The Beginner's Guide to Stravistix or How I Learned to Stop Worrying and Love the TRIMP

[kleph]

After joining the herd on Strava I poked around these here threads for some information about what the heck was going on. As you might expect, a lot of the posts were from @DopeyBadger and, more often than not, they mentioned this arcane and mysterious entity, Stravistix.

Basically, Stravistix is an extension for the Chrome and Opera browsers that takes the information from Strava and uses it to create detailed analyses of your fitness. You can find the plug in here: http://thomaschampagne.github.io/stravistix/

So I downloaded it and ran my Strava data through it and looked in awe at this huge chart of information it spit out. The problem was I had no idea what any of it actually meant. Well, I kind of had an idea what some of it meant but I had a lot more questions than answers.

The next step is an appeal to the master himself.

TRIMP (link) and HRSS (link).

The short version of those is that TRIMP and HRSS are methods to calculate the stress (or training load) of a workout using duration and relative HR (based on someone's resting HR, max HR, and LT HR). Also, "Fitness" builds slower than "Fatigue" as both are measures of training load averages over timeframes of 6 weeks and 2 weeks. So it takes longer to build fitness, and fatigue can be erased at a faster rate.

See? Clear as day. Day-ish. Better than complete darkness, anyway.

I also came across a few entries from fitness bloggers talking about it as well. This link is easily the best of the bunch and has a pretty good breakdown of what Stravistix is showing. http://www.runningmatters.me.uk/2017/01/06/stravistix-making-strava-even-awesome/

At this point I'm liking Stravistix as a solid overview of my progress. The issue I've got is I don't really understand what it's showing me except in a general way. So, in hopes @DopeyBadger or others might have insight to help me along, I've started this thread.

 

[DopeyBadger]

The first thing you are definitely going to want to do is check your setting on Stravistix. Accurate data makes accurate output. So max HR, LT HR, resting HR are extremely vital inputs. These should help give you an accurate representation (if you change them after you sync, then make sure to re-sync). If these inputs are off, the data is going to be thrown through a loop. One thing I don't like about it is the use of "Fitness" for training load 6 week average. Because one could confuse that higher "Fitness" equals faster times, but that doesn't necessarily have to be true. It's simply a reflection of your training load over the last 6 weeks (and fatigue is the last 2 weeks).

The key number to watch is "Form" as this is Fitness - Fatigue. This shows you whether your training is consistently building (optimal), stagnate (neutral), not enough (fresh), or way too much (overload). The goal is to spend as much time in "optimal" as possible. Using this, you can find what works best for you for training, too much training, and nailing the freshness for race day.

 

[kleph]

To make it clearer to anyone following along from home, I'm going to quote your information on this point from the link in the original post.

Fitness = The Fitness curve is the long-term average daily training load. Default period is fixed to 42 days (or six weeks), so it will take ~6 weeks for your Fitness to be accurate. It's basically the foundations, the potential availability of the athlete... You can't get into big rides/runs or races without a minimum of fitness acquired. If that curve is too regular or too flat, it is not usually a sign of good training

Fatigue = The Fatigue curve is the short-term average daily training load. Same way than Fitness but default period is fixed to 7 days (one week). Conceptually, fatigue is easy to understand: It's the tired feelings which limits your performance. This curve varies much faster than Fitness curve. On Fatigue curve, you can see the fatigue climbing sharply in response where you performed workouts with a high stress. But also go down quickly as you take few days off.

Form = The Form curve is simply the difference between Fitness and Fatigue, it's the right balance of stress you should look at to avoid over training for example.

When Form = :

• +25 < Form : Transition zone. Athlete is on form. Case where athlete has an extended break. (e.g. illness, injury or end of the season).
• +5 < Form < +25 : Freshness Zone. Athlete is on form. Ready for a race.
• -10 < Form < +5 : Neutral Zone. Zone reached typically when athlete is in a rest or recovery week. After a race or hard training period.
• -30 < Form < -10 : Optimal Training Zone.
• Form < -30 : Over Load Zone. Athlete is on overload or over-training phase. They should take a rest!

Fitness is the long term gains over time. Fatigue is the short term tired feelings. Form is the balance between Fitness and Fatigue.

This makes sense to me. I've always said that training is the consistency from one training run to the next. If you've got consistency, then you'll continuously improve. If you start to miss runs or adjust things, then you'll probably stagnate. It also means that no one run defines the training cycle. That 150 minute LR is important, but so is the Tempo run proceeding it and the speed workout after it. It just further underlines that training is not just one single important run, but rather the cumulative effect of all the training.

