TSS & IF versus training adaptions

Since 2001, over 90,000 cyclist's have joined Cycling Forums to discuss topics from general cycling to equipment, training, racing and travel or vacation destinations (especially in europe during the tour de france). We also feature an great deals in our online store, 100's of articles, classifieds and product reviews.

I have just started using TSS and IF to track my training and am wondering if someone can speak to the difference between training effect and TSS. That is, if I accumulate TSS points, can I presume there is a strong training effect, and, conversely, if I don't get many TSS points in a week, that there isn't a good training effect? I understand it primarily is a way of preventing overtraining and tracking training load. I am trying to understand the link between load and effect. It seems to me that a well structured 10 hour a week program can get more training effect than a poorly structured 20 hour a week program despite having less TSS points. (My training is closer to 20 hours, this is just an example). Similarly, it seems to me that the training effect is going to be dictated by the content of the training (Is it Tempo, Threshold, VO2max, Anaerobic Power, Sprints, etc) more than the overall load. I am obviously going to avoid weeks where TSS>1000. Is there anything else I can/should use them to monitor. Provided I focus on the correct things in my build (get the right intervals done at the right time), should i then go about accumulating more TSS points with tempo and endurance rides to enhance FT or is there an argument for just resting to maximize supercompensation from intervals?

Training Stress Score measures just that -- stress. The effect that the stress (or load) produces within the body depends on the nature of the stress (eg. IF). 2 rides with equal TSS values may produce quite different effects on the body if their IF values are dissimilar.

I think you should watch both IF and TSS for a typical buildup training plan. While the overall IF would probably build over time, TSS could either be steady, increasing slowly during the build phase, or a series of building steps with rest weeks in between mesocycles, depending upon your training approach. A nice thing about TSS is that it allows you to not lose track of your training load as the intensity changes or new workouts are substituted for others along the way. In that sense, TSS isn't the primary focus of a structured training plan, but keeping an eye on it helps you keep the mix about right when you start adding intensity.

Very interesting question. I don't claim to be an expert, but this is how I use the various charts and measures. My primary training objective is a target number of minutes in L4-L7 per ride and per week and I use a separate chart to track that. It can be misleading if I get minutes in a level that is not part of an interval (e.g., a brief acceleration on a short climb), but I manage that on the ride itself. I use IF to measure the intensity of a given ride, with an expectation of a degradation in IF as a function of total ride duration. Actually, I plan rides to attain a target NP and IF. And I use TSS exclusively to measure total stress, with a 1000 pt cutpoint for focus (i.e., <1000 pts, I'm not worried about overtraining so I don't really care what the actual number is).

should i then go about accumulating more TSS points with tempo and endurance rides to enhance FT or is there an argument for just resting to maximize supercompensation from intervals?I don't. As far as I am concerned, the only thing that matters is total time at L4-L7, unless I want to just enjoy the scenery.

I don't. As far as I am concerned, the only thing that matters is total time at L4-L7, unless I want to just enjoy the scenery.

Well I guess that focus is specific to whatever events you are targeting? I am concerned primarily to increase FT as this is the main limiter for my target events. Consequently I am doing the bulk of training emphasising NP ~L3 but without being tied down to a certain zone. For example, on rolling terrain I'll push quite hard up short hills, etc. I'll also do some 2X20's, etc. I'm going to do the more specific L5+ stuff later, though I end up spending some time in those zones anyway from riding rolling terrain and doing group rides.

This is what I'm all about for now: http://www.rongilcreast.com/ACoggan_New_Training_Levels_chart.htm

Well I guess that focus is specific to whatever events you are targeting?Well, to a limited extent, yes. But, like you, I am primarily focused on increasing FT and the core determinants are L4 and L5 intervals. I just don't believe I increase power at L3 or below. For me, the primary purpose of a training ride is time in L4-L7. Most of my rides are 3+ hours and an IF of .85 is pretty high relative to my MP/duration curve. But, if I rode for the entire ride at a constant power of 85%FT, I would consider it a largely worthless ride. My view of a training ride is it's a string of L4-L7 intervals with some other junk thrown in between the intervals.

This is what I'm all about for now: http://www.rongilcreast.com/ACoggan_New_Training_Levels_chart.htm
Is there any basis behind that figure? The strain and volume curves appear to be somewhat arbitrary mirror images of each other, and the effect curve doesn't appear to be a mathematic combination of the other 2 curves at all. I'm sure it's intended as a diagram to point to the location of the 'Sweet Spot' but it could point anywhere on the figure depending upon how the effect curve is drawn. For example, the apex of the effect curve would be elsewhere if you multiplied the strain value by the volume value at each intensity.

