You need to set up your swim zones in TP, just like you would for running or biking. They're based on sec/100m.

Not strictly true (I don't think) but you are in the right ball park. The zones are irrelevant other than for calculating things like 'Time spent in X zone'. The important metric to set up is threshold pace.

---

https://www.pbandjcoaching.com
https://www.thisbigroadtrip.com

I think Training Peaks is great. For individual sports the TSS - and because you are interested in that i assume you are also interested in the Performance Management Chart (PMC) - are quite accurate. Adding in running and cycling starts to make it more of an art than a science as one sport does not 100% impact the other yet the TSS build would suggest that it does. Running and cycling somewhat gel together, but swimming doesn't

When you add swimming I think it completely skews it. A TSS of say 80 in swimming adds on to a TSS of say 80 for a bike. Yet they they are effecting different systems so the cumulation is not focussed on one area and therefore a combined TSS of 160 would not represent a true cost. Cycling and swimming hit very different elements of the body.

A solution that works for me (and by that I mean gets me in a ball park) is not uploading data for the swim, but instead entering the TOTAL TIME IN THE POOL and the TOTAL DISTANCE of the swim. By doing this it adds in all the breaks and rests whereas your watch ignore them . Adding these in reduces the TSS by quite a chunk because your average pace is lowered a lot. This is a good thing, as the TSS becomes more realistic towards the actual cost of the swim within the framework of your other training. Not perfect, but I found it to be a better indication.

I have done this with all my athletes in Training Peaks and found that the PMC plays out fairly well in comparison to how they are feeling.

Definitely more of an art than science but I have found it works pretty well that way.

---

https://www.pbandjcoaching.com
https://www.thisbigroadtrip.com

You just need to set your swim threshold pace in Trainingpeaks. This is the pace that you can hold for a 1000M or 1000y TT in the pool when fairly well rested (like middle or end of a recovery week). You then record your MOVING time and total distance. Best to just exclude kicking and drills, or sort of just average it in at a relative swim intensity into the moving time.

OWS of course will have more variability, but then again, your moving time matches your total training time so it balances out a little.

---

TrainingBible Coaching
http://www.trainingbible.com

I have an excel spreadsheet I used since I could never figure out how to get TP to do it for me. DM me and I'll be happy to send it to you. It's pretty basic but it gets the job done.

It is not completely random, but I generally assign a TSS value per 1,000. It depends on the speed of the athlete. Faster athletes get a lower TSS per 1,000. I might adjust that up or down slightly based on the type of workout.

Picking round numbers here. A faster athlete might 'generally' complete a 4k session in 1 hour. So, they might get a TSS of 20 per 1k of workout. A slower athlete might complete 3k in 1 hour. So, they might get a TSS of 25 per 1,000. If it is a 'harder' swim I might bump the total up by 5. An 'easier' swim I might lower the total for the workout by 5.

This gets away from the question of how long did the swim take, is rest included, etc. I've not looked closely at the Garmin swim files. But, if you kick during the workout and it's not included then what. Mostly for the swim, I want an approximation that is consistent. And, I don't use TSS as an end all be all for training. More of guideline or loose guideline. I think that the key is to be consistent with the metric. If the feedback is 'that was the hardest 3.5k swim I've done'...then maybe it gets a bump up in the total.

Edit...after ALL of your years of training you want a metric now!!?

---

Brandon Marsh - Website | @BrandonMarshTX | RokaSports | 1stEndurance | ATC Bikeshop |

Same as Brandon. Every workout I do, I set it as being at 10s slower on average than my approximate threshold pace. So for me, my threshold is set at 1:20/100m, which is a bit slower (it's more like 1:15/16) than what it actually is for swimming, but which makes sense if you think of it as more like "pace for a 1hr TT" (aka "swim FTP," which doesn't really exist, as opposed to swim threshold, which means more like 400-800m pace or so). That keeps it roughly inline with the other sports. Then I simply do all of my workouts at a theoretical average pace of 1:30/100m.

This gives a sTSS score that roughly matches up with my other sports. Speaking with other athletes, this is pretty similar to the approach that they have used. Basically, you just come up with a consistent conversion for volume->sTSS.

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

I enter total time + an estimated average hr, for most workouts I enter HR equivalent to 65% of max HR, if it is an easy distance set I enter 50%.

Thus 1hr swim usually equates to 40tss.

What would be wrong with this rough method?

My goal is just to have swim included in the total CTL approximaton of training load. Have entered swims the same way for last years so numbers are comparable.

Similar to what Brandon does, I'm just too lazy to work that out everyday so I do the following. Note how at the end of the day his faster swimmer example got 80 TSS for the hour, and how the slower swimmer got 75 TSS. I got into the habit of looking at that as 1 TSS per minute MOVING plus a fudge factor for how hard the swim was as a whole. And I only try to gauge that after a little while, from 5 hours to a day. As a very general rule, every swim gets a 10 TSS "bonus". But if it was really easy I tone that down to 5 TSS, if really hard I bump it up to 15 TSS. I have been using WKO+ version 3 since it was first released (and version 2 before that). Once I import the swim to WKO+, it shows me my MOVING time right away as I use the rest timer on my Garmin. So it is pretty straight forward to come up with a consistent "guess" that works for the whole training period. WKO version 4 is something I'm still trying to understand why it has been released publicly.

---

After millions of years of selective breeding, here we are!

OMG I am so happy this discussion is taking place. I've always ignored the TSS for swim on TrainingPeaks. My last swim of 2,300 meters gave me a sTSS of 2.3 (I don't have a HRM for the pool). It makes my PMC look funny if I include swim. I'm fixing this straight away. Duh. Why didn't I think of this years ago?

---

Hillary Trout
San Luis Obispo, CA

Your trip is short. Make the most of it.
https://www.slogoing.net/

since TSS was designed for biking, I don't think the scoring of swims is yet accurate enough to be relevant. Your swim TSS also has a much lower correlation to swim performance. For that reason, I base my training decisions on bike + run TSS only (I exclude the swim). I still record and track swim TSS, but do so separately, and purely to see interesting trends or correlations.

To answer your question, you should enter an accurate swim threshold speed in your zones setup, calculated by doing a 1500 TT (or the easier, predictive method of a CSS test 400/200 TT). But, as I said, I would highly recommend also having a run/bike PMC that excludes swim.

---

____________________________________

Are you ready to do an Ultraman? | How I calculate Ironman race fueling | Strength Training for Athletes |

I'm with you. Total time in the pool including rest/kick and whatever else intrigues me. Total distance. That's it.

An hour swim usually gets me in 30 sTSS range. (Although I haven't checked my threshold pace in a while..)

---

Team Zoot - Great Lakes

Some real examples. I've been swimming with my Garmin just to check. This most recent workout had 30x50, and I was too lazy to stop/start each lap, so I just ran it as a straight 1500, which gave me a lower sTSS for that particular part of the workout than I would have had otherwise. But still, you can see the values using the "pick a consistent pace about 10s/100 slower than threshold and just use that for the whole workout" works quite well. Comparing across other workouts, I'd say this was consistent. Within a few % in either direction...

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

There is so much noise in swimming I don't see how any real value can be gained through a scoring algorithm.

Examples of the noise (and this is just a start):
1) The push off of first 25 of any send off is done with shoulders and body aligned everyone swim significantly faster on their first lap.
2) Oxygenation always becomes the limiter. By "how much" and at what point is anyone's guess.** But either way on a set of 100's you are most certainly more oxygenated the first 50 of each send off than the second.

Take those two factors and this makes 24 x 50's on :50 holding :35 significantly easier than 18 x 75's on 1:05 holding 52.5 (and 8 x 150's on 2:30 holding 1:45 is a different same planet.).

The only effective way I can see coming up with an algorithm is to use some combination of HR, perceived effort and then verifying through lactate acid testing.

** Gary Hall did some experimenting where the swimmer would expel all oxygen underwater, hold for a few seconds and then push off. I think he was trying to train his swimmers to sprint purely anaerobically. But to anyone reading this thinking "hmmm, smart" don't do it. Because it will likely lead to your untimely death through shallow-water blackout.

Reminds me of the standard but very unpleasant - "25 underwater - 25 sprint" 50s...

EDIT: though to add to your larger question, I think an algorithm can be useful "enough" especially for triathletes who tend to train in a *relatively* consistent manner. I.e., triathletes don't typically do a ton of anaerobic stuff. And, likewise, I think intervals tend to be structured in relatively the same way FOR A GIVEN ATHLETE. For me, I rarely do interval sets with intervals longer than 200. And anything really long almost always tends to be pull. Having spent some time with a very good swim club, my own training tends to be much more confined in terms of its scope than theirs. I can see how a metric for them would largely be impossible. But for me? It seems to work "well enough." My swim fitness is a pretty direct correlation to how much I'm swimming, because what I do when I'm swimming doesn't really vary a huge amount. That's true of both running and biking as well.

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

Unrelated to the main topic of the thread, but we used to do "blow out" sprints all the time when I was in college (this was before shallow water blackout was a commonly know thing in the swimming community). While the anaerobic sprinting is part of it, it was as much or more about body position and body line, as emptying your lungs removed a huge buoyant force from your body.

---

I wrote this, you should read it:
https://www.slowtwitch.com/...n_Swimming_6700.html

I am just a few years older than you (well maybe, I turned 40 yesterday) so they hadn't "caught on" when I was in college. I am rather alarmed that they ever caught on.

