Training Load - ATL, CTL, TSB

Gringo78 · 11-03-2010 12:38pm #1

Anyone use the the Training Load Plugin on Sportstracks? I've 'discovered' it lately and now find it a very useful tool for gauging current training levels and planning future training/tapering/peaking.

The science bit is kinda explained in this link:

http://stuartmultisport.com/Documents/Morton%20et%20al,%20Modeling%20human%20performance%20in%20running.pdf

Basically the graph on page 4 shows for a given jump in daily training intensity, you will initially see a performance drop as fatigue due to the extra training load kicks in. Howver, after 47 days you will be back at the performance level you were at when you initially made the jump in training load and from there on you will see a gradual increase in performance from day 47 out to maybe day 120-140. After that you start to stagnate and need to increase the training intensity again to get any further performance increases.

Or, at any time you can lower then intensity (i.e taper) and you will get a temporary jump in performance for about 2-3 weeks due to not being fatigued.

Its an interesting article, if you can wade through the sciency bits but I took a few things from it:

1. Your performance will suffer for about 7 weeks after an increase in weekly mileage or intensity.

2. It'll likely take 60 days before you see a noticeable increase in race times after a big increase in training intensity however, from there on the increase will be continuous (as long as the overtraining doesn't lead to sickness)

3. increasing the volume of training in the 3-4 weeks prior to tapering for a marathon is not really going to do anything for you, you're almost better off training at the same intensity as there wouldn't seem to be a benefit and you risk getting injured. If you're going to increase the volume of your training you need to do it at least 7-8 weeks prior to tapering.

Anyone see any evidence of the above in their own training?

TFBubendorfer · 11-03-2010 12:54pm

Gringo78 wrote: »

1. Your performance will suffer for about 7 weeks after an increase in weekly mileage or intensity.

2. It'll likely take 60 days before you see a noticeable increase in race times after a big increase in training intensity however, from there on the increase will be continuous (as long as the overtraining doesn't lead to sickness)

3. increasing the volume of training in the 3-4 weeks prior to tapering for a marathon is not really going to do anything for you, you're almost better off training at the same intensity as there wouldn't seem to be a benefit and you risk getting injured. If you're going to increase the volume of your training you need to do it at least 7-8 weeks prior to tapering.

Anyone see any evidence of the above in their own training?

These things are only true if your body conforms to the formulas being used in that graph.

I do actually use that plugin but found it a useless predictor for race performance, both for marathons and shorter races. I still have it installed, probably because I'm a numbers geek and like the graphs, but I'm certainly not using it to shape my training or tapering.

tunney · 11-03-2010 1:11pm

ATL, CTL and TSS and the rest do work as predictors of peaks and performance.

But not for running, only for cycling and only when used with regular testing and a power meter.

Sports track seem to have ripped off, (and infringed on copyrights????), WKO and trainingpeaks.

Gringo78 · 11-03-2010 1:14pm

TFBubendorfer wrote: »

These things are only true if your body conforms to the formulas being used in that graph.

I do actually use that plugin but found it a useless predictor for race performance, both for marathons and shorter races. I still have it installed, probably because I'm a numbers geek and like the graphs, but I'm certainly not using it to shape my training or tapering.

True enough, but I find it shows how training is a long range thing, it takes a good 8 weeks for your body to adapt to a sustained training load.

I wouldn't actually use it a performance predictor, but you can compare the training load you were doing in a particular block of training compared to e.g last year. How the daily intensity is made up will of course need to be specific for the event you are training for i.e you can build up a high weekly training intensity by running big long LSR's but if your chosen event is 3km races then that ain't gonna work.

If the training load predictor shows a continual improvement in CTL, you should see a continual improvement in performances, if not you are training wrong. It also shows you when you are fatigued. It is suggested to tweak the numbers depending on the evnt you are training for i.e ATL should be calculated on a rolling 4-5 day basis rather than 7 days if you are training for more long distance events.

Gringo78 · 11-03-2010 1:17pm

tunney wrote: »

But not for running, only for cycling and only when used with regular testing and a power meter.

Does it not have any merits for running?? I mean the numbers it generates are arbitrary and cannot be converted to anything to equate to pace or anything, but they do show the quantity of training you are doing and how long you have been doing it for?

tunney · 11-03-2010 1:37pm

Gringo78 wrote: »

Does it not have any merits for running?? I mean the numbers it generates are arbitrary and cannot be converted to anything to equate to pace or anything, but they do show the quantity of training you are doing and how long you have been doing it for?

