Polarized training is an endurance training intensity distribution in which about 75 to 80 percent of sessions stay below your first lactate threshold, most of the rest go above your second threshold, and very little time is spent between the two. It is often shortened to "80/20," a label that hides whether the 80 percent counts sessions or minutes.
The idea comes from training logs of elite endurance athletes, who turned out to spend surprisingly little time at the comfortably hard pace most amateurs default to. If your easy days drift into that middle band, you pile up fatigue without getting the adaptations of either very easy or very hard work. Checking your own distribution against a polarized target takes heart rate zone data from workouts, which Apple Watch already records.
01Three zones bounded by two thresholds
Polarized training is defined on a three-zone model. Zone 1 sits below the first lactate threshold or first ventilatory threshold, where blood lactate stays close to resting levels. Zone 2 runs from there to the second threshold, near maximal lactate steady state and the anaerobic threshold. Zone 3 is everything above the second threshold.
This research zone 2 is a different intensity from the "zone 2" in podcasts and on five-zone watch faces. On a five-zone heart rate zones scale, the popular zone 2 usually sits inside research zone 1. Research zone 2 is the band that polarized training keeps small, the tempo and threshold effort that feels productive and leaves you tired the next day.
02Seiler's observations in elite athletes
Stephen Seiler and Gro Kjerland tracked 11 nationally competitive junior cross-country skiers for 32 consecutive days in a study published in 2006. They recorded 318 endurance sessions with continuous heart rate. By heart rate, 75 percent of sessions fell in zone 1, 8 percent in zone 2 and 17 percent in zone 3. Session RPE gave almost the same split, 76, 6 and 18 percent. Blood lactate from 60 consecutive sessions agreed, with 71 percent at or below 2.0 mmol/L and 22 percent above 4 mmol/L (Seiler and Kjerland, 2006). The authors called the pattern polarized and concluded that elite endurance athletes "train surprisingly little at the lactate threshold intensity."
Seiler's 2010 review pulled together descriptive studies of nationally and internationally competitive athletes training 10 to 13 times a week. They converged on about 80 percent of sessions at low intensity, around 2 mmol/L blood lactate or below, with about 20 percent built around high-intensity work such as intervals near 90 percent of VO2 max. The same review found no convincing evidence that adding more high-intensity work to already well-trained athletes produced long-term gains (Seiler, 2010).
03Sessions or minutes
The 80/20 number changes meaning depending on what you count.
The session-goal method labels each session by its purpose. A 60-minute interval workout with a 20-minute warm-up, 16 minutes of hard reps and a cool-down counts as one hard session. The time-in-zone method counts minutes by heart rate, so the same workout contributes mostly easy minutes and a few hard ones.
Sylta and colleagues compared the two directly on 570 sessions from 29 elite cross-country skiers during an altitude camp. By time in zone, 96.1 percent of training fell in zone 1 and 4.0 percent in zones 2 and 3 combined. By session goal, 86.6 percent of sessions were zone 1 and 13.5 percent were zone 2 or 3. The conversion factor for the high-intensity range was about 3.0, which means a hard session share of 15 percent corresponds to roughly 5 percent of minutes (Sylta et al., 2014).
Tønnessen's study of 11 Olympic and world champion skiers and biathletes shows the same gap over a full year. About 91 percent of endurance training time was low intensity, while 23 percent of sessions were classified as high intensity (Tønnessen et al., 2014). Rosenblat and colleagues note that Seiler and Kjerland's 75, 8 and 17 percent session split becomes 91, 6 and 3 percent when the same data are counted as heart rate time in zone (Rosenblat et al., 2025).
For your own training, a polarized week counted by session is close to 90/10 when counted by minutes. If your watch shows 80 percent of your minutes below the first threshold, you may be doing more hard or moderate work than the elite athletes whose logs defined the term.
