How do you apply endurance sports science in practice? Learn the complete process from knowing the balance between evidence-based and experience-based, interpreting and contextualising research, recognising poor research and interpretations, to applying it to improve your triathlon performance. 

• The three-legged stool: scientific evidence, coach's experience, athlete's desire and experience. 
• Don't proclaim yourself "evidence-based" and disregard experience (yours or others') for the sake of it. 
• Interpreting research studies, understanding different study designs and terminology, like statistical significance and effect size. 
• Statistical significance does not equal the absolute truth - research results is the observation of what happened, not the prediction of what will happen in the future. 
• The importance of context (demographic, environmental, athletic ability, etc.).
• The importance of humility - don't be so quick to criticise. A great training plan is based on improving the individual athlete's performance, not on pegging them into the hole of the average research subject response. 
• If you're not measuring, you're not managing. 

3:59 - 

• Training shouldn't be a case of you finding the one solution just by reading a lot of research and optimising your training plan based on that.
• I like the analogy of the three-legged stool when it comes to planning training and sports performance:
  • The three legs are scientific evidence, the coach's experience and the athlete's desire and experience.
    • Scientific evidence is what we'll talk about later in this episode.
    • The coach's experience is just as important - if they have coached a lot of athletes and can say this usually works for people like you, try it!
    • The same thing goes for the athlete's experience, but also what they want to do. For example, if they hate running on a treadmill and have the ability to run outside, it might be in their best interest to plan for running outside, even if the coach is a big proponent of treadmill running.
  • The more experienced the athlete the more important this becomes - they know what's worked and what hasn't worked for them in the past.
  • You don't want one leg of the stool to become too dominant because it'll tip over.
• Don't use "evidence-based" as a shield.
• It's important to consider what's happening in the real world, and use your performances in training and racing to adjust your training, rather than just basing your training on research.
• Test yourself and measure your performance regularly.
  • Also, use other performance benchmark sessions - there are many options.
  • E.g. if you use Training Peaks you can monitor your peak powers for various durations regularly.
• Real life example: In my coaching I have a spreadsheet which has one tab for each athlete that I coach, and I have a list of all sorts of tests, races and key workouts.
  • E.g. I have one row just for FTP-tests, and whenever the athlete does an FTP-test I'll add the results in the next column (date, power, average and maximum heart rate).
  • I can then easily overview how things have progressed.
  • I can also correlate what happened when we saw progression, or regression, and work out why that happened.
• Consider the three-legged stool when you're assessing your training plan to make sure any one leg isn't overpowering the others.
• There is no one right way to train. There are too many variables and confounding factors.
  • We need to combine the scientific evidence with the art of coaching and the athlete's experience to get the best results.
• We're not after being evidence based, we're after performing well!
• You should be open and honest about what the reason for any training/nutrition/recovery intervention is - is it based on research or your experience?
  • Both are fine, but you don't want to claim it's evidence based if it's just your experience. It's important to be clear and honest about that.

11:35 -

• I'll try and keep this simple, but if you have any questions comment below and I'll be quick to answer any questions that come up.
• You need to consider study design, statistical significance and effect size when interpreting studies.
• Study design:
  • There are different grades of how well designed a study is and how much credit you can give based on it's design.
  • Case studies or expert opinions are the weakest grade of scientific evidence.
    • They are still valuable when added to other studies out there or your own experience.
  • Non-controlled studies are in the middle - they will have one experimental group.
    • E.g. 20 runners, testing to see if increasing from 4 to 6 runs per week will improve their 5K time. All runners will do a baseline test before and after the study, and see if the results are statistically significantly different.
    • The problem is the lack of control group (e.g. one group remaining at 4 runs per week) so you can compare the difference.
  • A study with a control group is the best type of study design, particularly randomised controlled studies.
    • E.g. using the example above if there was a control group you may see they improve just as much as the group that increased their training volume - maybe they were all new runners so they'd improve just as much running 4 times per week as 6 times per week.
• Statistical significance:
  • There will almost always be a difference between study groups
    • E.g. using the above example, say one group improved running time by 1 minute, and one group by 59 seconds - there is a one-second difference, but what does that mean?
    • If it's a one second difference you can't say that one method is better than the other - this difference is simply don to the variations in the sample.
    • This is why you use statistical methods to see whether the change is likely the result of chance or a true effect.
  • Statistical significance basically tells you that if you repeat the test 100 times, 95 of the times you'd find the effect, 5 times you wouldn't.
    • You usually draw the line at 95%, any smaller and it would be too difficult to find an effect.
  • Statistical significance is a composite score that's based on the true difference.
    • E.g. using above example, if the true difference was an improvement of 20 seconds after adding 2 runs per week, that's just one part of statistical significance.
    • The others are the variance - some people may improve by 10 seconds, some not at all, some by 40 seconds etc.
    • The bigger the variability the more difficult it is to get statistical significance.
  • The sample size is also important.
    • If you repeat the same study and get the same difference but one study has more participants, this study may find statistical significance and the smaller study may not.
    • It doesn't mean that there isn't an effect, but it tells you the likelihood based on the variability of the phenomenon and the sample size of the study.
• Effect size:
  • In this example, effect size is the difference in 5K running times.
    • If one group improved by 30 seconds and one by 5 seconds, you can say the 25 seconds is the difference here.
    • If you found a 25 second difference on a 5K and you also find statistical significance, that's a good thing.
    • If you find in another study an effect size of 10 seconds that is statistically significant, that effect size is pretty small.
  • So there might be a true difference but you need to weigh up whether it's worth adding the extra runs, and potentially extra injury risk, for such a small gain.
• Keep in mind that not all research is good research. The interpretation that the authors make isn't always good, they can also make mistakes.
• These days there are lots of journals that make it easier for you to get your research published.
• If somebody has a biased opinion they may be able to find one study that backs up their view and this can quickly become publicised on social media.
• Research shows what happened and if it was due to chance or not, it does not predict the future.
  • Just because there was a difference in running performance, it doesn't mean the same thing will happen in the future if someone else does it.
  • It might be true on average but not for every person.
  • It is also very contextual - a lot of variables go into this.
    • E.g. demographics, background of participants, current fitness level, what other exercise they did during the study, nutrition, sleep, etc.
• A positive research result doesn't mean 'this is the way you should train'. It needs to be put in context by reading the full study.
  • If the context seems applicable to you it means it might be a method worth trying, but it still may not work for you.
• If there is a study and the context fits you perfectly, and if there's one or two studies with a decent effect size, it may be worth trying.
  • However, if it's one or two studies with a good effect size but the context doesn't fit you well, you may want more evidence before trying something new.
• It comes down to your philosophy. If there's no potential harm and some potential benefit some people might consider it a no-brainer to try it.
• Some other people might have the philosophy that there needs to be no potential harm and a high likelihood of success, which would require many repeated studies showing the same effect, in the demographic of the athlete and in the context of their life.
• Don't just read the abstract! Read the entire study.
  • Almost any published research study finds some statistically significant results.
    • Sometimes research isn't published if they don't find results.
    • And many studies test multiple different hypotheses so some hypotheses will show results even when others don't. It may be due to chance even when they have statistical significance.
      • E.g. you have a statistically significant reduction in lactate levels, but so what if your time didn't improve!