How do you measure intensity? What metrics do you use to measure effort?

With cycling you can do tests and get a Functional Threshold Power and then use this FTP to calculate an intensity factory and a training stress score for individual training sessions. A watt is always a watt regardless of whether its windy, uphill, downhill whatever. This numbers are used to calculate the ATL and CTL.

In running all you really have is pace. Doesn't really work as well.

Gringo78 · 11-03-2010 3:14pm

tunney wrote: »

How do you measure intensity? What metrics do you use to measure effort?

I wear a HR monitor on all my runs....the plugin assigns a TRIMP score based on the HR. A higher HR will give an exponentially higher intensity score so 6 miles straight easy will have a lower intensity score than say a 6 mile interval session where the average pace falls out at the same as the 6 mile easy run but the HR is high & low rather than steady.

Again, its not accurate, the numbers are arbitrary, but its a comparison tool. A 7 minute mile will have a lower TRIMP score for me in 2010 than it did in 2009 because my HR is lower now at the saem pace. I think HR is about the only metric you can use for running.

TFBubendorfer · 11-03-2010 4:40pm

Gringo78 wrote: »

I think HR is about the only metric you can use for running.

I think that exactly is the problem. I did one long run early in my training cycle and got a certain TSS score. Towards the end of the training cycle, now in much better shape, I did a longer run at faster pace and lower HR, and accordingly the TSS was a good bit lower. But the equivalent TSS for a bike workout would have been higher rather than lower, because I had created more power per time unit, and over a longer time frame. But there is no present way of measuring power output in running.

I guess that's where Tunney's right, it doesn't work for running.

As an aside, sportstrack themselves didn't rip off WKO because it's a third-party plugin. The guy who wrote it, however, DID rip off WKO. As far as I remember he said as much in the documentation. I think he fiddled with the formula until the graph looked pretty much the same as the WKO one.

tunney · 11-03-2010 5:12pm

TFBubendorfer wrote: »

I think that exactly is the problem. I did one long run early in my training cycle and got a certain TSS score. Towards the end of the training cycle, now in much better shape, I did a longer run at faster pace and lower HR, and accordingly the TSS was a good bit lower. But the equivalent TSS for a bike workout would have been higher rather than lower, because I had created more power per time unit, and over a longer time frame. But there is no present way of measuring power output in running.

Actually no - assuming constant retesting and re-evaluation of FTP then there would be a reduction in TSS for the second bike.

TFBubendorfer wrote: »

I guess that's where Tunney's right, it doesn't work for running.

Functional Threshold Pace - sorta works (its done off pace rather than HR) but then doesn't work as wind, surface and all the rest affect pace (gradient is included in the calculation) and its sorta useless.

The reason I am saying it does work for running is there is no way to measure the work done running.

TFBubendorfer wrote: »

As an aside, sportstrack themselves didn't rip off WKO because it's a third-party plugin. The guy who wrote it, however, DID rip off WKO. As far as I remember he said as much in the documentation. I think he fiddled with the formula until the graph looked pretty much the same as the WKO one.

RedB · 11-03-2010 5:36pm

tunney wrote: »

The reason I am saying it does work for running is there is no way to measure the work done running.

Does the calorie (Kcal) reading on HRM's not give this or is it too rough to be meaningful?

Gringo78 · 12-03-2010 12:15pm

RedB wrote: »

Does the calorie (Kcal) reading on HRM's not give this or is it too rough to be meaningful?

The kcal reading will just tell you how much you burned, but not the rate or intensity at which you burned it. Its equivalent to really the distance run, no more.

TFBubendorfer wrote: »

I think that exactly is the problem. I did one long run early in my training cycle and got a certain TSS score. Towards the end of the training cycle, now in much better shape, I did a longer run at faster pace and lower HR, and accordingly the TSS was a good bit lower. But the equivalent TSS for a bike workout would have been higher rather than lower, because I had created more power per time unit, and over a longer time frame.

Thats the point though, you are faster now but the longer run, at a faster pace, and at a lower HR was easier for you to do, stressed you less and actually gave rise to less training effect than the long run early in your training cycle.

Two runners go out and run 8 miles at 7min/ml pace - Runner A is a faster runner (say 2:40 marathoner) than the other and so for Runner A its an easy paced run, for Runner B (3:05 marathoner) its a PMP run with a higher stress score. HR would tell you that. If Runner A repeats that run every day for a week, he's tapering and will be nice and fresh, if Runner B repeats that run for a week he'll be in a heap after 7 days.

Training Load - ATL, CTL, TSB

Comments