04Polarized, pyramidal and threshold models
Three distributions come up repeatedly in the research. All three are described by the share of training in each of the three zones.
| Model | Shape | Typical split by time, zones 1, 2, 3 | Example |
|---|---|---|---|
| Polarized | Most in zone 1, then zone 3, least in zone 2 | about 80, 5, 15 | Neal et al. (2013) polarized group, 80, 0, 20 |
| Pyramidal | Most in zone 1, less in zone 2, least in zone 3 | about 80, 15, 5 | Elite distance runners in preparation (Casado, 2022) |
| Threshold | Zone 2 larger than in either model above | 35 percent or more in zone 2 | Muñoz et al. (2014) between-thresholds group, 46, 35, 19 |
Casado and colleagues set the threshold model boundary at more than 35 percent of volume in zone 2, while acknowledging that the cut-offs have not reached full consensus (Casado et al., 2022).
Treff and colleagues proposed a single number to separate polarized from non-polarized training. The polarization index is PI = log10((zone 1 / zone 2) × zone 3 × 100), with each zone entered as a fraction of total training. A distribution counts as polarized when zone 1 is larger than zone 3, zone 3 is larger than zone 2, and the index exceeds 2.00. An 80, 5, 15 split gives log10(240), about 2.38 (Treff et al., 2019). An 80, 15, 5 pyramidal split gives about 1.43.
05What the randomized trials found
Esteve-Lanao and colleagues ran one of the first controlled comparisons in 12 subelite runners over five months. Both groups kept zone 3 near 8 percent of training time. One group trained 80.5 percent in zone 1 and 11.8 percent in zone 2, the other 66.8 and 24.7 percent. The group with more easy training improved a simulated 10.4 km cross-country race by 157 seconds, against 121.5 seconds for the group with more zone 2 (Esteve-Lanao et al., 2007).
Neal and colleagues gave 12 trained cyclists two six-week blocks in random order. The polarized block was 80, 0 and 20 percent of time across the three zones at 6.4 hours a week. The threshold block was 57, 43 and 0 percent at 7.5 hours a week. The polarized block produced larger gains in peak power output, 8 against 3 percent, and in high-intensity exercise capacity, 85 against 37 percent (Neal et al., 2013).
Stöggl and Sperlich randomized 48 well-trained runners, cyclists, triathletes and cross-country skiers, with a mean VO2 peak of 62.6 mL/kg/min, to nine weeks of polarized, threshold, high-intensity interval or high-volume training. The polarized group trained 68, 6 and 26 percent of time across the zones. It gained 11.7 percent in VO2 peak, 17.4 percent in time to exhaustion and 5.1 percent in peak velocity or power. The threshold and high-volume groups showed no significant change in VO2 peak (Stöggl and Sperlich, 2014).
Muñoz and colleagues tested the idea in 30 recreational runners over ten weeks. The polarized group, at about 77, 3 and 20 percent, improved 10 km time by 5.0 percent. The between-thresholds group, at about 46, 35 and 19 percent, improved by 3.6 percent. The difference was not statistically significant. Among the runners who most closely followed their assigned distribution, the polarized advantage was 1.29 standardized effect size units (Muñoz et al., 2014).
Filipas and colleagues compared polarized with pyramidal training in 60 well-trained male runners over 16 weeks, holding training load constant. Every group improved. The group that trained pyramidal for eight weeks and then polarized for eight improved the most, about 1.5 percent in 5 km time and 3.0 percent in relative VO2 peak (Filipas et al., 2022).
06What the meta-analyses found
Rosenblat, Perrotta and Vicenzino pooled randomized trials of polarized against threshold training in 2019. Only four studies met the criteria and three entered the meta-analysis. Each scored 4 or 5 out of 10 on the PEDro quality scale. The pooled effect on time trial performance favored polarized training, with an effect size of -0.66 and a 95 percent confidence interval of -1.17 to -0.15 (Rosenblat et al., 2019).
Casado and colleagues reviewed ten studies of highly trained and elite distance runners in 2022. These runners typically followed a pyramidal distribution, with marathoners more pyramidal and 1500 m runners more polarized. The review recommends a shift from pyramidal training in the preparatory and precompetitive periods toward polarized training in the competitive period (Casado et al., 2022).
