Testing: Is the 20-minute test that necessary?
I already wanted to just do a little post on something I've been thinking about for a while, and then after I came across a social media post by another athlete I felt I really had to put the pen to paper and write down my thoughts. For cycling in particular you often hear about performance testing throughout the season, using the 20 minute FTP test, or 8 minute test, or doing the Andrew Coogan's Power Profile test (5s-5min power). Recently, I also came across another performance test done by the UCI to help identify potential pro-level cyclists who don't live in countries that have the funding to have national/provincial cycling programs. Essentially, their test is very similar to the Coogan power profile, it's just two 6sec efforts (in case you mess up the first one), 1 30 second test, followed by a 4 minute "endurance" test (I'll come back to this in a little bit). You also hear about doing test sets in swimming, and to some degree in running - although when I was distance running only I never actually called it a "test set" that seems to be a term that the triathlon community has coined. Basically the point of doing these mini tests is to track your progress over time or in the UCI test's case to identify future talent. I'm going to focus this post more on the tracking improvement piece of testing. So if an athlete does a 20 minute FTP test in January and achieves 5.5W/kg, and in April achieves 5.6 then in theory they have improved. You'll notice by the way that I referred to the FTP result in W/kg. Reason being if the athlete weighed 70kg and put out 385W for the 20 minutes (5.5), and then in April weighed 69kg and scored the same 385W (5.6) then despite the wattage being the same they are putting out more power relative to their weight and therefore have improved. Conversely, to go to other way, if the Athlete did the 385W at 70kg in mid-season form, then in the offseason started hammering the Christmas cookies, weighed in at 80kg and put out 400W for the 20 minutes, while they may want to think that Christmas Cookies are definitely the fuel of champions, really they put out 5W/kg which is significantly less than there mid-season form (and that extra 10kg may be problematic on the run who knows).
So the point of the test is number one to track progress over time. In order for something to be valid and truly track changes over time, variables that could affect the test results need to be accounted for. In other words, when you test in January and later in April: where you tested, on what bike, IN WHAT POSITION (I'll come back to this too), conditions in the room (heat in the house, the fan speed, fan placement), music or no music, what you ate/drank (caffeine or no caffeine) leading into the test, your training the days leading into the test, the list goes on and on, but basically everything from one test to the next needs to be identical. If you go into the January test fresh as a daisy and scored that 5.5, but then in April you feel dead tired from a massive training block and score 5, did you really get worse, or were you just too tired. If any variable is different from one test to the next, the test is almost meaningless and a total waste of 20 minutes spent in a world of hurt. The athlete I mentioned earlier on social media had posted something saying they had bombed there FTP test because they were dead tired going into it after a long day of work, but it's okay cause they would test again in a month's time. If your going into a test, a test where the purpose is to completely empty the tank, with already an empty tank, what is the point of testing. If you go and do the test next month say maybe on a weekend where you haven't worked during the day, you're feeling fresh, and you get 50 watts higher, did you improve? Maybe, but because the conditions were different from one test to the test, the result is meaningless, and you can't really draw any conclusions to whether your training from the last 4 weeks was good or not. If the first test was done dead tired, then guess what, for the next test to mean anything you also have to be dead tired. In other words, the first test was a waste of time when maybe a recovery spin, and then waiting another until another day to test when you feel fresh and ready to really get after it.
Then there is the training you do going into the test, that needs to be similar for the test to really by a reliable source of insight into progress over time. While the 20 minute test is going to hurt no matter what, you can absolutely train yourself for the 20 minutes to make the results a little better on the day. So for example, in the weeks leading into a test you can do say 4x5minutes at 108% FTP on 1:30 rest, or 3x7 minutes at 105% on short rest, or 2x10 minutes, and more, and then smoke your FTP test because you've trained well for it. Essentially instead of training for the 70.3 distance or full ironman that you're training for and using the test to track your progress towards those goals, you've trained for a 20 minute TT. But my question is, if in January you do that type of training leading into the test and you kill it, then in April as the first race of the season nears and your in a build phase, you begin doing more 70.3 race specific efforts of say 90 minutes at pace, or 2x1 hour at pace, you do the test and score the same or maybe worse, does that mean you are not in as good shape? I don't think so, I just think the training emphasis has shifted a little. So I think in order to make a test to carry a good amount insight throughout the season maybe the answer is to periodize your testing just like you periodize your training. So maybe when your in a phase of training where your doing lots of threshold/VO2 max type training using an 8-minute or 20 minute to track improvement over that training period. When your training shifts to more sweet spot/tempo efforts in a build phase, change the test to maybe a 1 hour TT, or maybe the max average power across say 3x45 minute intervals (2:15 of total effort so roughly 70.3 pace), or whatever mix does a good job of reflecting and testing the training emphasis.
