Custom distances for Pace Curve & Native support for manual sprint workouts

Feature Requests
1

Hi everyone,

I’m a track & field coach working primarily with sprinters, jumpers, and throwers. I’ve been using Intervals.icu to manage my athletes and I’m really impressed with the platform. However, I’ve run into significant limitations when it comes to analyzing sprint performance.

The Problem (Part 1): The Pace Curve
The current Pace Curve and Critical Speed (CS) model in Intervals.icu is incredibly useful for endurance athletes. However, it’s built on best efforts over standard distances like 1km, 2km, 3km, 4km, and 5km. This means the model is optimized for efforts lasting 2+ minutes.

For my sprinters, their key performances are on 10m to 400m, which last only a few seconds. While the Pace page does extrapolate values for these short distances, the model isn’t designed for this domain, making the extrapolated data unreliable for training prescription.

The Problem (Part 2): Manual Workouts
To make things more challenging, many of my athletes’ key sprint workouts are performed without a GPS watch. Currently, Intervals.icu treats manually entered activities as second-class citizens when it comes to analysis. While it’s possible to create custom fields with scripts to parse workout descriptions (e.g., 4x30m Z7), this approach is:

  • Complex: It requires JavaScript knowledge that many coaches don’t have.

  • Not Scalable: Creating and maintaining these scripts for multiple athletes is time-consuming.

  • Prone to Errors: Manually entered data can be inconsistent, and scripts can break with platform updates.

My Specific Use Case
To properly analyze a sprinter’s training, I need two things:

  1. A Critical Speed (CS) and D’ curve built from their specific test results (e.g., 10m, 30m, 60m, 100m, etc).

  2. A way for manually entered workouts to be analyzed using this curve, without requiring custom scripts. For example, if I log a workout as “6x30m max effort, 3min rest”, the platform should be able to estimate the metabolic cost (D’ consumed) based on the athlete’s CS/D’ model.

Feature Request
I would like to request two interconnected features:

  1. Custom Distances for Pace Curve: The ability to define custom distances (e.g., 10m, 30m, 60m, 100m, etc) for the Pace Curve calculation. This would allow coaches to select the specific distances relevant to their athletes’ sport.

  2. Native Analysis for Manual Sprint Workouts: The ability for Intervals.icu to treat manually entered sprint workouts as valid data for analysis. Ideally, the platform could:

    • Allow users to define the structure of a manual workout (intervals, distances, intensities).

    • Use the athlete’s custom CS/D’ curve to calculate metrics like D’ consumed, intensity distribution, and training load for these manual entries.

Why This Matters
This feature would open up Intervals.icu to a whole new group of users (sprint coaches and athletes) who currently have to look elsewhere for this specific analysis. It would also add a powerful new dimension to the platform for existing users who want to analyze their short, intense efforts.

I’ve searched the forum but haven’t found a request quite like this. Has anyone else found a workaround for this? And David, if you’re reading this, would this be a feasible feature to consider?

Thanks for your time and for building such an incredible platform!

2

image
image1575×520 97.7 KB

For your second request I don‘t understand what the AI want to tell. Do you mean workouts in the sense of planned workouts - or activities, meaning finished activities?

3

Hi,

Thanks for your response. I think there might be a slight misunderstanding about which part of the platform I’m referring to.

You’re right that the “Best Efforts” table lets you track performances over various distances, including 50m and 100m. That’s a useful feature for seeing personal bests.

However, my request is specifically about the Pace Curve itself – the model that calculates Critical Speed (CS) and D’.

The Pace Curve on Intervals.icu is built using your best 1k, 2k, 3k, 4k, and 5k times, specifically selecting efforts that take between 2 and 15+ minutes. This is clearly stated on the Pace page itself, right below the curve.

This means the underlying model is optimized for endurance efforts lasting several minutes. While the curve does extrapolate values for shorter distances like 50m, those are mathematical projections, not data-driven calculations based on actual sprint performances.

For sprinters, this is a critical limitation. A 50m or 100m sprint (lasting 5-12 seconds) is physiologically completely different from a 1km run (lasting 3-5 minutes). Using the same model to analyze both is like using a marathon pace calculator to predict a 100m time – the underlying assumptions don’t hold.

So my request remains: the ability to define custom distances for the Pace Curve model itself (e.g., 10m, 30m, 60m, 100m) so that the CS/D’ calculation is actually based on relevant sprint data, not just extrapolated from endurance performances.

Thanks for engaging with the discussion!

4

Regarding second question. To clarify: I’m referring to Activities (finished, executed sessions), not planned Workouts.

Here’s the scenario:

  1. I plan a sprint workout on the calendar (e.g., 6x30m max effort, 3min rest).

  2. The athlete performs the session without a GPS watch.

  3. I go to the calendar, click “Mark as done”, and a manual Activity is created.

Important assumption: I take it for granted that the athlete executes the workout exactly as prescribed. If there are any deviations (e.g., they only did 5 reps instead of 6, or they had to adjust the intensity), I will manually edit the Activity before marking it as done, so that it accurately reflects what was actually performed.

The problem is that this manual Activity is essentially a blank slate for analysis. It doesn’t have the detailed data streams (speed/power) that a GPS Activity would have. So, while the platform can calculate load and intensity for a GPS Activity, it can’t do the same for a manual one.

What I’m asking for: the ability for Intervals.icu to analyze these manually created Activities in a meaningful way, specifically for sprint work, using the assumption that the prescribed workout was executed as planned (or manually adjusted to reflect reality).

For example:

  • Allowing me to define the structure of the manual Activity (intervals, distances, intensities) when I mark it as done.

  • Using the athlete’s custom CS/D’ curve (from my first request) to calculate metrics like D’ consumed, intensity distribution, and training load for these manual entries.

The goal is to get useful analysis from every session, not just the ones recorded with a GPS.

Does this make sense? Thanks again for engaging with the discussion!