace, a real pattern called cardiovascular drift, not sloppy pacing. A session that starts easy can finish moderate without the athlete changing anything. Aerobic decoupling covers this mechanism in full. In one study of amateur Ironman triathletes, training looked reasonable at 68% easy. On race day, heart-rate distribution flipped to 59% moderate. Effort drifted under pressure, the same way easy sessions drift under fatigue.
Where the Critique Goes Too Far
None of this means polarized training doesn’t work. The critique is narrower than “80/20 is a myth.” The 2014 Stöggl and Sperlich trial remains strong evidence that mostly-easy-plus-real-hard beats a threshold-heavy default, even with imperfect execution. A 2007 trial found a more-polarized group improved more than a less-polarized one. Filipas’s 2022 trial saw pure polarized beat pure pyramidal on 5K time, though sequencing the two beat both. Even so, 91% of the pooled 437-athlete dataset trained more than 60% of their time at low intensity, in every sport studied. Polarized’s real edge shows up in well-trained, higher-volume, supervised athletes, not necessarily in yours.
What Actually Works Under 10 Hours a Week
- Stop chasing an exact percentage. Aim for mostly easy, real hard, small middle. Every study above missed its own target and still worked.
- Measure time-in-zone for the whole session, not the plan. A workout labeled “easy” isn’t the same as one that stayed easy for 60 minutes. Check the back half.
- Pick one measuring stick and stick with it. Heart rate and power time-in-zone disagree by roughly 12 points on the same data.
- Protect the moderate zone from creeping in. Easy days that drift into moderate cost recovery without building fitness. See the junk-miles trap.
- Use moderate work on purpose, not by accident. Sweet-spot and tempo sessions belong in a plan. See sweet-spot training for where.
- Under 8 hours a week, don’t force strict polarized. The low-volume evidence doesn’t support it over a pyramidal or focused approach.
- Recheck your thresholds before trusting your zones. A 30-watt swing in where Zone 2 starts changes everything downstream.
Take a runner I’ll call Dana, 39, training seven hours a week. Her plan called for 80% easy, 5% moderate, 15% hard. She thought she was hitting it because every easy run was labeled “easy.” That’s intensity blindness. Her real heart-rate time-in-zone told a different story: 61% easy, 29% moderate, 10% hard. Her “easy” long runs were finishing well above her easy heart-rate ceiling. She started backing off pace once heart rate crept up, instead of holding it. Eight weeks later, her split moved to 76% easy, 12% moderate, 12% hard. Her next long-run time trial came in faster at a lower heart rate.
That’s the gap AthleteOS closes. Its session analysis turns real heart rate and power data into an honest time-in-zone number for every workout. Your AI coach compares that to what the plan called for, then says it plainly: “your easy days landed in your moderate zone three times this week.” Just the split you trained, next to the one you meant to.
Deciding which model to build toward? Polarized versus pyramidal training covers that question. This piece is about making sure the number you pick is the number you hit.
See your real time-in-zone breakdown, not your calendar’s guess.