Analysing a Triathlon Season – One Last Look at 2010

There’s a lot to be gained from understanding what contributed to success (or failure) in the past. Typically fuelled by insecurities about performance I bolster my confidence searching for patterns in my present training that match past success. If I pay attention my training diaries can show me the sessions and schedule I respond well to

Despite the season being well and truly over I’ve not spent much time considering what lessons can be learnt from it. I scratched the surface in my review for Endurance Corner just considering the Performance Management Chart (PMC). I’d have left it there were it not for further questions on what to look for when examining training logs. The PMC is a good starting point, but there was more to be learnt by picking through the data in WKO.

The Performance Management Chart

2010 Performance Management Chart

A familiar chart on these pages, though I’m not sure if 2010 has made it here in its entirety before. Rather than revisit specifics from the year (you can read that in the Endurance Corner column) I’m going to consider how I use this chart to assess training and what I can take forward from this. There are a few simple principles I can learn from the past year and use to help shape 2011.

I particularly focus on Training Stress Balance (TSB) as a good indicator of periods of intense training and their impact on recovery. TSB drops as I apply increasing training loads, I expect to spend time with negative TSB whilst in a solid training block. Periods of negative TSB are a requirement of developing fitness, how they are managed is what maters. The further TSB dips or longer spent there the greater the impact on performance and recovery.

  • TSB < -70: guarantees I’ll need a significant recovery period shortly after, often achieving TSBs as high as 50.
  • TSB -70 – -50: has less impact on recovery in the short term, but longer periods likely require similar recovery to TSBs below -70.
  • TSB -30 – 0: is typical of a good training block, recovery will be needed at points, but nothing of significant impact.
  • TSB 0 – 10: gives more chance of a surprising performance in training. It’s a good place to be for challenging sessions or training races.
  • TSB > 10: has generally worked well for races. Too high and whilst freshness makes me feel good, lost fitness will impact performance.

In general a higher Chronic Training Load (CTL) can lead to a better performance, but the higher the number the greater the work to reach and sustain it. On a fulltime training schedule I could reach numbers that couldn’t be managed with more limited time, I had could do the work and importantly recover from it. Whilst high CTL is a good thing it’s hard to predict actual performance from it. On a limited schedule perhaps I’ll reach 110 rather than 150, but I’ve seen I can get good results at either point.

It’s the process of building a higher CTL that counts, the training I perform to reach higher fitness and my ability to sustain this level (and progress beyond it). This leads to race performance so when tracking CTL I should consider what the rate of growth, plateaus and periods when it falls tell me. What lead to an unplanned decline, was it something I did in training? Had I spent too long at low TSB and been forced to break consistency? These are the questions I consider to help guide the season ahead.

Weekly TSS

2010 Weekly TSS

Looking deeper I can consider the training stress (TSS) for each sport over the year. Higher training stress leads to greater increases in fitness at the price of fatigue. High training stress quickly raises CTL and quickly drops TSB, too much or too fast and the fatigue is high. Weeks of exceptional TSS will likely lead to significant recovery and a period of much lower TSS.

  • TSS > 1500: requires recovery in the following weeks, TSB is likely to have heavily dipped.
  • TSS 900 – 1200: is sustainable as part of a serious build. A month in this region followed by a taper leads to a good race.
  • TSS 700 – 1000: is manageable all year round on a sensible schedule, this is where the bulk of a longer build lies.
  • TSS 400 – 700: an easier week during a build or a decent week back from a recovery period.
  • TSS < 400: an easy week, casual sessions during recovery periods.

Training stress is a product of duration and intensity, there’s only so far you can take either in a week. Higher TSS weeks require more time and are only realistically manageable in short blocks or on training camps. I know that 700-1000 TSS is an achievable weekly target that offers good returns in terms of fitness and recovery. As races approach I can consider a peak with an increase in weekly TSS towards 1200. There’s no harm planning occasional serious week so long as I’m prepared to follow with an easier recovery period.

Weekly Power Distribution Chart

2010 Weekly Power Distribution and TSS

Breaking each week into the percentage of time in each of the Coggan Power Levels gives a more detailed look at my bike training. I’ve overlaid bike and run TSS to give an impression of the workload involved at the time. Not racing with power unfortunately means some data is missing (along with a couple of weeks of flat batteries!) Previous charts considered the impact of training on fitness and fatigue, this is about the composition.

Much of the year is spent with roughly similar distributions of zones. When I wasn’t focussing on the bike (June/July) or training volume was high (January/February) lower power zones dominate. During more bike focussed blocks there’s a shift towards slightly higher levels with a progressive increase of work in tempo and threshold. I consistently spend a small proportion of time above my threshold.

Largely through choice ride time will be more limited in 2011, this naturally shifts my approach to bike training. Training stress will drop, but I can control the reduction by increasing intensity. At lower volumes I’m in a better position to recover from each session and perform the next well. My ability to ride far is established, but my ability to ride hard could use more work. Managing the composition of bike (or run) training helps minimise the impact of reduced volume and allows a focus on specific aspects of fitness.