The balancing between Fitness and Fatigue is important too. You could do a massive training week and you'd slightly improve your Fitness, but you would greatly increase Fatigue. Too much fatigue and you increase the chance for injury.

 

[kleph]

For an example of what we are talking about here is a snapshot of @DopeyBadger chart with Fitness, Fatigue and Form on it.

 

[kleph]

To boil it down to basics (as I understand it) TRIMP and HRSS are methods to calculate the stress (or training load) of a workout using duration and relative Heart Rate. Stravistix allows you to look at charts using either one.

"Fitness" is a rolling average of your TRIMP or HRSS over the last 42 days. Also known as Chronic Training Load.
"Fatigue" is a rolling average of your TRIMP or HRSS over the last seven days. Also known as Acute Training Load.
"Form" is then the difference between the two.

Since these measures are established by heart rate, this is why there is a need to have accurate baseline numbers which @DopeyBadger mentioned above.

Edited to include both measuring methods.

 

[kleph]

Freshness, Neutral, Optimal and Overload are measures of Form.

• +25 < Form : Transition zone. Athlete is on form. Case where athlete has an extended break. (e.g. illness, injury or end of the season).
• +5 < Form < +25 : Freshness Zone. Athlete is on form. Ready for a race.
• -10 < Form < +5 : Neutral Zone. Zone reached typically when athlete is in a rest or recovery week. After a race or hard training period.
• -30 < Form < -10 : Optimal Training Zone.
• Form < -30 : Over Load Zone. Athlete is on overload or over-training phase. They should take a rest!

 

[DopeyBadger]

For an example of what we are talking about here is a snapshot of @DopeyBadger chart with Fitness, Fatigue and Form on it.

One thing this chart shows is the ebb and flow of a training plan (and what an injury looks like). The purple line is the "Fitness" which moves slowly, and the yellow line is "Fatigue" which spikes consistently based on the individual workouts. Then over months you can see the training plan building, then the race, then the recovery, then the next training plan. My training load has been pretty consistent for years, yet my PRs have consistently dropped.

 

[kleph]

I really like this because it can be difficult to tell how much progress you are making just by looking at distance or pace. This provides a nice snapshot of one's overall effort over time and that shows progress better. Obviously every system is only as good as it's data and the HR monitor on fitness apps are notoriously wonky, but I'm looking for changes over time, and I feel comfortable this gives a measure of that.

For example: I know I am in better shape than I was a few months ago but my pace hasn't improved that much. At the same time, I'm handling long runs better but they still leave me worn out. Now I also know my heart rate isn't as high as it used to be, even on the most difficult runs. This system uses that to show me how my physical condition has improved in a visual way I can understand.

The problem is I still have no sense of what the numbers mean. To wit: my "Fitness" is 26.4 but @DopeyBadger is 43.4. So is mine bad? Do I want to reach a specific level before taking on a major race? Is the number in TRIMP better or worse than HRSS? Which should I be using to get a better sense of my overall progress?

 

[PaDisneyCouple]

wow. so it sounds like if you want to get into analyzing fitness at a high level, strava has the tools to do it. this is very interesting!

I'll keep checking in.

 

[DopeyBadger]

To wit: my "Fitness" is 26.4 but @DopeyBadger is 43.4. So is mine bad?

No. It merely means that I do more training then you. It doesn't mean that I'm any more fit then you. For example:

-Two runners complete the same duration workouts in a training plan.
-The effort of those duration workouts is the same when evaluating relative HR.

The faster runner does 150 min at LR pace.
The slower runner does 150 min at LR pace.

The faster runner does 20 miles.
The slower runner does 12 miles.

The faster runner is a 3:00 marathoner.
The slower runner is a 6:00 marathoner.

But if the conditions above are met (same duration and same relative HR), then they should have the same "Fitness". So fitness is not a measure of speed, abilities, or progress. It's simply a measure of training load. So my number being higher just means I have a larger training load.

Now in a lot of cases, the more training load you have the faster you will individually improve. But the intricacies of the training plan matter for that to be true.

Is the number in TRIMP better or worse than HRSS? Which should I be using to get a better sense of my overall progress?

HRSS will be more accurate if you indeed know your lactate threshold HR. It may or may not be the pre-set value. But having a higher relative LT HR should enable you to have a higher capability over the long term.

 

[kleph]

Here is the last six weeks of my data using the HRSS measure. I don't have my LT HR but I'm actually not concerned about immediate accuracy as much as overall trends. My understanding is that will still be expressed in these data.



And here it is with TRIMP. As I said, the same general trends are shown.