Another point on things is that a good test is not only one that is valid, it is specific to your goals/event. To my point earlier about positioning, if you do your FTP test in an upright climbing position on the bike, unless your planning on racing in that position, there is no point to the test. Because in addition to tracking improvement, an FTP test, is used to set your training zones. So say you score 400 watts (just to make math easy) on your 20 minute test, and that would mean for a 70.3 where you plan to hold 85% FTP (or something near that) you should aim for around 340W. However, most people can hold a higher power in the upright climbing position than in their aero position (just due to larger glute activity in upright position meaning more power). So if you held 400W in an upright position, and then on race day you want to hold 340W, well maybe in an aero position 340W is equivalent to 90-95% FTP if it was tested in that aero position, in which case good luck running off that. My point is, if you're using FTP tests to set your training/racing zones the position you test in needs to be the same as the position you race in primarily. You could always do two tests one upright, one aero if you want so that you have a rough idea of your training power zones in each position and that may come in handy if the course has some long steady climbs in it (such as Ironman 70.3 Worlds this year where there was a long roughly 15 minute climb).
What about the mindset going into the test? For a test to really be accurate the effort level needs to be the same from test to test. The best way the effort can consistently be the same from one test to the next is if you going absolutely as hard as you can from start to finish (basically all out is all out and less subjective than going comfortably hard for example). Say your doing a swim test set where your doing 10x200 on say a 2:50 pace time. If going into the test you say to yourself something like my aim is to average 2:32 for each interval and you pace it perfectly and get 2:32 each and every time. A month down the road same test, 10x200 on 2:50, this time you say I'm going to aim for 2:30, again you amazingly got exactly 2:30 average. My question is do you know for sure that you improved from month 1's test, to month 2's test? Well, you don't know for sure, what if in month 1 you went a little harder threw out the pacing strategy and were able to average 2:28s. Having the 2:32 target actually was maybe below what you were capable of doing on that day. If your using the test to really assess did I get better or not, then I would discard any pacing strategy and you need to go all-out start to finish, and whatever the time is, is the time. Sure you can have a rough idea of where you should be so that for a 20 minute test on the bike for example your legs don't fill completely with lactate 5 minutes in and you need to stop, but what I'm saying is don't put a very specific number on it. It's why if I was doing a run test, I wouldn't do it on a treadmill, because you are preselecting the speed and who knows if you could have gone a little faster than the speed you selected. If I was doing a bike FTP test, I may set a very large range to score between but I wouldn't say something like I'm aiming for 336W, maybe saying something like anywhere from 330-350.
With that point about pacing for tests, it brings me to one last point, and that for most tests like the 20-minute FTP test there is absolutely a learning effect. You learn how hard you are able to push yourself for 20 minutes at a time, and therefore you get better and better at spreading your effort nice and evenly across the 20 minutes (since for a accurate and meaningful FTP test your power in the first minute and last minute should not be drastically different - there should not be enough energy for a sprint to the finish). Now the learning effect would start to diminish more and more, the more frequently you test. But how frequent should you be testing. Remember conditions from one test to the next should be similar, so if you want to test every month to slowly eliminate the learning effects, then most of training is pretty much timed around your tests. So for an FTP test for example, you may want a couple easy days on the bike going into it, and probably a couple easy days at least one after it. Now you've dedicated 4-5 days of every month to for just your bike FTP test, because you probably don't want to do a massive run or swim the day before either. That's fine and dandy to have four days of pretty easy training (plus one test) at the start of each month to do your test but what I think could be easier and is really my whole point to this blog post is instead of having to do a set 20 minute test on multiple occasions throughout the year you can use the numerous 20 minute, 30 minute, 60 minute powers, etc. from the massive amount of data you've accumulated in the field, so to speak, from weeks and months of training to determine training zones and whether or not you're improving or not. Now I'm not saying for everyone to do away with testing altogether. In fact, really what I'm saying is that in reality each and every day your out training you are testing yourself in some way. If you aren't challenging yourself to improve on a daily basis then what's the point of staring at the black line at the bottom of the pool for a mind-numbing amount of time. I'm just giving my opinion on the matter, and that is that I believe there are too many cofounding variables for these set tests in particular the 20 minute FTP test to really be a good representation of improvement over time, and I rather just use my own feel and knowledge from the massive amount of data that I accumulate in every training session to know how hard is too hard, what is a sustainable 2 hour power for a 70.3, or how hard can I dig on a 15 minute climb based on literally thousands of hours staring at my bike computer or running watch during training and racing.