Silva Oliveira and colleagues pooled 17 studies with 437 participants in 2024. Polarized training produced a small advantage in VO2 peak over other distributions, a standardized mean difference of 0.24 with a 95 percent confidence interval of 0.01 to 0.48. The advantage was larger in interventions shorter than 12 weeks and in highly trained athletes. Time trial performance did not differ, with a standardized mean difference of -0.01 (Silva Oliveira et al., 2024).
Rosenblat and colleagues, with Seiler as senior author, then pooled individual participant data from 13 studies and 348 athletes in a network meta-analysis published in 2025. They reclassified each athlete by the distribution actually completed according to heart rate, since the prescribed and executed distributions often differed. Polarized and pyramidal training produced no difference in VO2 max or time trial performance. The mean VO2 peak difference was -0.11 mL/kg/min, well within normal test-to-test error. Performance level changed the picture. Competitive athletes tended to gain more VO2 max with polarized training, and recreational athletes tended to gain more with pyramidal training (Rosenblat et al., 2025).
The same authors reexamined the 2019 meta-analysis. Three of its pooled studies had "threshold" groups that actually followed a pyramidal distribution, so the earlier result actually compared polarized with pyramidal training (Rosenblat et al., 2025).
Burnley, Bearden and Jones argued in 2022 that no evidence shows a specifically polarized distribution is optimal. They point out that elite athletes train pyramidally when training is counted by time, and they recommend prioritizing zones 1 and 2 over zone 3 (Burnley et al., 2022).
The consistent finding across these studies is that a large base of zone 1 training beats a program built mostly on zone 2. The difference between polarized and pyramidal training is small and depends on the athlete, and the most direct trial favored doing pyramidal work first and polarized work closer to competition.
07Who it suits
The strongest support comes from trained and competitive endurance athletes. The benefit showed up in Stöggl and Sperlich's athletes averaging 62.6 mL/kg/min, in the competitive subgroup of Rosenblat's 2025 analysis, and in the highly trained subgroup of Silva Oliveira's analysis.
For recreational athletes, the evidence points to a large share of easy training with some freedom in how the rest is split. Muñoz found no significant difference in recreational runners, and Rosenblat's recreational subgroup, with a mean VO2 peak of 53 against 67 mL/kg/min in the competitive group, leaned toward pyramidal training.
If you train four times a week, one interval session is already 25 percent of your sessions. A strict 80/20 split by session would allow less than one hard day a week, which limits how the rule applies at low volume. In that situation, count by minutes and keep the easy sessions truly easy.
A polarized block also fits the weeks before a goal race, based on Filipas's sequencing result, Casado's periodization advice and the tendency of Tønnessen's champions to train more polarized in the competition phase.
08Measuring your distribution with a wearable
A heart rate monitor gives you time in zone, the method most research on executed training uses. Its accuracy depends on two things, the sensor and the zone boundaries.
Optical wrist sensors vary by activity. Gillinov and colleagues tested wrist-worn monitors against ECG in 50 adults. A chest strap agreed best, with a concordance of 0.996. Apple Watch came next at 0.92 across all conditions. All devices except one performed well on the treadmill, and none were accurate on an elliptical with moving arm levers (Gillinov et al., 2017). Running produces cleaner data than activities with heavy wrist or arm movement.
Zone boundaries are the second source of error. Apple's Workout app shows five zones that it calculates as percentages of your maximum heart rate from your health data (Apple, n.d.). Percent-of-max zones do not locate your thresholds. The boundary between research zones 1 and 2 varies from person to person, and a fixed percentage can place a runner's easy pace in research zone 2 without any warning. Test your first threshold with a lab test or a heart rate drift test, as described in the endurance training guide, and set your zones from it.
Heart rate also lags effort. It takes time to rise at the start of each interval, so short repetitions record fewer minutes in the top zone than the effort deserves. This is part of why the session-goal and time-in-zone numbers differ by roughly a factor of three for hard training.
