What's Cool In Road Cycling

Variable vs Constant Intervals – What’s the Difference?

Not all intervals are created equal, even if the average wattage is kept similar. Making intervals that vary in intensity can make them really different from constant wattage intervals.

van poppel
Tough short cobbled interval for Danny Van Poppel

Last month, I discussed my masochistic infatuation with the Rønnestad intervals, consisting of 13 repeats of 30 s all-out and 30 s recovery at around 50% of your threshold power. I then discussed a study from Bent Rønnestad comparing these efforts to longer efforts at a steady wattage, showing that they elicited greater improvement gains even in very well-trained cyclists (Rønnestad et al. 2020).

I won’t repeat myself with my idolatry of the Rønnestad intervals, but the past month I’ve been playing with a variation, kind of a mash-up between the Rønnestad with a “Hour of Power” effort of steady tempo with short bursts of 15-20 s every 2-3 min.

  • 20 min efforts, 2-3 sets per session.
  • All-out efforts of 30 s. I have lots of quiet country roads, and I’ll randomly start them after crossing each intersection or else every 90-120 s, whichever comes first. I’ll stand up and sprint the first 10-15 s.
  • Recovery still at a “tempo” effort. For me, I peg my recovery effort at about 190-200 W.

Main lower coloured line is my power output. Purple line is Xert’s Maximal Power Available (MPA). Grey line is Difficulty Score.

What I end up with is about 10-12 hard 30 s efforts. They’re at higher wattages (~340 W) than when I do Rønnestads. The recovery wattage of 190-200 W is also higher than the 120-140 typical of the shorter efforts.

However, due to the longer duration of the recovery efforts, overall average wattage (~235 W) is lower than with Rønnestads (255-260 W).

Regardless, they’re harder than a traditional “Hour of Power” where the on efforts aren’t as intense and the purpose is to just push a bit above Threshold Power. They’re not as bleeding eyeball intense as the Rønnestads, but they’re still plenty hard and I’m also pretty much guaranteed to reset my Fitness Signature (when the purple MPA line gets drained down to your power output).

Bossi et al. 2020

The above prelude is mainly to demonstrate that there are many variations of a similar workout. The constant is that I infinitely prefer these variable intervals to a steady state interval, and my previous article comparing variable versus steady intervals suggest that the former leads to better fitness improvements even in very elite cyclists.

A new 2020 study by Arthur Bossi, a Brazilian PhD student studying at the University of Kent in the UK, with Bent Rønnestad as a co-author, studied this same idea of comparing variable versus steady-state intervals (Bossi et al. 2020). Rather than looking at the training adaptations over time, this study was interested in what were the physiological responses to one specific session of either variable or constant intervals.

Specifically, they wanted to see whether there was a difference in the time spent above 90% VO2max, as they theorized that this was one reason why variable intervals may be of greater efficacy.

Some of the main details here include (major differences with Rønnestad 2020 in italics):

  • Also very highly fit and trained cyclists (14 males with VO2max = 69.2 mL/kg/min).
  • Total duration of ON and OFF efforts were matched. Recovery duration and intensity were equal between variable and steady conditions.
  • Rather than self-paced, the ON efforts for both variable and steady state were clamped at set wattages based on Maximal Aerobic Power (MAP).
  • Each condition involved 6×5 min interval sets with 2.5 min recovery at 30% MAP (see figure below).

Main Findings

The study design was sound and so was the methodology overall. A key with this study, like with Rønnestad et al. (2020) was the use of very fit cyclists. They would likely be very close to their genetic ceiling, so any significant findings here likely would be magnified in less fit cyclists like the majority of us.

  • No difference in mean power output, heart rate, lactate levels, or cadence for each session between variable and constant conditions.
  • No differences in session RPE responses, so overall each type of interval felt roughly similar in intensity.
  • A big difference in the time spent above 90% VO2max, with 410 (207) s for variable intervals versus 286 (162) s for constant intervals.
  • The variable intervals had a slightly higher ventilatory response. Namely, while breathing frequency and tidal volume (the volume breathed per breath) did not differ, the slight differences in each combined for a higher overall ventilation (volume breathed per minute) for the variable compared to the continuous interval. This meant that some of the higher duration at >90% VO2max may be due to greater energy cost from breathing more.


Whether you want to join me on the variable effort interval bandwagon or not, the key idea to take away from this study is that there are lots of ways to tune and refine an interval session. This study shows that simply looking at average wattage does not really provide full information. As with the Rønnestad study, the higher bursts of effort with variable intensity intervals places stress upon a wider range of both muscular and metabolic demands. This isn’t necessarily always the goal of a particular workout or interval session, so there truly is a time and a place for everything.

Ride fast and have fun!

A longer interval for Bob Jungels


Bossi AH, Mesquida C, Passfield L, et al (2020) Optimizing Interval Training Through Power-Output Variation Within the Work Intervals. Int J Sports Physiol Perform 15:982–989. https://doi.org/10.1123/ijspp.2019-0260

Rønnestad BR, Hansen J, Nygaard H, Lundby C (2020) Superior performance improvements in elite cyclists following short-interval vs effort-matched long-interval training. Scand J Med Sci Sports 30:849–857. https://doi.org/10.1111/sms.13627



Like PEZ? Why not subscribe to our weekly newsletter to receive updates and reminders on what's cool in road cycling?

Comments are closed.