As I am about to go into marathon training, my concern is the apparent plateau of the fitness level. I feel comfortable I'm not currently overloading myself due to the form information (I kind of went overboard at the start of the month and it wasn't pleasant). But miles and, hopefully, pace is about to increase substantially. If fitness isn't increasing, the only thing upping fatigue will accomplish is pushing Form down to overload. No?

Edited to include correct chart.

 

[DopeyBadger]

As I am about to go into marathon training, my concern is the apparent plateau of the fitness level. I feel comfortable I'm not currently overloading myself due to the form information (I kind of went overboard at the start of the month and it wasn't pleasant). But miles and, hopefully, pace is about to increase substantially. If fitness isn't increasing, the only thing upping fatigue will accomplish is pushing Form down to overload. No?

So actually, the training you were doing from May 27th-June 5th area was appropriate training per this measurement (assuming the six weeks prior is also good data). Doesn't mean the training itself was appropriate, but simply that the combination of how much and at what HR was increasing at a steady rate. The fitness was increasing slowly, and you weren't overstepping with too much too soon.

But it appears that you didn't change the stimulus and starting around June 10th the training was too consistent. Not enough polarization in pacing or mileage. So May 27th-June 5th was good training. And June 10th-July 1st hasn't really been spent getting better.

Now, pacing is appropriate when relative to your current fitness. Don't merely increase pace to try and get fitter (or increase Fitness). It's my opinion that training at current fitness yields the best long term results. So if a recent 5k suggests a marathon time per a race equivalency calculator, then that's the set of paces you use. Now if you mean "increase the pace" by including more "hard miles" then again it's all relative. You want to keep the ratio of hard/easy to be relatively in the area of 80/20 (it can ebb and flow). So more Marathon paced miles means you need significantly more easy miles to compensate.

Keep in mind when working with Stravistix, it's merely a tool in the toolbox. It's not the end all be all. You shouldn't make sweeping dramatic choices based on the data. Use it as a tool along with other resources to help improve the quality of your training.

Here is a recent chart from me:



After coming off a stress fracture, I was finally able to start running again in mid-March. The amount of mileage I was doing was relatively low, but consistently building. I kept the pacing easy. Then in late April, I was able to actually start training again. You can see that time period from mid-March to April was really good training (per this measurement). I increased the mileage at a nice and consistent rate and kept my training in "optimal" a majority of the time. Then, at the end of April I got stung on the foot by a wasp (the day I was allowed to start training again). You can see the "Fatigue" drop quickly from an inability to train for a week. You can also see the "Fitness" drop some, but it's less effected by the little things like a week off. It helps emphasize that training is the big picture. A missed day or two here or there will not destroy training, it's the consistency in training that makes it.

Once the foot was healed, I could return to training again. It took some time to build the fatigue back up, but after a few weeks I was back to optimal training load. Then in late May, we went on a trip to WDW and I took time off again. Fatigue dropped, but Fitness didn't move much. Restarted training and Fatigue jumped back up. Training was optimal, until I was ready for my HM in mid-June. Then I intentionally dropped the load to try and get my form into "Fresh" and I barely made it come race day.

After the mid-June race I had a week of only easy running. You can see that stagnated the training. So that's a bad thing right? Not necessarily. It was warranted because I just raced. Which means this Stravisitix data may not be necessarily reflecting the muscle damage done from a hard race. Now in late June, I started my marathon training plan and you can see that it's just starting to pick back up again and I anticipate the Fitness and Fatigue shall continue to climb. The hope is that the training stay near optimal from here until October. I'll use this data in conjunction with other things I track (like HR v pace, effort for certain paces, consistency of pacing within a workout and within a week's set of workouts, etc.) to make sure I'm on the right track for race day.

 

[kleph]

Keep in mind when working with Stravistix, it's merely a tool in the toolbox. It's not the end all be all. You shouldn't make sweeping dramatic choices based on the data. Use it as a tool along with other resources to help improve the quality of your training.

I think we need to emphasize this point. I'm finding Stravistix offers a very helpful perspective into how I am progressing but it's not the absolute authority on where I stand physiologically. The other measures of progress, which can be frustrating because they are incomplete, still have critical value; I'm running faster, I'm running further and I'm avoiding injury. I'm fitting all this into my available time. My goal with this metric is to continue the progress in all of these areas as I start adding the demands of marathon training.

As for some of the particulars you are addressing. I picked the six-week window because a lot of my work over the last six months is getting back into shape. I felt I was doing good by the end of April but then did a WDW trip in early May and everything dipped. The chart shows pretty much my recovery from that to now. A few things have occurred in the last six weeks that, after looking at the Stravistix chart and taking your comments under review, I can put into context.

