Friday, March 30, 2012

And the results are in!

Dear Swimmers

Thanks for turning up on Wednesday at 9.30am or this morning at 5.30am / 6.30am to retest over the 400m and 200m Time Trials for the culmination of our 10 Week CSS Development Program. Here's my breakdown of our little experiment for your reference (apologies for the length!).

Abstract (for a more detailed description of the 'experiment' and our findings, please read to the end):

127 athletes from the Swim Smooth Perth Squad were tested on 13th January 2012 for their Critical Swim Speed (CSS) pace per 100m using the 400m and 200m Time Trial method first proposed in 1993 by Ginn*. A 10 week CSS Development Program was designed and put in place with the view that it might be possible to improve the swimmer's performance by reducing the CSS time per 25m by 0.5% per week using the Finis Tempo Trainer PRO to help control pacing accuracy for the incremental improvements proposed. The swimmers typically swam 2 to 3 times per week for 2.5km to 4km in a given session. One session per week was set aside to purely work on developing this aspect of their performance by adapting to progressively faster CSS times. By 30th March 2012, 57 of the original 127 athletes had been re-tested and 25 new athletes had also been tested for their current CSS pace.

*Ginn, E. (1993), "The application of the critical power test to swimming and swim training programmes", National Sports Research Centre

Further athletes will be re-tested in the first week of April 2012 (this is a preliminary report, due to be updated). 

The hypothetical goal of an average 4.5% reduction in CSS time per 25m for the squad (0.5% per week compounding over 10 weeks) was not achieved, but an average reduction of 2.59% was observed.

45 out of the 57 athletes re-tested improved their CSS pace with the training intervention, 42 out of the 57 improved their 400m times and 40 out of the 57 improved their 200m times. The greatest improvement was seen by Alen Pezzin dropping his CSS time by an amazing 13.62%! Well done Alen - sensational result!

Other significant improvements above the goal of 4.5% were made by:
  • Alen Pezzin, 13.62%
  • Bill Carmody, 9.92%
  • Lorraine Driscoll, 8.82%
  • Michael Serich, 8.25%
  • Chris Foley, 7.80%
  • Ian Murray, 7.46%
  • Caroline Claydon, 7.05%
  • Trevor Magee, 7.02% (a sensational bounce-back from a horrific shoulder injury from cycling!)
  • Craig Jameson, 6.27%
  • Chris Jameson, 6.11%
  • Nicole Klemm, 5.88%
  • Jess Huston, 5.76%
  • Bill Moody, 5.29%
  • Belinda Bennett, 5.12%
  • Lindsey Shepherd, 4.98%
  • Anne-Marie D'Arcy, 4.63%
  • Amy Callow, 4.58%
Swimmers needing to move up a lane or two (!) are:


Chris Foley - lane 2 to 3
Bill Moody - lane 2 to 3


Jess Huston - lane 4 to 4.5
John Turner - lane 4 to 4.5
Michael Serich - lane 3 to 4.5
Alen Pezzin - lane 3 to 4.5
Amy Callow - lane 1.5 to 2
Caroline Claydon - lane 1.5 to 2


Gillian Evans - lane 1 to 2


If you recall, CSS stands for 'Critical Swim Speed' and closely equates to your continuous 1500m pace in race-like conditions. It's an excellent measure of your current level of aerobic fitness and can be estimated by doing a 400m Time Trial and a 200m Time Trial and comparing the two. Put very simply, the 400m swim assesses the aerobic / endurance component of your current fitness and the 200m swim assesses the anaerobic / sprint component of your current fitness. This is a bit of a gross simplification of what the equation looks for, but essentially allows you to extrapolate out what pace a swimmer should then be able to maintain for longer distances based on the slope of this curve. It has been used in sports science since the early 1990s as a non-invasive way of testing and monitoring for improvements over time and is much easier to get you guys to do than a 1500m Time Trial every 4-6 weeks!

You can read more about the validity of CSS testing at: and in more simple language at:

As you know, we've been using the new Finis Tempo Trainer PRO in our Fresh 'n' Fruity CSS sessions for the last five months to accurately gauge pacing over a set distance. Before the Finis unit we were using the Wetronome and have been using this in a similar fashion for the last 4 years in the squad. The beepers are set to beep once every time you should be at each 25m mark, e.g. if you wanted to swim 1'40" per 100m, it should be set to beep every 25 seconds (4 x 25 seconds = 100 seconds = 1'40"). Simple - you either stay with the pace or you don't - a little like the red World Record line at the Olympics overlaid to visually show how close to a new record a swimmer is - just like the red line has a habit of getting in front of the Olympians right in the last moment, so too does the beeper often annoyingly get ahead of you too!

