Training structure: plan for long-term performance gains

One of the main goals of a committed endurance athlete is to improve performance year on year. However good your performances were in 2025, most of you reading this will no doubt be hoping to turn in even better performances this year. However, once you’ve built up a base level of fitness, simply training longer and harder won’t cut it for guaranteed improvements – and could even lead to injury and burnout. What’s needed is a strategic approach that blends the right combination of training volume, frequency and intensity (training load) with excellent recovery. And to do that means following a planned program where training loads are monitored and adjusted, not just within a season, but from season to season.

Getting the recipe correct

If you’re serious about improving over the longer-term – ie from one season to the next, you need to keep a extended record of your training loads over that period to see what worked and what didn’t. This also enables you to check that your training loads are progressing appropriately. One of the most popular ways to monitor is to record training intensity data in terms of power output, which is now the go-to method for many professional and serious amateur cycling pros and triathletes^(1,2). In particular, a power measure known as ‘Mean Maximal Power’ (MMP – see box 1) over a given time period, is a particularly valuable tool for analyzing and monitoring a cyclist’s peak performance during training sessions, competitions or competitive seasons⁽³⁾. Recent research has observed that a 3% improvement in MMP between two different seasons would be within the expected range for professional cyclists⁽⁴⁾.

Box 1: What is mean maximal power?

Mean Maximal Power – often abbreviated MMP - is the highest average power output an athlete can sustain for a specific, uninterrupted duration. So for example, a well-trained a cyclist might have an MMP of 450 watts for 30 seconds, 380 watts for a minute, 350 watts for two minutes, 300 watts for ten minutes and 275 watts for 30 minutes. However, to allow fair comparisons between athletes of different body sizes, MMP is often quoted in watts per kilo of body weight; taking the example above of a 300 watt MMP for ten minutes, an 80kg cyclist would have an MMP of 2.75 watts per kilo for that 10-minute duration.

By measuring MMP over different durations, it’s possible to build up a ‘physiological profile’, which reveals how that athlete fares over a certain duration compared to shorter or longer durations. This can then point to where an athlete might need to focus training intensities. For example, if a clubman cyclist has above average MMPs for durations over ten minutes but below average MMPs for shorter durations, this would suggest he/she needs to focus more on training at high intensities (eg more high-intensity interval sessions) to achieve the balance needed for racing.

Training intensity and MMP connection

For the reasons given above, monitoring training sessions and performance using power meters is now used extensively, both in triathlon, road cycling and other sports such as rowing^(5,6). This makes perfect sense; if you can train to produce more power over a certain duration of time, you are going to perform better over a distance that takes that amount of time to cover. For example, if you are training to improve your 10-mile time trial performance, working on maximising your 20-30 minute MMP is definitely going to help.

Strangely however, the relationship between training intensities used in an endurance program and the MMP profiles that this training generates is very poorly researched. A 2020 study looked at professional cycling, and investigated the effects of adopting a more polarized approach to training⁽⁷⁾. The polarized approach to training says that moderate-hard intensity training – ie near to your lactate threshold - is rather unproductive for making fitness gains. Instead, it proposes that endurance athletes are better to spend a large proportion of their training time working at low intensity (ie in the ‘easy’ purely aerobic zone – zone 1) and a small proportion at a very high intensity (near maximum power output – zone 3) with very little time spent training at or near to lactate threshold (zone 2 - see table 1).

Table 1: The structure of a polarized training intensity distribution