Where this idea really starting taking shape was in October training for Ironman 70.3 Miami. I was planning on doing a FTP test about 1 month out from the race. Going into the test, over a 7 day span I did 4 massive VO2 max/suprathreshold workouts on the bike, ranging from 4x5minutes, 2x10, 3x7, and 24 minutes with 16 minutes of VO2 max work in it (a 2:1 work to rest ratio). After the 7 days I felt ready to do a the test, but with the race now about 3 weeks out, I thought, why do I need to waste precious time to test if I'm better or not. Looking at the numbers I did in those workouts, it was the most power I ever pushed in all four of them by far, and I was 100% certain that my FTP was higher than when I last tested, and had a really good idea of what I could hold for the 20 minutes, or the 1 hour power that the test represents. So I went into TrainerRoad, TrainingPeaks, and Zwift and just manually changed my FTP to what I predicted based on those 4 workouts the number would be. Did I fall apart in workouts after that, No. Did they feel maybe too easy, No. Instead they felt just right, and when I moved into more race specific workouts the power felt just as it should be.
So in summary all I'm saying is to actually use that Garmin on your bike or wrist during run/swim training which is collecting millions of data points every second. It is literally giving you the answers to what you are capable of every single day in training, instead of just relying on a 20 minute test to let you know how things are going. Think if your doing 5, 10, or 20 hours on the bike every week, think how many 20 minute segments you've accumulated which can help you determine your 20 minute test potential or your goal 70.3 race pace. Look at softwares like Xert Online. Xert takes every session you have and gives you (in my experience) a very accurate estimate of your current FTP without ever actually doing a FTP test.
My final point and this goes back to the UCI's test, is if you are doing some testing know what you're testing to determine how specific that test is to your goals. In that test's case the "endurance" test is a 4 minute long test. Now if you've studied exercise physiology you will know that a 4 minute long test will definitely be influenced by your anaerobic capacity. In fact, I would bet out of the 4 minutes, you are probably doing the first 60-90s with a heavy anaerobic contribution and only really the final 2 minutes or so will be testing your aerobic system's capacity.
An Aside: So you know I am well aware that nothing is ever purely one system or another, I took an entire 4th year level course on this false assumption (got A+ in it - course was the Physiological Mechanisms of Fatigue taught by Dr. Belcastro it was excellent course and highly recommend it). In grade school and high school we learn that it is this very step like process of going from Creatine Phosphate for few seconds to glucose in glycolysis, to pyruvate into the Kreb's cycles, then in the ETC etc., but actually these systems are all going at the same time, in fact by-products of each are fueling one another. Don't worry I'm not going to get into this, but basically know that in reality the old thing of well it takes the aerobic system time to "warmup" or something is not true, its not you get 10 seconds or whatever of just ATP-PCr, then minute of lactate, than aerobic takes over. They are cooking all at the time time.
But anyways for the test, they use the test to just purely test a person's VO2 max since you are working at such a high work rate for the primary aerobic portion of the test. It is not really something they are testing over time to track changes, because in reality the VO2 ceiling won't change all that much even with lots of training. Instead, they use this test to scout talent. They want to know what is this person's absolute VO2 max, we can train there 1 hour power, or lactate threshold, but the ceiling of their potential is pretty set. So going back to my point of knowing what the purpose of the test is, will you do repeated 4 minute tests throughout the season to determine whether an athlete is improving or not, probably not because the 4 minute test is testing one's max VO2 power which won't change massively. While the 4 minute test may be good to see VO2 max, it is not all that specific to the 70.3 you are training for. Nice to see and get a number, but knowing how much power you can hold at max effort for 4 minutes won't come into play during a 2 hour bike leg (that's why the 20 minute test is more useful it is a little more specific to the event without being long enough to wipe you out for a week afterwards). This goes to swim and run testing too. Know what you are testing, and in doing so make it specific to your event. Will a 200m swim TT, be a great assessment to determine how you will do in a 2km swim -No (*but it can be useful in pro field terms of assessing take out speed). Will a max effort 400m run be a great indication of how well you can run a half marathon - No. Think about what system you are using during the event, and that is the system you want to test. For a half-marathon runner for example, you are working very very close to your lactate threshold, so your lactate threshold is what you want to be testing throughout the year, not necessarily your absolute VO2 max which is pretty set (yes it can increase a little with training but I'm talking very little like a couple % whereas speed at lactate threshold is certainly trainable).
So that's it for my post/rant on testing. Basically I'm of the school now that I rather use the massive amounts of data accumulated throughout the season to decide on a sustainable race pace or how much I can push for a certain stretch of time, instead of relying on tests (specifically the 20-min FTP test) that are just influenced by too many variables to really make them all that valid and reliable. However, if you are testing, make it as specific to your event as possible between the physiological demands of the event, and then also things like the position you plan to ride in on the bike. Also keep variables as close as possible from one test to the other, and making sure to really just go as hard as you can start to finish instead of setting a specific number to aim for in order to make the effort from one test to the next consistent.
My next post will be a recap on my season, and some of my plans for next year. I'll also give a sneak peek into the setup of the training room I spent more than a month putting together and I really enjoy so much. It's the reason why I'm supplementing Calcium+Vit D daily now since the grand majority of my day is spent training underground (in the basement).
Until then, happy training.