Question on Swim Training Load Calculation – Large Variance Between HR and Pace Priority

Training
1

Hi everyone,

I’ve been experimenting a bit with the Intervals.icu swim metrics and was hoping to get your thoughts on how to best configure training load settings for swimming.

Initially, I had my Training Load Priority set to the default for swimming: Power > HR > Pace. With this setup, I noticed that my swim training load was consistently around 20 TSS, regardless of whether I swam 2k or 3k. It also seemed significantly lower than the load I was seeing for cycling or running sessions. That made it nearly impossible to stay “in the green” for training load balance if swimming was a core part of my routine.

However, when I started creating planned swim workouts, I noticed that the estimated workout load was much higher — often in the 40–60 TSS range for 2–3k swims. That’s when I noticed my (default?) workout load prioirty order was set to Pace > HR > Power. These loads now also scaled more reasonably with distance and intensity.

I switched my Training Load Priority to Pace > HR > Power, and that changed things dramatically. The calculated loads for completed swims are now much higher.

But now I’m wondering, am I tricking myself with this setup?

I’d love to hear how others navigate this:

  • Do you use pace-based load for swimming?
  • Is the large discrepancy between HR-based and pace-based swim load normal?
  • Any thoughts on best practices for the settings to make the data most usefull/fair and consistent with load from cycling/running?

Would really appreciate any advice or shared experience — thanks in advance!

1 Like
2

Just leaving my thoughts here:

1. Go with Pace for Swimming.
I would also use pace. It is more reliable and accurate than HR. There are multiple reasons for that:

  • HR Lag: HR is too slow to react during interval-based training, which is the cornerstone of most swim plans. The HR Load is reduced because of that.
  • Lower HR in general: Your HR in the water is naturally lower than on land for the same effort due to the body’s cooling effect and the dive reflex.
  • Hardware Issues: Wrist-based HR are often inaccurate in the water, and chest straps always slip off, if you do not have a bathing suit over it.

Pace does not have these problems, but you need to check that your watch measures the distance correctly.

2. The “Consistency vs. Accuracy” Decision

Now you get to the big question: how to handle load across different sports. You have two primary philosophies to choose from (that I could think of).

Option A: The “Best-in-Sport” Method (Highest Per-Sport Accuracy)
This is what you are leaning towards. You use the most reliable metric available for each individual sport:

  • Swim: Pace
  • Cycle: Power
  • Run: Pace or HR
    Those are pretty much the methods usually used, as far as I know.

Pro: This gives you the most accurate load calculation for each individual workout.
Con: The Training Stress numbers (TSS) aren’t perfectly comparable across sports because they are calculated differently. This means your overall “Fitness” chart in Intervals.icu should be viewed as a general guide, not an absolute measure. Many athletes successfully manage their load by monitoring the trends for each sport individually.
That is also the the way I do it. Just set your weekly goal load per activity and not in general. Then try to increase those individual values. I also focus more on the load than on the fitness chart, that is generated from it. Finally I decide the load of the next session/week by how I feel and my biomarkers, not the fitness chart.

Option B: The “Unified” Method (Highest Cross-Sport Consistency)
The goal here is to use the exact same calculation method (e.g., HR-based load) for all activities.

Pro: Your load numbers are then perfectly comparable. A “100 TSS” ride is, in theory, directly comparable to a “100 TSS” run or swim. This makes the overall Fitness chart and ramp rate much more meaningful.
Con: You lose per-sport accuracy. As we discussed, HR-based load will likely underestimate your swimming intensity, so you’d be trading accuracy for consistency. I did this for the last half year, and the fitness chart also lost its meaning because of the inaccuracy. That is why I switched my planning and will do the same for my load calculation, to Option A.
(As an example for consitency, I created a custom TRIMPc calculation for this very reason, aiming for a better balance across sports and training intensities).

My Recommendation:
For most athletes, Option A is the most practical. Stick with your change to Pace > HR > Power for swimming. Trust that number. As previously said, it requires a bit of understanding of load management. There are many posts on this topic here. I think it spending time understanding it will help your training a lot.
Option B will only come back to bite you. Sure the chart is pretty but your training planning will suffer, if you base it on that.

One Last Thing: A Sanity Check
Before settling on any method, always ensure your fundamentals are correct (the standard answers of every post):

  1. Check Your Zones: Make sure your Threshold Pace (CSS) for swimming is accurate and up-to-date. An incorrect threshold will give you meaningless data, no matter which method you use.
  2. Check Your Watch: As a quick check, make sure your watch is counting your pool laps correctly.

If you want to dive deeper into the topic, there are a lot of post around “training load”, “external vs internal load” or “load management”. You will probably stumble over some things I wrote here.

3 Likes
3

A question here : pace is wildly unreliable when swimming, as recorded by a Garmin (vivoactive 5). It might work for people swimming open water, but for pool swimming in a swim team it simply doesn’t work. It assumes same stroke style for every 50 m and a pause or wall kick-off. This is not how swim team training work. IM swims change style midway, there are other people in the lane causing small pauses and overtakes which the watch registers as a lane completion. My lates swim registers 100 m paces from 0:12 to 1:35 min.
In the case of pool swim with real-world swim team training sets, how should thevtraining load priority be set? I assume the HR is the only reliable reading.

5

Yes you are completly correct there. Had the same problems. But the problem with the standard HR is, if you do interval based training like most swim teams train, that it takes the average over the whole session. The load calculation becomes terrible. A sprint interval training (for example lactate production) will have not even a third of the load of a easy recovery swim. That makes the normal calculation useless. Maybe check out the TRIMPc i made or read up on it, if you actually want to quantify by hr with interval training. It should resolve those issues, but you will have to use the manual tracking for the 100 IM’s, the drill mode or whatever it is called.

I think triathlete swimmers should be heartrate, normal swimmers pace, competitive swimmers or swimmers in a swim club will probably need to use TRIMPc. I have not found something better until now. Maybe systems like triton could do that, but that would have to be implemented on a club level.

A sidenote: The HR is not as accurate as you might think unless you are using a verity sense or hr-strap. Unless you are female, you can probalby not use an hr strap under the suit.
The arms get thrown around in swimming, the sensors already have hard times with wrist based calculations, and now we catapult blood into them every second stroke.

Hope this helps.

(Link to the TRIMPc post if you want to read up on it)

2 Likes