1) The fatigue spike in early June was accompanied by some awful long runs where I didn't prepare properly. I didn't hydrate, I didn't eat well prior, I went out too fast at the start. I suffered. That was followed by a weekend where I was so exhausted I had to skip the long run completely (ergo the big dip in fatigue). I slept 10 and 12 hours over the course of several nights and, at the time, couldn't figure out why. Now I know.

2) The leveling off in fitness also corresponds to a dramatic improvement in my HR efficiency. My Garmin tracks time in HR Zones and I saw that, starting in mid June, I simply wasn't hitting Zone 5 anymore. This is as my paces were improving and the resumption of the long runs (the latter of which are visible on the Fatigue measure). This is a big reason I'm excited about Stravistix because now I can understand the dynamics of that and plan my training better. Which leads to the next observation...

3) In mid-June I began toying with the idea of an October marathon and penciled in a candidate. That sent me back into looking at training plans and on collusion course with your thread on the subject which started my curiosity about Stravistix. I've identified a number of elements I want to incorporate in my plan -- interval runs, hill runs, pace/long run combos -- and the last couple weeks has been experimenting with them. I have been deliberately conservative on these as I figure out if my routes, exertion and recovery are all in line before diving in deep end.

4) I am planning a 10K on Wednesday for which I hope to get a better gauge of where I am at. That will help me determine what my pacing needs to be for my training plan. With that in hand, I should be able to proscribe targets for my various runs in order to push myself on the hard efforts and keep it in check on the easy ones. That should allow me to transcend this plateau on the fitness metric.​

I am having fun with Stravistix but my training plan is deliberately not as intricate and detailed as what you produce. There is a reason for that. My goal in October is completing the first marathon I've done in ten years. So the emphasis is on getting strong enough to run the race in it's entirety, speed is a secondary goal. The key to that is not optimizing my training but avoiding injury. I've had my experience with tackling too much with a training plan when I wasn't ready and I don't plan on ever doing that again.

Or, to put it a different way, I'm having fun playing around with these running dork toys but it's not time to go full-on dork yet.

 

[kleph]

For anyone who is still following along now @DopeyBadger and I have resorted to WALL-O-TEXT mode, please let us know your own thoughts and experiences with Stravistix. Upon using it the first time I was flat lost as to what the data meant. This thread is, hopefully, an explainer for anyone else trying to sort it out. My hope is that our discussion on our specific readouts provide a way for folks to figure out what the details of their own charts might mean.

 

[DopeyBadger]

I think we need to emphasize this point. I'm finding Stravistix offers a very helpful perspective into how I am progressing but it's not the absolute authority on where I stand physiologically. The other measures of progress, which can be frustrating because they are incomplete, still have critical value; I'm running faster, I'm running further and I'm avoiding injury. I'm fitting all this into my available time. My goal with this metric is to continue the progress in all of these areas as I start adding the demands of marathon training.

For me, my gold standard is tracking HR vs Pace. I believe it gives the most accurate representation of where you are "right now" when used appropriately. It's just a matter of cleaning up the data. The Garmin devices kind of do this on their own by providing a VO2max estimate which is the watch doing the same thing I do. And rightfully so, you should see ebb and flow in the Garmin VO2max (or HR v Pace data) as the temp or terrain changes. It gives you output to suggest where you are in relation to that run vs others.

Something to keep in mind with the Garmin VO2max though. It produces a "race prediction" output that is extremely aggressive even when your HR max, body weight, and HR rest are accurate. When these inputs are not accurate, the error on this will be even bigger. But the "race prediction" it spits out assumes you have maximized all other capabilities. It generally places the marathon prediction at 82-85% VO2max which is extremely high and typically only elites (hence those who have near maxed potential out) are able to achieve the values. But if you keep things in perspective, you can see that if you don't change the inputs (weight and HR max/rest) then any improvement seen in Garmin's VO2max value is a true improvement. Although adjusting weight is a justifiable change as your VO2max will improve with a lesser body weight (assuming you don't lose too much body weight). It's all relative in relationship to VDOT expressed as liters/min which uses weight in its calculation.

2) The leveling off in fitness also corresponds to a dramatic improvement in my HR efficiency. My Garmin tracks time in HR Zones and I saw that, starting in mid June, I simply wasn't hitting Zone 5 anymore. This is as my paces were improving and the resumption of the long runs (the latter of which are visible on the Fatigue measure). This is a big reason I'm excited about Stravistix because now I can understand the dynamics of that and plan my training better. Which leads to the next observation...