I've considered taking my Max Power v. Duration curve and plotting it out as Max TSS v. Duration graph to see what intensity would allow someone to achieve the maximum TSS for a ride. Would that indicate an individual's 'Sweet Spot'?

I've considered taking my Max Power v. Duration curve and plotting it out as Max TSS v. Duration graph to see what intensity would allow someone to achieve the maximum TSS for a ride. Would that indicate an individual's 'Sweet Spot'?Couldn't you just plot max IF vs. duration for the same result?

Couldn't you just plot max IF vs. duration for the same result?No, I don't believe so. That'd tell you that high IF's are possible for short durations and low IF's at long durations, but not much else. Essentially, that'd just normalize the Mean Max Power curve to FT.

In general, the higher the IF for a given TSS, the greater the training effect.

SST seems to balance the two really well.

SST seems to balance the two really well.
:confused: Is that the inverse of TSS?

:confused: Is that the inverse of TSS?
sweet spot training

Is there any basis behind that figure?

It's a theoretical construct meant mostly as a teaching tool ("a picture is worth a thousand words"). That said, my gut says that it portrays reality fairly accurately, even if I can't prove it.

The strain and volume curves appear to be somewhat arbitrary mirror images of each other

They are mirror images, but their shape wasn't chosen arbitrarily: they both are 4th order functions, based on the fact that not only does metabolic strain increase at approximately that rate with increasing intensity, but time-to-fatigue also decreases at approximately the same (inverse) rate.

and the effect curve doesn't appear to be a mathematic combination of the other 2 curves at all. I'm sure it's intended as a diagram to point to the location of the 'Sweet Spot' but it could point anywhere on the figure depending upon how the effect curve is drawn. For example, the apex of the effect curve would be elsewhere if you multiplied the strain value by the volume value at each intensity.

I've played around a lot in Excel with how to best present the effect curve, but can't say that I'm really pleased (or set upon) any particular representation as of yet. The key point, however, is that when you consider the trade-off between intensity and volume that there seems to be a "sweet spot" that lies somewhere at or below 100% functional threshold power, but above the level 2/3 border.

've considered taking my Max Power v. Duration curve and plotting it out as Max TSS v. Duration graph to see what intensity would allow someone to achieve the maximum TSS for a ride. Would that indicate an individual's 'Sweet Spot'?

No, what that would show is that as you ride for longer and longer periods, you "earn" more TSS, but not in a linear fashion since intensity must be reduced. There's a difference, though, between the stress imposed by training (which is what TSS is intended to reflect), and the resultant improvements in performance. IOW, more is not always better...

I have just started using TSS and IF to track my training and am wondering if someone can speak to the difference between training effect and TSS. That is, if I accumulate TSS points, can I presume there is a strong training effect, and, conversely, if I don't get many TSS points in a week, that there isn't a good training effect? I understand it primarily is a way of preventing overtraining and tracking training load. I am trying to understand the link between load and effect. It seems to me that a well structured 10 hour a week program can get more training effect than a poorly structured 20 hour a week program despite having less TSS points. (My training is closer to 20 hours, this is just an example). Similarly, it seems to me that the training effect is going to be dictated by the content of the training (Is it Tempo, Threshold, VO2max, Anaerobic Power, Sprints, etc) more than the overall load.

In general I think you grasp the concepts: TSS is meant to reflect/help you manage your overall training load, but the "composition" of those TSS points is what determines how much you improve in terms of any particular ability. For example, a pursuiter might maintain a very high average daily TSS of ~120-130 points/d by doing lots of level 5/6/7 intervals, but that isn't necessarily the best way to prepare for, say, a stage race or longer TT. Conversely, the same individual might rack up the same ~120-130 points/d doing lots of "sweet spot" training and be essentially equally stressed, but would perform much better in longer events (but not as well in a pursuit).

To put it another way: training stress score does not = training performance score.

In general I think you grasp the concepts: TSS is meant to reflect/help you manage your overall training load, but the "composition" of those TSS points is what determines how much you improve in terms of any particular ability. For example, a pursuiter might maintain a very high average daily TSS of ~120-130 points/d by doing lots of level 5/6/7 intervals, but that isn't necessarily the best way to prepare for, say, a stage race or longer TT. Conversely, the same individual might rack up the same ~120-130 points/d doing lots of "sweet spot" training and be essentially equally stressed, but would perform much better in longer events (but not as well in a pursuit).

