ERG Mode Explained

ERG mode is the single feature that separates a smart trainer from a resistance unit with a knob, and it is also the feature people most often fight. Understanding what it is actually doing makes the difference between workouts that run cleanly and workouts that collapse ten minutes in.

What ERG mode does

In normal use — often called simulation or SIM mode — a smart trainer sets its resistance based on a simulated gradient. You shift gears, and the effort changes exactly as it would on a road. Your power is whatever you choose to produce.

In ERG mode, the relationship inverts. The app tells the trainer a target wattage, and the trainer continuously adjusts its own resistance to hold you there. If the target is 250 W and your cadence drops from 90 to 80, the trainer increases resistance so the power stays at 250. If you spin up to 100, it reduces resistance. Your gear lever effectively stops doing anything.

The point is execution. A structured session is written as a sequence of specific wattages for specific durations, and ERG mode guarantees you actually do that rather than an approximation of it. There is no drifting 15 W under target because the interval hurts, and no accidentally riding 20 W over and burying yourself before the last rep.

The spiral of death, and how to avoid it

This is the failure mode everyone meets eventually, and it is worth understanding because the fix is counter-intuitive.

You are holding a hard target. Your cadence starts to fall. The trainer, doing exactly its job, raises the resistance to keep the power constant. Higher resistance makes it harder to turn the pedals, so your cadence falls further. The trainer raises resistance again. Within a few seconds you are grinding at 50 rpm against a wall, and then you stop.

Three things prevent it. Use a smaller gear. In a low gear the flywheel spins faster for a given power, which gives the trainer more range to work with and makes cadence easier to hold. Watch your cadence, not your power. Your power is handled; your job is to keep the pedals turning at a steady rate. Recover deliberately. If you feel cadence sliding, spin up hard for a few seconds before the trainer can chase you down — recovering early is far easier than recovering late.

If you do get caught, the escape is to stop pedaling for two or three seconds, let the trainer reset, then spin back up. Fighting it never works.

Which gear should you use in ERG mode?

A small-to-middle gear, and here is why it is not arbitrary. Trainer resistance comes from a combination of flywheel speed and applied braking force. At low flywheel speed the trainer has to apply a lot of braking to hit a given wattage, which makes the feel harsh, the response slow, and the spiral of death much more likely. At higher flywheel speed the same wattage needs less braking force, the response is quicker and the ride feels smoother.

A middle gear on the small chainring is a good default for most road setups. It keeps flywheel speed up, keeps the chain line straight, and leaves plenty of headroom if the workout has a hard block later. Some trainers also perform better in a specific gear range; if yours feels particularly sluggish, experiment across a few sessions.

The lag, and why it exists

When an interval changes from 150 W to 300 W, the trainer does not switch instantly. There is typically a two-to-three second ramp, and it is not a fault — it is deliberate. An instantaneous jump would feel like hitting a wall and would be hard on the drivetrain.

What this means practically: for intervals of thirty seconds or more the lag is irrelevant. For fifteen-second efforts, a meaningful chunk of the interval is spent ramping, and the session no longer measures what it was designed to measure. That is the main reason to switch ERG off for short, sharp work.

Most platforms also have an ERG mode power smoothingsetting. It averages the displayed power to make the graph look clean. It is worth turning off, at least once, so you can see what your trainer is really doing — a smoothed line hides both your surges and the trainer's response, which makes it harder to learn to ride ERG well.

When to turn ERG off

Sprints and neuromuscular work. A maximal sprint is not a target wattage, it is an all-out effort, and ERG will actively fight you by reducing resistance as your speed climbs. Use simulation mode.

Very short intervals. Anything under about thirty seconds spends too much of its duration in the ramp.

Racing and group rides. Obviously — in Zwift the whole point is that the terrain and the pack determine your effort.

When you want to enjoy the ride. ERG is relentless. There is no easing over the top of a rise, no freewheeling, no rhythm — just a machine holding you at a number. Some riders find that focus valuable and some find it draining, and both reactions are legitimate. If ERG is the reason you dread indoor sessions, ride in simulation mode and pace yourself instead.

ERG mode and accuracy

Two things are often confused here. ERG mode controls the resistance; the trainer's power measurement is what tells it whether it has hit the target. If your trainer reads 5% high, ERG will faithfully hold you at a number that is 5% off — and it will do so consistently, which is why training on an inaccurate-but-repeatable trainer still works.

If you also own a power meter, expect the two to disagree slightly, because they measure at different points in the drivetrain. Most platforms let you use the power meter as the power source while the trainer handles control, which is the tidiest arrangement. Our accuracy guide explains the difference between accuracy and repeatability, and how to calibrate a power meter covers keeping both honest.

One practical note: a trainer that has drifted out of calibration will hold ERG targets that feel wrong — too easy or brutally hard at a familiar wattage. Before assuming your fitness has collapsed, run a spindown.

Does every trainer have ERG mode?

No, and this is the single most useful thing to know when shopping. ERG requires a controllabletrainer — one an app can command. Any trainer described as supporting FE-C, FTMS or a manufacturer's control protocol will do it. A wheel-on trainer with a lever will not, no matter what sensors you attach to it, because there is nothing for the app to control.

This is the real dividing line in the category, and it matters far more than the accuracy figures brands compete on. Our smart trainer vs dumb trainer guide makes that case, and the controllable field is ranked in best smart trainers. If you are building a first setup around structured training, that is the feature to buy.