With the new beepers it has been possible to more accurately set the pace to be precise to 1/100th of a second, so as opposed to just full second increments being available (giving steps of 4" per 100m), we've been able to set paces to 0.04" per 100m accuracy! Amazing! This then prompted the idea - how much improvement in someone's threshold pace (a key determinant of their endurance ability over any distance >400m) is possible over a given time frame like 10 weeks?

Hypothetically, coaches have often discussed the potential to improve by ~5% in a whole season (i.e. 9 to 12 months) with a consistent approach to their athlete's training programs. However, not until now has such a device been available in swimming to control the accuracy and performance of a swimmer in a given training set. Given that to make a 5% improvement in a whole season actually requires just incredibly small steps in improvement to be made each session, I proposed that it may be possible to see an average of 0.5% improvement per week over 10 weeks following a very structured program designed specifically to bolster improvement in CSS pace for masters swimmers / triathletes of varying ability typically swimming 2-3 times per week for 2.5 to 3.5km per session. This improvement - when compounded - equates to a goal of 4.5% reduction in CSS time per 25m over the 10 weeks. Given the variety of abilities within the group, this seemed like a reasonable, albeit optimistic, goal to aim for.


The plan was to use the beepers to set a progressively faster CSS times per 25m by 0.5% each week for a key session every Wednesday or Friday morning. This typically equated to just 0.1" to 0.25" per 25m after the swimmers were all tested in mid-January 2012. Such a small margin was deemed to be barely perceivable by the swimmers and as such the goal of 0.5% improvement per week was set upon. Each week the main set was split up into two parts - the 'pace control set' (a series of 5 to 10 x 100m intervals with a full 25m beep cycle recovery, i.e. ~18 to 30" rest) and the 'maintain over distance set' (a total of 1000m of longer intervals with shorter relative recoveries aiming to stay as close to the set pace as possible). Even up to Week 10, most swimmers found the first set doable (given the 4:1 work to rest ratio) and were encouraged to use this first set to really dial in the pace awareness at the new speed each week with the view of avoiding 'blowing-up' for the second part of the main set. The second part was always kept 'blind' to the swimmers so that they did not hold back in anticipation on the 100m intervals. Sticking to these times after week 5/6 in particular proved to be quite challenging, especially for the faster, more experienced swimmers.

127 athletes were initially tested for their CSS pace and allocated set training groups (lanes) according to their ability within the three Swim Smooth Perth squads at 5.30am, 6.30am and 9.30am. The fastest swimmer in each lane was to use the beeper to control the pace for the whole group and up to 5" difference in CSS pace per 100m was seen within each lane between the fastest swimmers in the lane to the slower swimmers. In theory, every swimmer should have followed the same protocol at their own set CSS pace using their own beeper, but given the drag effect created by in-line lane etiquette (even when the swimmers were 5 to 10 seconds apart) and with up to 10 swimmers per 50m lane, this was seen to be impractical.


The hypothetical goal of an average 4.5% reduction in CSS time per 25m for the squad (0.5% per week compounding over 10 weeks) was not achieved, but an average reduction of 2.59% was observed 

45 out of the 57 athletes re-tested improved their CSS pace with the training intervention, 42 out of the 57 improved their 400m times and 40 out of the 57 improved their 200m times. The greatest improvement was seen by Alen Pezzin dropping his CSS time by an amazing 13.62%!

NB. swimmers highlighted in purple did not complete the initial test, but were tested along with the rest of the group on 30/3/12 for the ongoing benefit.


CSS closely equates to your continuous 1500m pace in race-like conditions and the calculation (as seen at: ) attempts to extrapolate that from the 400m and 200m Time Trial speeds. If a swimmer has a fast 200m time but seems to be much slower relative to other swimmers in his / her lane over 400m an 'Aerobic to Anaerobic Ratio'* of over 4.0% is typically observed. This says he / she is more anaerobic in their physiology and is either naturally suited or has trained (or both) for short distance swims. Other swimmers have a much lower drop-off between his 200m and 400m times and therefore are much more likely to be good distance swimmers. As shown in the chart below, the calculation is predicting this trend will continue and that the pure distance swimmer will start to get progressively faster than the middle distance swimmer over distances of more than 400m.

*this is a super simplified way of describing the drop-off in times between the 200m and 400m swims.

...really this has to be the case or the fastest 100m sprinter in the world would also be the best at 400m and 1500m – when in fact these three distances need different athletes with different physiologies. The sprinter needs pure power and a strong anaerobic system. The middle-distance swimmer needs a compromise between anaerobic and aerobic and the distance swimmer needs a great aerobic system.
The key thing here is that if the middle distance swimmer wants to maximise his distance swimming he / she needs to do more CSS work and not focus on training his anaerobic system with lots of short, sharp sprints. 