To put it another way: training stress score does not = training performance score.
Is it fair to compare training in different levels, but that have the same desired effect, using tss? For example, can you compare tss from a "sweet spot" workout vs. a "pure" threshold workout (intervals done at ft or very near) and conclude that one offers more training affect than the other based on the tss being higher?

I am not sure I understand that sweet spot thing. That is level 3 training. Optimal Level 3 training aimed at lactate threshold power improvement?

Is it fair to compare training in different levels, but that have the same desired effect, using tss? For example, can you compare tss from a "sweet spot" workout vs. a "pure" threshold workout (intervals done at ft or very near) and conclude that one offers more training affect than the other based on the tss being higher?

I would certainly say that that would be fairER, but I wouldn't go so far as to argue that it is a completely fair comparison. First, TSS is predicated on the fairly simple notion that there is an interaction between intensity and duration, and therefore we should multiply them together. It provides a quite reasonable metric by which to assess overall training load, but I wouldn't claim that it is perfect. Second, at some point you will almost inevitably have to raise the intensity of your training to further increase your functional threshold power, even if it means reducing your overall training load slightly.

I've played around a lot in Excel with how to best present the effect curve, but can't say that I'm really pleased (or set upon) any particular representation as of yet. The key point, however, is that when you consider the trade-off between intensity and volume that there seems to be a "sweet spot" that lies somewhere at or below 100% functional threshold power, but above the level 2/3 border.
Ok, I think the relative impact of the different intensities (levels) on FT improvement is what I temporarily lost sight of. In that sense, the 'effect' curve is not a combination of volume and strain, but rather volume and a quantified degree of adaptation by level (ie, Andy's chart with the check marks). I could see the 'degree of adaptation vs. intensity' being represented by an 'S' curve starting low in L1, rising to a deflection point somewhere in lower L3, and flattening out in upper L4. Mathematically combining the 'S' curve with the volume curve would produce a peaking 'effect' curve similar to what is currently shown in the diagram, and might help explain why there is a 'sweet spot' at all. IIRC the accompanying article does explain it somewhat, but this might help make the diagram more 'stand-alone'.


No, what that would show is that as you ride for longer and longer periods, you "earn" more TSS, but not in a linear fashion since intensity must be reduced. There's a difference, though, between the stress imposed by training (which is what TSS is intended to reflect), and the resultant improvements in performance. IOW, more is not always better...
....because TSS does not necessarily address the degree of adaptation, sure. Thanks for the insight.

Ok, I think the relative impact of the different intensities (levels) on FT improvement is what I temporarily lost sight of. In that sense, the 'effect' curve is not a combination of volume and strain, but rather volume and a quantified degree of adaptation by level (ie, Andy's chart with the check marks). I could see the 'degree of adaptation vs. intensity' being represented by an 'S' curve starting low in L1, rising to a deflection point somewhere in lower L3, and flattening out in upper L4. Mathematically combining the 'S' curve with the volume curve would produce a peaking 'effect' curve similar to what is currently shown in the diagram, and might help explain why there is a 'sweet spot' at all. IIRC the accompanying article does explain it somewhat, but this might help make the diagram more 'stand-alone'.



....because TSS does not necessarily address the degree of adaptation, sure. Thanks for the insight.Andy,
Is this all detailed in the book? :)

Nice thread BTW ... good questions and replies ..

FWIW, I've been slowly/quietly building back my form since last summer and without doing a single 'real' L4 workout since September. That is, nothing over 0.95 IF for an extended period time. But I have been doing as much Upper L3 and a moderate amount of Lower L4 as I can stand (definitely SST and lots of TSS). Zero L5/6/7.

My FT is up from my last TT in late September by some 25-30W ... to me that proves that moderate intensity but substantial duration (SST) works. I had similar results last fall/winter -- upped threshold power without doing any real threshold work. Feels like cheating sometimes :)

I did my 1st 2x20 efforts last night ... set new short-term PB's for 10/15/20min on the 1st one which was quite a nice surprise (on -6 TSB as well so I was not overly fresh) Hopefully, by now increasing the intensity as AC mentioned, I can tack on another 20-30W by mid-April (~12 weeks of hardcore threshold training).

I give full credit to TSTWKT and SST concepts in planning my training. We still have to get on the saddle and do it but having confidence that it WILL work makes it easier to get through hours 3/4 of a hard tempo ride.

rmur

Oh sure, throw your prototype TSB's and TSTWKT's in our faces while we fumble along in the dark.... :D ;)

Any word yet on it will be released? :)