In any event: as a former world class swimmer what were your experiences measuring training load and fitness? Did coach on the elite level try and do this?

The SwimScore algorithm accounts for oxygen uptake kinetic base on interval and rest durations:



It gives different scores for these workouts, for example using 1:15 as threshold pace:

• 24 x 50's on :50 holding :35 -> SwimScore = 25 and xPace = 1:18
• 8 x 150's on 2:30 holding 1:45 -> SwimScore = 31 and xPace = 1:16

That's 24% higher for the second one, but I don't know if this reflect the difference you are talking about.

---

Ale Martinez
www.amtriathlon.com

this is something that I didn't know about. I read SwimTSS at training peaks and just shrugged. I'd love to read more about it.

So experience has taught me that 24 x 50s on :50 at :35 would be like 8 x 150s on 2:30 at 1:53.... what does the table say? (if it matters, SCM)

I think the blow outs were never for more than about 15m or so, but dumb nonetheless. My first experience with blow-outs was around 2004ish, and I retired in 2008 when I was 25.

I never swam somewhere that kept sophisticated track of load/fitness. In high school we did a lot of test sets (6x100 on 6:00, 6x200 on 8:00, T30, 18x100, on 1:40, 8x50 on 3:00) that were recorded and posted, so you could track progress over a season. College testing was more simple, but not recorded - Wednesdays were "dive day" with sets varying over the course of the year from 6x100's on 6 to 4x50 on complete recovery, varying by time of the season. Friday's were "V02 max" days - 5x{3x50 on :35 best average, 150 easy} or some slight variation. Post grad/Pro - David Marsh just kind of looked you in they eye and gave you a workout for the day. There was general structure to the season, but no predictability to the testing/loading/unloading of workouts.

As far as testing physiological parameters, at swim meets, USA Swimming Nat'l Team staff would take ear-blood for lactate post race, and you had to warm down til you were below some threshold (I don't remember the numbers) but at various points in the season, clearing took 10-40 minutes, depending on the race etc. I also got tested during dive sets at the OTC in Colorado Springs, to gage response/clearance ability, but to my knowledge the numbers were never recorded.

I've been out of the sport for 8 years now, so I'm sure that things have changed, and as the sport becomes more professional, more care is taken to record and track. For me it was practice times, volume and feel that were the indicators of training. Feel being the most overrated but easiest to fall back on.

Feel is crap. Don't trust it. Best times feeling like dog crap, dog crap times feeling like $1,000,000.

---

I wrote this, you should read it:
https://www.slowtwitch.com/...n_Swimming_6700.html

Thanks for the info. Alright, lets get to what is important: best set ever in practice. No hardest. Best.

I did 2 x 50's from a push on 6 min @ 23.2, 22.8

I know you DESTROYED THAT, so lay it out!

PS: You probably already know this but USA swimming did a lactate testing on over 5,000 swimmers. Care to guess who had the lowest lactate acid level by something like 20 standard deviations. (I am not exaggerating. The avg was smoothing like 11 Millimoles, STD of .3 and Phelps (the obvious answer) came in a 5.5). Though someone can probably look that up because I am going by memory..

I've got a few that spring to mind.

I was a 400im'er in college, 200 imer as a pro - dabbled in backstroke and mid-d free.

6x100 on 6:00 freestyle all 45.9 or better from the blocks scy
10x400 pull Pads/Bouy 5 on 4:00 5 on 3:55 averaging 3:40 scy
10x100 kick averaging 59 on 1:05 at 7500 ft altitude scy
4x[400IM on 5:00, 4x100 Fr on 1:20] 400im's all under 4:40 lcm

---

I wrote this, you should read it:
https://www.slowtwitch.com/...n_Swimming_6700.html

Regarding those 2 examples. Is that the sTSS that shows up in TrainingPeaks?

I honestly don't know how it's calculated, etc. I end up changing it for my athletes who have a garmin that tracks it.

That's a sTSS of 25 and 31 for a 1200 swim? I guess that I could see that since the swimmer is going faster than threshold for 14ish minutes.

Makes it complicated!! I like the estimation method. :)

---

Brandon Marsh - Website | @BrandonMarshTX | RokaSports | 1stEndurance | ATC Bikeshop |

With the above assumptions it gives me a little lower score (22), the exact same score happens for 8x150 on 2:30 at 1:50, it seems to be in the ballpark!

There was a nice pdf by Dr. Phil Skiba about SwimScore, I can't find it on-line but I cant send you a copy if you give me an e-mail, my review (in spanish) is here. The algorithm is implemented in RaceDay and GoldenCheetah.

---

Ale Martinez
www.amtriathlon.com