Probably the best example of this within the squad is that of Bill Moody. In 2009 Bill trained up for the Australian National Masters Championships in the 50m freestyle sprint event. He's a natural born sprinter with a very good turn of speed. 10 weeks ago his CSS pace was 1'44" per 100m with an 'Aerobic to Anaerobic Ratio' of 6.39%. Bill had a tendency of setting off too quick in interval sessions and during training for the Rottnest Channel Swim Event (20km) performed a 10km training swim which started off at a pace of ~16' per km but by the end had dropped to over 24' per km. Had this happened during the actual event, Bill ran the risk of not even finishing. The CSS intervention training and a lot of work spent at ~8" per 100m slower than CSS pace resulted in Bill dropping his ratio to 3.68% and improving his CSS pace to 1'38" per 100m. This took a lot of trust on Bill's part in my seeming "madness". He didn't get any faster over 200m, but for distance freestyle he didn't need to - his performances above 400m though really improved. A sensational result that ensured that he completed the Rottnest Channel Swim in a very respectable 7h05m - Bill actually got faster by going slower and getting REALLY good at pacing! This is arguably one of my proudest coaching achievements to date given how weird these ideas must have sounded to Bill at the time.

Whilst this protocol primarily looked at trying to improve the swimmer's CSS pace, it was also beneficial to see improvements in the outright performances over 400m and 200m. Given that some 'strange' results can occur (as detailed below), it was good to see that nearly 74% of the squad got faster in both 400m and 200m times.

  • A swimmer's CSS pace can improve despite their 200m speed in particular decreasing due to the slope of the extrapolated line essentially become more gradual. This can show one of two things: 1) more focus has been on CSS and endurance type work (as per the 10 Week Program); 2) the swimmer was too tired from the 400m exertion and correspondingly did not perform well over the 200m. This becomes particularly evident if the 200m time is more than half of the 400m time and can lead to erroneous results as in theory it's impossible for a swimmer's speed to not be at least as fast over 200m as it is over 400m as at some point within the 400m test the swimmer must have swum 200m as fast as (or faster than) half the 400m time. This occurred with 8 out of the 57 swimmers (Jon Turner, Anne Murrell, Suzi Scarff, John Edwards, Sally Steffanoni, Kim Annear, Roxanne Garven and Annette van Hazel). As such, for the purposes of the analysis, their 200m times were doctored to 50% of their 400m time less 2".
  • It is possible for a swimmer to get slightly faster over 400m but a lot faster over 200m and for his / her CSS pace to appear to decrease or be exactly the same as it was at the start of the 'experiment'. This occurred with 3 out of the 57 swimmers (Michael Japp, Adam Wheeler and Sally Howe). It can be explained by either the swimmer not performing as well on the 400m swim (possibly due to pacing) or typically from having some time off, i.e. endurance drops off but freshness increases and over short distances the swimmer is able to knock out a pretty quick time. It can also occur if the swimmer's training has been skewed much more towards shorter, faster intervals well above CSS pace, i.e. sprinting.
Of course, ideally we'd have liked to see everyone improve (that was the aim after all!) but a 79% improvement rate in CSS across the entire group is a pretty good result all round. Some possible suggestions for those who didn't improve or only improved a touch are:

  1. How consistently were you swimming 2-3 times per week and attending the Fresh 'n' Fruity Test Set each week?
  2. Could you have just had an 'off day' (we all have them!) - if you felt like you've been progressively well recently, it would be worth re-testing next week and see if you can have another go at it.
  3. Have you been on holiday during this last 10 week period, or been sick / injured? Obviously the program was cut in half by the Rottnest Channel Swim - some of you would have taken a few weeks off after this - could that explain it?
  4. Are you already swimming pretty quick (i.e. lane 4 or 4.5) - could you already be nearing your ceiling of improvement? Was 0.5% per week simply too much to ask?


I do hope you've all found this 10 Week CSS Development Program to be at the very least intriguing! If you're still reading now, firstly, well done (!) and secondly I hope it's helped improve your knowledge and interest of how to improve your efficiency for distance swimming purely from a training / physiology perspective, i.e. irrespective of any specific stroke technique development work we did during this period. It's been a very useful exercise for me personally as I simply didn't know what people were capable of within a 10 week time frame, only that you were all likely to develop at different rates, despite all following a relatively similar program with the hope that you could all attend as consistently as possible. The findings (thus far) show that 2.59% improvement is possible within this time frame as an average, but equally that much more and much less than this amount is equally possible.

I plan to use these results to further develop the squad's fitness training structure to compliment the large amount of technique work that we also conduct.

For the next few weeks at least we'll ensure that we maintain and develop this fitness further, albeit with some new session ideas.

Thanks for reading and thanks for being part of this exciting project!



No comments:

Post a Comment

Please add your comments here: