The best road for intervals: a checklist for finding a stretch that fits the rep

How long the stretch of road has to be

Multiply the rep length by the speed you hold at that effort, then add about a minute of road at the same speed. An 8-minute threshold rep at 32 km/h needs 4.3 km of clear road; with the margin, call it 5 km. A 20-minute rep at the same speed needs about 11 km. Short reps need far less.

Use your speed on that road at that effort, not your average ride speed. On flat roads most of the power goes into air resistance, and a validated road model predicted measured power to within a 2.7 W standard error from speed, grade, wind and mass [Martin et al. 1998]. The practical reading: wind changes how much road a rep uses. The same watts that cover 4.3 km in still air can cover more than 5 km with a tailwind and under 3.5 km into a headwind. Size the stretch for the fast direction.

The margin covers two things. The first 10 to 20 seconds of a rep are a ramp while you get to the target, and the last part must end before the hazard, not at it. A minute of road at rep speed gives room for both. For a set of short reps the sum works on the whole block: a set of 13 reps of 30 seconds on and 15 seconds easy lasts under 10 minutes, so it needs roughly 5 km of continuous road. That format earned its place in the research, with a larger VO2max gain than effort-matched 5-minute reps over 10 weeks [Rønnestad et al. 2015].

Rep length is the input, and the research favours longer work bouts for most self-coached riders. In 35 trained recreational cyclists, 4×8 minutes produced a VO2peak gain of 11.4%, about double the gain from 4×4 or 4×16 [Seiler et al. 2013]. A steady climb needs less road: at 13 to 15 km/h, 8 minutes uses under 2 km. On a 9.8% grade, 4- and 10-minute power was not different from flat power, though the two were not interchangeable [Hovorka et al. 2022]. Our spoke on how long a hill should be for hill repeats covers the climb version.

The checklist: what makes a road work for intervals

Six checks decide a road. Length, from the sum above. No lights, stop signs or junctions on the work section. A safe place to turn, or a loop. A surface that lets you hold power. Low traffic at the hour you ride. And wind that you account for, not ignore.

Interruptions come first because they are the one check you cannot pace around. In Jeffries' study, 20 riders did a 20-minute time trial indoors and on a closed racing circuit. Mean power was the same, 280 W indoors and 284 W outdoors, but within-rider variability was about double outdoors: a standard deviation of 69 W against 33 W [Jeffries et al. 2019]. That circuit had no traffic and no lights. A public road adds both. Count only the stretch between two points where you never need to brake.

The turnaround matters more than riders expect. A U-turn on a fast road in the middle of a hard set is the most exposed moment of the ride. Look for a car park, a wide lay-by, a quiet side road or a roundabout you can circle. A loop removes the problem, and a loop of 5 to 10 km with one clean stretch often beats a straight road with a bad turn. Broken tarmac, gravel and drain covers make you sit up and lose power.

Traffic is a question of the hour, not the road. A road that is empty at 6 a.m. on a Sunday can be a commuter route at 5 p.m. on a Tuesday. Wind is the last check, and it matters most on an out-and-back. Our reasoning, not a study: hold the hard reps into the headwind or uphill, where the bike has resistance to push against, and use the tailwind leg for the recovery. With a flat tailwind you can run out of gears before you reach the watts. A good road does not cost you the session. Over 40 km, mean power outdoors was 312 W against 303 W indoors, not a significant difference [Smith et al. 2001].

How to find candidate roads near you

Start with the rides you already did, then widen the search. Your own ride files show where you held speed without stopping. Heatmaps show where other riders go. Climb databases list length and grade. Segment search finds stretches of the right length. None of them knows your workout, so you still run the length check yourself.

Your own history is the best source because you know those roads at your hour. Open a few past rides and look at the speed trace. A long section with no drop to zero is a candidate; note where it starts and where it ends. Heatmaps such as the Strava global heatmap show which roads riders use most. Climb databases such as PJAMM and climbfinder list each climb with its length and average grade, which is the quickest way to find a steady 2 km at 5 to 7%.