09How Titan shows your distribution
Titan sorts workout heart rate into five zones set as percentages of your max heart rate. The default zones start at 53, 74, 81, 87 and 93 percent. Unless you set a custom max, Titan uses the higher of your highest heart rate recorded in Apple Health over the last 30 days and 220 minus your age. You can edit each zone's starting percentage. Tap your avatar on Today to open You, open Your Training, and under Heart Rate tap Heart Rate Zones. Titan applies its own zones to the heart rate samples in Apple Health, so its zone minutes can differ from the Workout app on your watch. See heart rate zones for the full settings.
Titan's three training load groups line up with the three research zones.
| Titan group | Titan zones | Research zone when your zones are calibrated |
|---|---|---|
| Low aerobic | Zones 1 and 2 | Zone 1, below the first threshold |
| High aerobic | Zone 3 | Zone 2, between the thresholds |
| Anaerobic | Zones 4 and 5 | Zone 3, above the second threshold |
The match holds only if your Zone 3 starts near your first threshold and your Zone 4 starts near your second.
The Heart Rate Zones card on Today shows how many workout minutes you spent in each zone that day. Tap it to see each zone's share, then tap a zone to see which workouts added time to it.
For a longer view, open Trends, which starts on Year, and choose 30 Days, 60 Days, 6 Months or Year from the period menu. The Cardio distribution chart shows weekly minutes in Zones 2 through 5. It leaves out Zone 1, so your easy share looks smaller there than it is.
Below the Training load chart, Training load focus splits the period's training load into Low aerobic, High aerobic and Anaerobic, each with its load and share. Each workout minute is weighted by its zone number, so a Zone 5 minute counts five times as much as a Zone 1 minute. Load shares therefore run lower for easy training than time shares. Only heart rate recorded during workouts counts.
A worked example
Take a 360-minute week split 115, 173, 18, 38 and 16 minutes across Zones 1 to 5. By time that is 80 percent easy, 5 percent moderate and 15 percent hard, a polarization index of 2.38. Training load focus shows about 62 percent Low aerobic, 7 percent High aerobic and 31 percent Anaerobic.
Now let the easy runs creep uphill and the long run finish at tempo. The same 360 minutes lands at 90, 150, 90, 25 and 5. By time that is about 67, 25 and 8 percent, a polarization index near 1.35. Training load focus shows about 50, 34 and 16 percent.
Low aerobic is the largest group in both weeks, so the top row alone does not tell you which distribution you trained. Compare the two smaller rows. Anaerobic above High aerobic points to a polarized period. High aerobic above Anaerobic points to pyramidal or threshold training. The drifted week also carries 785 load against 747 for the same hours.
10Common mistakes
Letting easy days drift. Rosenblat's 2025 analysis found that athletes assigned to low-intensity programs often accumulated more zone 2 heart rate time than planned. Hills, heat and group runs push easy sessions into the middle zone. Cap easy sessions at the top of your calibrated Zone 2 and slow down or walk steep grades to stay under it.
Counting sessions and minutes interchangeably. If you plan by session and check by minutes, a correct polarized week will look too easy. Pick one method and compare against the matching target, about 80/20 by session or about 90/10 by time.
Using uncalibrated zones. Percent-of-max zones built on 220 minus age can shift your research zone boundaries by many beats. Set a custom max if you know it, and move the zone starts to your measured thresholds.
Treating 80/20 as a guarantee. The trials show a large easy base works better than threshold-heavy training. No trial has identified one split that is best for everyone, and the gap between polarized and pyramidal training is small.
