TSB - the ignored metric?

TR uses most of the Coggan/Allen metrics but I never see or hear anything to do with TSB. What consideration of this was made when devising the TR plans? The effect on individual fatigue is surely going to be very different if one person begins a plan at +10 compared to someone starting from -10.

I’d imagine there’s an assumption that people will be starting from a position of reasonable freshness. You don’t embark on a new training regime the day after riding yourself into the ground for 8 weeks without any rest. And that will obviously be the case if people are going through base / build / specialty, because the recovery weeks are built in.

What makes a bigger difference is what CTL you start out at. If your ATL is 100 but your CTL is 85, then that’s only -15. If you did the same 7-day workload but start at CTL 40, you’re going to be deep in overtraining territory. Which is why I’m sure Coach Chad would encourage cyclists to select a volume that doesn’t hugely increase the amount of work they’re used to.


Had I done a search for the subject of TSB I’d have found a few threads on it already from about 1 year ago with various official answers such as

  1. Chad does consider it
  2. We don’t want to over complicate things
  3. Use another app if you want that data
  4. 6 week average TSS is a rough equivalent
  5. There is something better coming (presumably this was plan builder)

There will no doubt be lots of TR users who do already have another source for their PMC and I assume if TR wanted to create a PMC within their product they’d have definitely done it by now given how easy it would be to implement but i do still find it odd that nobody ever asks about TSB and it’s never mentioned in the podcast.
I may plug in the data from a random plan and see where it takes my own figures in the chart which I maintain for myself.

1 Like

In the book “training and racing with a power meter” there is some good info on performance at various TSB.

1 Like

Hi, yeah, I fully understand it, that’s not my point. It’s just the fact that it’s never talked about.

1 Like

I think this is key.

“Trainer Road users” are very different from “Trainer Road forum users” as a group. One of TR’s selling points - in fact, one of their most important selling points - is that if you follow the plans, you know that such matters as form, fatigue, progressive overload, targeting energy systems, loading / rest cycle, etc have all been worked out for you.

Note that the first thing that Plan Builder asks you is “How many hours per week have you trained in the last six weeks?”. I don’t think it’s a coincidence that they ask about 6 weeks - they’re basically trying to get a rough ballpark figure for your CTL so they can prescribe a plan that will give you an appropriate CTL ramp rate that will challenge you but not send your TSB into the red.

Isn’t that CTL? TSB is (roughly) 7-day average - 6 week average, or CTL - ATL in TrainingPeaks terms.

You can do it fairly quickly by syncing your TR calendar to intervals.icu. I tried out 3 scenarios…

In order to create much TSB (or “form” in intervals) I needed to add a SSB HV 2, which if I started it today would take me from 84 to 94 CTL, and reach a TSB of -16.

When I add General Build MV, my CTL goes from 84 down to 74, and my TSB only goes down to -8.

But if I take 4 weeks totally off the bike to let my CTL decline, and then do General Build MV, it’d take me from a CTL of 51 to 64, and a TSB of -18.


Now intervals.icu supports expressing TSB as a percentage of fitness (CTL), so if you operate with CTL above or below the “normal” range it may give you a better, or at least more understandable, picture.


Yes, I thought about mentioning this but wondered if it was getting too far into the weeds… I used absolute values for the scenarios above to keep it consistent with TrainingPeaks, but use percentage of fitness as my default metric. Personally think TP should do this too.

One limitation to these metrics (and also a limitation to Plan Builder) is that they only look at the last 6 weeks, and don’t consider the long-term training load of the athlete.

For example, early this year I had 2 weeks pretty much off the bike with a bad stomach bug, followed by another 1-2 weeks of lower volume and intensity while I recovered. At the end of this, my CTL had naturally plummeted. But because I’ve done years of fairly high volume training, I was able to go straight into a high-TSS block of work, which in TSB terms sent me way too far into the red - when in reality it was a tough but definitely manageable workload.

1 Like

Yes, I’m sure it would take some data crunching, but a metric defining deep fitness would be cool.

For me the exact opposite happened. I had a 30hr training camp in February followed by two of my biggest months ever in March and April (3 weeks on, 1 week off) . After recovering from the training camp I was never in the red but it still took two weeks to recover after April. If I had just looked at TSB when I was trying to restart (I was super positive but I didn’t feel fresh at all) the first time I would have thought I dug too deep of a hole. I listened to my body, took more rest, and now I’m back to it (with a sizeable FTP bump, may I add).

1 Like

Ah I didn’t know about the ICU sync function, I use that too so will explore. I’m at work now but in a nutshell how do you do it, presumably through ICU?

  1. Go to Log In to TrainerRoad and click on Google. Then click “show instructions” and then copy the URL that appears.

  2. Go to intervals.icu and your activities calendar. Click “options”, then “ADD CALENDAR”, then paste in the link. This will then paste the activities and expected TSS of your training plan into intervals.

  3. Then if you set the end date on the “fitness” tab to a few weeks into the future, you can see what your training plan is going to do to your form, fatigue etc.


That is excellent, thanks so much for this.

It’s called FTP.

Perhaps “deep fitness” isn’t quite the right phrase. Maybe “long-term tolerance to workload”.

Anyway, you can have a 220 FTP and cope perfectly well with a high volume plan, or have a 280 FTP and fall apart.


@Nate_Pearson I asked on Strava what your TSB was at the time when you felt you had dug yourself into a bit of a hole. Did you ever look it up?

I would strongly disagree with that.

Depending on how you test FTP, results may vary.
For example, I had roughly the same FTP in 2017 and 2019, in 2017 I had a higher 20-minute power but in 2019 I had higher power for all durations 30 minutes or more.
I would call that deep fitness.

If I were to propose such a deep fitness metric, it probably would have something to do with the slope of the best fit line of your power numbers for efforts longer than 20 minutes or an hour.


On one hand, I think TSB is a useful metric and I do buy into the idea of what Joe Friel says regarding optimal training happening after about -10 TSB. I’ve been doing SSB1 HV these past few weeks and I’m not anticipating any changes. Partly it’s due to the fact that I had done SSB1HV and SSB2HV over the winter (followed by a week vacation in late Feb with a VO2 progression block done throughout March and April), so I’ve had a pretty decent base, and I started the block at 74 CTL and after Sunday’s workout I was at 81, so not a huge jump over 5 weeks. My TSB never went really deep and just after this past weekend of 2 big workouts did it go to -15. While I think I got the benefits of doing sweet spot in a lot of ways, the level of work really isn’t going to elicit any changes from an FTP standpoint.

But I also think that speaks to the fact that I was training a system that was pretty well developed and if I may have had different results if I had done similar TSS but with more demanding threshold/over-threshold work.

Calendar synced succesfully, thanks.

So you have stamina in WKO that describes exactly that.

So is stamina the ignored metric, then? What’s a good number? A bad one?