How do you have your HR zones setup? I use the HRR method (Heart Rate Reserve) which takes into account both resting HR and max HR. A zone 5 run seems quite aggressive as I'm not sure I hit more than a few seconds in any given week in Zone 5 (for me being a HR above 90% of HRR). I generally only hit those values at the very tail end of a 5k/10k race.

I also use Daniels as a guide:



For the most part, his values jive with my own as my easy is 120-137, LR is 138-142, M Tempo is 148-152, HM is 154-156, LT is 156-158, and I generally don't get above 159 unless it's the tail end of the race.

4) I am planning a 10K on Wednesday for which I hope to get a better gauge of where I am at. That will help me determine what my pacing needs to be for my training plan. With that in hand, I should be able to proscribe targets for my various runs in order to push myself on the hard efforts and keep it in check on the easy ones. That should allow me to transcend this plateau on the fitness metric.

Good idea.

I am having fun with Stravistix but my training plan is deliberately not as intricate and detailed as what you produce. There is a reason for that. My goal in October is completing the first marathon I've done in ten years. So the emphasis is on getting strong enough to run the race in it's entirety, speed is a secondary goal. The key to that is not optimizing my training but avoiding injury. I've had my experience with tackling too much with a training plan when I wasn't ready and I don't plan on ever doing that again.

So then the goal should be more easy and less hard. That'll help keep you injury free but still progressing. Since the goal is a marathon, then you're looking for mostly aerobic training anyways since the marathon itself is 99% aerobic. So in general, if almost all pacing is slower than M Tempo, then you should be in a good place.

 

[kbenson13]

For anyone who is still following along now @DopeyBadger and I have resorted to WALL-O-TEXT mode, please let us know your own thoughts and experiences with Stravistix. Upon using it the first time I was flat lost as to what the data meant. This thread is, hopefully, an explainer for anyone else trying to sort it out. My hope is that our discussion on our specific readouts provide a way for folks to figure out what the details of their own charts might mean.

I truly admire and enjoy everything he has to say, but I would say @DopeyBadger defaults to wall-o-text mode rather than resorts to it!

Of course, he always has a ton of valuable things to say and I devour his walls-o-text voraciously. But I just thought it was funny the way you worded that!

Being serious, I'm glad y'all started this thread. I just downloaded StravistiX and after reading these posts know what the heck I'm looking at. Thanks for doing the hard work on this. I'm excited to see my fitness slowly and steadily trending upwards (though nowhere near either of yours). So so far, so good....

 

[DopeyBadger]

I truly admire and enjoy everything he has to say, but I would say @DopeyBadger defaults to wall-o-text mode rather than resorts to it!

It's a problem. At least I can admit it.

Being serious, I'm glad y'all started this thread. I just downloaded StravistiX and after reading these posts know what the heck I'm looking at. Thanks for doing the hard work on this. I'm excited to see my fitness slowly and steadily trending upwards (though nowhere near either of yours). So so far, so good....

And something to keep in mind for anyone who recently purchased a HR monitor, the data isn't useful on Stravistix for at least 6 weeks. It takes six weeks to accumulate the "Fitness" data value since it is a rolling average. It will initially say you're consistently "overloading" merely because your "Fatigue" which averages every 2 weeks will finish its average earlier.

Now for someone who recently downloaded Stravisitx who has been using a HR monitor for at least six weeks, once you upload all the past data you should be good to go immediately and don't have to wait six weeks.

So, HR data from the last six weeks = good to go. New to HR data = Stravisitx needs six weeks before useful.

It also means if you get a really erroneous run (like the HR monitor captures cadence instead of HR), then it'll take six weeks to flush that out of the "Fitness" value. Thus your "Form" will be slightly off until six weeks post erroneous data.

 

[kbenson13]

Interesting. So even though I've nicely trended up none of that data means anything! But since I'm just into my seventh week of my training program (and seventh week of HR data) it will all start getting good from here...

 

[DopeyBadger]

Interesting. So even though I've nicely trended up none of that data means anything! But since I'm just into my seventh week of my training program (and seventh week of HR data) it will all start getting good from here...

Since you are in your 7th week of HR data you can now consider the data it shows you currently (as in today's form value) to be an accurate representation.

 

[kleph]

Since you are in your 7th week of HR data you can now consider the data it shows you currently (as in today's form value) to be an accurate representation.

One thing that I was unsure of when I started playing around with the app was if I needed to set a starting value for my fitness and fatigue. It defaults at zero as you see here.



There may, in fact, be a best practice for this I am unaware of. But I found that no matter what initial numbers I put in, they would all eventually fall back to the same levels after a two-month or so period.