Segments are a search aid, not a verdict. Filter a segment search by distance for stretches of 3 to 6 km, then check each one on street-level imagery for lights, junctions and a turn. This is the same principle that runs through doing intervals outside: pick the road for the workout, not the workout for the road. A 4 km stretch with two junctions fails a threshold session and still suits Zone 2, which our spoke on a Zone 2 ride without stops covers.

There are tools that do part of this search. Gritmaps, a free beta that reads intervals.icu, builds loops that line up climbs and headwinds with the effort blocks of a workout, which is the closest any tool gets to placing reps. Suggested Routes in AdaptCycling, a paid feature with a 14-day trial, takes a smaller step. On an outdoor day it offers loops sized to today's planned ride from your own speed, using your past rides and new loops from OpenStreetMap. It checks for straight stretches long enough for the reps and for 3 to 8% climbs near the start, and keeps busy roads under a cap. It picks a loop where the work fits. It does not place each rep or check for lights, so the checklist above still applies.

How to use a road once you have found it

Fix one landmark as the start of every rep, and pick it so the rep ends before the first hazard. Use the stretch in your fast direction for the work and the way back for the recovery. When a light or a car stops you mid-rep, the rep length and the stop length decide whether you continue or restart.

A fixed start makes the session repeatable. Choose a post, a sign or a driveway about 15 seconds of riding before the clean stretch begins, and start the rep there so the ramp falls on the approach. Recover on the way back, and extend the recovery if the ride back needs it. For a 4×8 at 5 km per rep, the full session with recoveries uses the same stretch four times, so a loop or a short return leg is worth more than a longer road.

No study has measured what a red light does to an interval, so the rule here is reasoning from how the effort works. The value of a threshold session is the time you accumulate in the target range; 32 minutes at about 90% of peak heart rate beat 16 minutes at a higher intensity [Seiler et al. 2013]. So in a long threshold or sweet-spot rep, a stop of under about 30 seconds costs little. Continue when you can, and add the lost time to the end of the rep. A VO2max rep is different. Oxygen uptake takes one to two minutes to climb near its peak, and that part drops fast when you stop. If a stop catches you in minute three of a 4-minute rep, the minutes that count are gone. Take a full recovery and do the rep again.

Short reps sit between the two. If a stop breaks a set of 30-second reps, finish the set after it and count the stop as a recovery. If a road interrupts most sessions, it is the wrong road for that workout. Use it for endurance, or take the hard session inside. Our guide to indoor cycling training explains how the trainer holds the watts when no road nearby can.

Quick answers

Sources cited in this guide

1. 01
   Jeffries et al. 2019. An Analysis of Variability in Power Output During Indoor and Outdoor Cycling Time Trials. International Journal of Sports Physiology and Performance.
2. 02
   Smith et al. 2001. Reliability of mean power recorded during indoor and outdoor self-paced 40 km cycling time-trials. International Journal of Sports Medicine.
3. 03
   Martin et al. 1998. Validation of a Mathematical Model for Road Cycling Power. Journal of Applied Biomechanics.
4. 04
   Seiler et al. 2013. Adaptations to aerobic interval training: interactive effects of exercise intensity and total work duration. Scandinavian Journal of Medicine & Science in Sports.
5. 05
   Rønnestad et al. 2015. Short intervals induce superior training adaptations compared with long intervals in cyclists - an effort-matched approach. Scandinavian Journal of Medicine & Science in Sports.
6. 06
   Hovorka et al. 2022. Effects of Flat and Uphill Cycling on the Power-duration Relationship. International Journal of Sports Medicine.

More inside How to do intervals outside

Other articles in this series

1. 01

   How long should a hill be for hill repeats?

   Rep length sets the hill. A table from 30-second to 10-minute repeats: how long, how steep, and which workout each hill serves.

   →
2. 02

   VO2 max intervals outside: hill or flat road?

   Uphill and flat VO2 reps are not the same work. What the research says about each, and how to ride 5x4 when you have no hills.

   →
3. 03

   Threshold intervals without hills: rolling and flat roads

   No long climb nearby? How to hold threshold and sweet spot on flat or rolling roads: pace by range, use the wind, fit reps to the road.

   →
4. 04

   Zone 2 outside: how to plan a ride without stops

   Lights and junctions break an easy ride into pieces. How to plan a Zone 2 loop that stays steady, and what to do when you cannot.

   →