I picked the charts above because they tell me about how much I trained and what I did with the time. I can then this back to the PMC and see the impact it had in terms of fatigue, recovery and performance. Planning the season ahead I consider what delivered a good level of CTL, enable consistency by avoiding fatigue and contained the right balance of intensity for my goals. The result is a set of principles to help guide me.

  • CTL around 110 – 120 is a good level to race at.
  • TSB below -50 should be avoided except on training camps.
  • The majority of my build should work towards a weekly TSS cap of 1000
  • The final month of race preparation should raise the cap to around 1200
  • Composition of power training should increase emphasis on Threshold and Tempo at a reduced volume.
  • Growth in CTL should be kept gradual over the six months to avoid spikes in fatigue and potential impact on consistency

The most important lesson from 2010 though is sometimes it’s easy to focus on the numbers, the heart rate, power, fitness or training stress. Record the data, plan sessions, set goals, but most of all enjoy the process. Training should always be fun.

Ironman Training Library

From nutrition to pacing - a collection of CoachCox blog posts focused specifically on Ironman training and racing.


  • Brett

    Hey Russ, love you blog. I got onto it through Endurance Corner who I am now working with and it is pretty useful, thanks. I was wondering if you log your swim training in WKO? If so you got any tips on the best way to do it and approximate TSS scores etc?

  • Hi Brett,

    Thanks for reading, glad you like the blog.

    I log swim training in WKO, but have to be honest and admit I log time, distance and session details, I’ve not yet attempted to estimate TSS for each session. However it is something I should be doing and as we’re in a new year, I’m going to take the opportunity to start.

    TSS is modelling the impact that training has upon our bodies in terms of improving fitness and creating fatigue so however we calculate it the results need to tally well with this. With the bike and run we’ve power and pace to determine it by relating the values in training to our threshold values we can calculate a more accurate figure. Swimming we can utilise something like Critical Swim Speed (CSS) to represent a threshold pace and as a basis for all calculations.

    That wouldn’t be a problem if all we ever did in swimming is get in the pool, swim for an hour and get out. You could take your pace, compare against threshold and produce a TSS for swimming. It gets more complicated because we have sets and potentially a variety of strokes. There’s going to be a degree of estimation in there. Some possible options:

    • Quick and dirty: Assume a TSS per hour of swimming – estimate TSS for each session based on this rate.
    • Quick and less dirty: Assume a TSS per 100m of swimming – estimate TSS for each session based on this rate (slower swimming/more rest means lower TSS).
    • Slow and more accurate: Assume a TSS per 100m swum at CSS – calculate TSS for pace of each 100m swum (increases impact of harder swimming).

    There’s still the issue of drills, kick and other strokes of course. My inclination is to try a version based around number 2. Have an estimated TSS per 100m of drill, kick and each stroke. Break down sessions into their components and sum up the different TSS. It’s far from perfect, but I feel that would be more than accurate enough. I’m going to give it a try over the coming weeks and report back.


  • Willem

    I was left wondering how many training hours are associated with their TSS/CTL values.
    For example, in your Kona build, one can see a stable TSS score of 500+ and a CTL of around 110. How many training hours did go into that training block?

    I often get the impression that the whole performance management chart gives “beter” results if one is not a time-crunched athlete but has significant time to put a lot of volume (which gives a relative high amount of TSS).

    — Willem

  • Hi,

    It’s been quite a while since I produced this data (or trained to that level for that matter). Without exact figures too hand I can tell you at that point in time I wasn’t working full time and had a lot of opportunities to train. I’d also had a couple of years training regularly at 20+ hours per week so had the capacity to handle high training loads each week. The reality is, if I was working a regular job, it would have been much harder to achieve the hours required to hit that level of CTL consistently.

    I’d largely agree, the impact of volume on CTL tends to outweigh the impact of intensity. The time-crunched athlete is also going to have limits on how much intensity they can handle and hence how much TSS they can gain from doing so. If you’re limited to 10 hours a week, you’re not going to be able to increase the training load of those 10 hours in a way to match the training load of someone training 16 hours per week.

    The majority of age groupers I now coach are sufficiently time limited that there will be limits to their TSS accumulation. I try not to focus on specific values of CTL. While it’s true that higher CTL probably equals greater fitness (simply because it takes more training to achieve), it also matters how the CTL is achieved. Volume alone won’t cut it. So I aim to put together a program built around a mix of volume and intensity that should yield consistent growth in CTL for the athlete. We may have a limit on how it can go, but we train as effectively as we can in the time available.

    If you’re looking to Kona qualify or win age groups, then chances are you need to train enough to achieve higher values of CTL. The training still needs to be appropriate and be more than just about the volume, but the overall level of work done needs to go up. It’s not impossible to perform at that level on limited hours, but it takes a fair bit of history and/or talent to do so. Equally though it’s quite possible to reach a high level of CTL and not perform at the highest level.

    I’ll always remind athletes it’s the performance that counts not the numbers in their charts. Those numbers might give an indication of what to expect and where they are, but it’s what happens on race days that matters.