11References
- Apple (n.d.). View Heart Rate Zones on Apple Watch. Apple Watch User Guide. https://support.apple.com/guide/watch/view-heart-rate-zones-apd897dccddf/watchos
- Burnley M, Bearden SE, Jones AM (2022). Polarized training is not optimal for endurance athletes. Medicine and Science in Sports and Exercise 54(6):1032-1034. https://doi.org/10.1249/MSS.0000000000002869
- Casado A, González-Mohíno F, González-Ravé JM, Foster C (2022). Training periodization, methods, intensity distribution, and volume in highly trained and elite distance runners, a systematic review. International Journal of Sports Physiology and Performance 17(6):820-833. https://doi.org/10.1123/ijspp.2021-0435
- Esteve-Lanao J, Foster C, Seiler S, Lucia A (2007). Impact of training intensity distribution on performance in endurance athletes. Journal of Strength and Conditioning Research 21(3):943-949. https://doi.org/10.1519/R-19725.1
- Filipas L, Bonato M, Gallo G, Codella R (2022). Effects of 16 weeks of pyramidal and polarized training intensity distributions in well-trained endurance runners. Scandinavian Journal of Medicine and Science in Sports 32(3):498-511. https://doi.org/10.1111/sms.14101
- Gillinov S et al. (2017). Variable accuracy of wearable heart rate monitors during aerobic exercise. Medicine and Science in Sports and Exercise 49(8):1697-1703. https://doi.org/10.1249/MSS.0000000000001284
- Muñoz I et al. (2014). Does polarized training improve performance in recreational runners? International Journal of Sports Physiology and Performance 9(2):265-272. https://doi.org/10.1123/ijspp.2012-0350
- Neal CM et al. (2013). Six weeks of a polarized training-intensity distribution leads to greater physiological and performance adaptations than a threshold model in trained cyclists. Journal of Applied Physiology 114(4):461-471. https://doi.org/10.1152/japplphysiol.00652.2012
- Rosenblat MA, Perrotta AS, Vicenzino B (2019). Polarized vs. threshold training intensity distribution on endurance sport performance, a systematic review and meta-analysis of randomized controlled trials. Journal of Strength and Conditioning Research 33(12):3491-3500. https://doi.org/10.1519/JSC.0000000000002618
- Rosenblat MA et al. (2025). Which training intensity distribution intervention will produce the greatest improvements in maximal oxygen uptake and time-trial performance in endurance athletes? A systematic review and network meta-analysis of individual participant data. Sports Medicine 55(3):655-673. https://doi.org/10.1007/s40279-024-02149-3
- Seiler KS, Kjerland GØ (2006). Quantifying training intensity distribution in elite endurance athletes, is there evidence for an "optimal" distribution? Scandinavian Journal of Medicine and Science in Sports 16(1):49-56. https://doi.org/10.1111/j.1600-0838.2004.00418.x
- Seiler S (2010). What is best practice for training intensity and duration distribution in endurance athletes? International Journal of Sports Physiology and Performance 5(3):276-291. https://doi.org/10.1123/ijspp.5.3.276
- Silva Oliveira P, Boppre G, Fonseca H (2024). Comparison of polarized versus other types of endurance training intensity distribution on athletes' endurance performance, a systematic review with meta-analysis. Sports Medicine 54(8):2071-2095. https://doi.org/10.1007/s40279-024-02034-z
- Stöggl T, Sperlich B (2014). Polarized training has greater impact on key endurance variables than threshold, high intensity, or high volume training. Frontiers in Physiology 5:33. https://doi.org/10.3389/fphys.2014.00033
- Sylta Ø, Tønnessen E, Seiler S (2014). From heart-rate data to training quantification, a comparison of 3 methods of training-intensity analysis. International Journal of Sports Physiology and Performance 9(1):100-107. https://doi.org/10.1123/ijspp.2013-0298
- Treff G, Winkert K, Sareban M, Steinacker JM, Sperlich B (2019). The polarization-index, a simple calculation to distinguish polarized from non-polarized training intensity distributions. Frontiers in Physiology 10:707. https://doi.org/10.3389/fphys.2019.00707
- Tønnessen E et al. (2014). The road to gold, training and peaking characteristics in the year prior to a gold medal endurance performance. PLOS ONE 9(7):e101796. https://doi.org/10.1371/journal.pone.0101796