24:30 - 

• I mentioned it with the previous example study where participants were changing their run schedule, and whether or not they were ordered to stay at home and rest, or allowed to complete other exercise during the study.
• This would affect internal validity.
• Internal validity is how well a study is controlled internally. How well the researchers controlled for different variables like recovery, nutrition, sleep, time of day, equipment, etc.
• Hypothetically if you control for absolutely everything in a study, to achieve the same results in the real world you may also need to put in the same level of control in your life. This reduces the studies external validity.
• E.g. always having the same nutrition.
• Internal and external validity are in a constant battle with each other.
• Ideally you want a good amount of internal validity, but you want it repeated across multiple studies that control for slightly different things.
• This would mean you've covered the spectrum of variables, but you still get positive results each time which gives external validity to the phenomenon over time.
• Nothing is a fact. Even if a lot of studies have shown it, error bars in research papers exist for a reason, there is so much variability.
• Also it's important to remember research is often behind coaches and practitioners.
• For example, don't discredit practitioners just because they say you should do a certain strength training exercise which hasn't got a heavy evidence base for improved strength and performance.
• It doesn't mean the coaches are wrong, it might just mean it hasn't been tested yet in studies.

27:07 - 

• We have to understand that in most cases we don't know anything.
• My goal as a coach, or as a self-coach, is to improve that specific athlete's performance, not to treat them as an average research subject.
• We need to individualise!
• This is intruitive, but it also means we may sometimes need to go against scientific evidence.
• Be humble about it, especially when looking at what other athletes around you are doing.
• E.g. if your friend is doing something, and you don't think it's evidence based, there may be a good reason why they're doing that.
• Their coach may be assessing what changes in performance occur after certain types of training, and hopefully adapting the training to have that individual response.
• This is not an excuse to not stay up to date! It's still important to know what the evidence is for all sorts of topics related to triathlon.
• Especially if you're a coach, you need to be constantly reading and educating yourself.
• But individualisation needs to come first - as we mentioned in the three legged stool at the beginning.
• Practicing humility can be difficult on social media. You often see people getting into arguments, pointing out flaws and errors in others training.
• Important to consider that each plan is likely to be suitable specifically for that individual.
• Accept how little we know, and be humble.

33:46 - 

• Know what your goal is, and what your objectives are.
• How you use science and experience will be different if you want to qualify for Kona, versus complete your first sprint triathlon.
• You need to know your objectives to know what you need to achieve that goal.
• E.g. You might need to raise your FTP a specific amount to qualify.
• "You're not managing that which you aren't measuring"
• My example with my coaching spreadsheet shows me trying to measure so that I can manage how my athletes train and hopefully help them improve.
• You should do the same! Particularly if you're self-coached. If you're coached, they will hopefully be doing this for you.
• E.g. If you start polarised training: do a baseline test before you make any changes, give it 8 weeks, and then retest and see if it's made a difference.
• If you don't progress it doesn't mean you need to abandon the idea, but you need to have a good reason to continue.
• When you want to include something that you've found in research, you need to look at the accumulated scientific evidence available as well as the accumulated experience available.
• Ask your coach, or if you're self-coached ask as many athletes as you can, and read anecdotal evidence on the internet.
• Form a hypothesis for how this will help you and how much it will help you.
• Will it help you make a 0.5% or a 5% performance improvement? If it's only 0.5% is it worth your time? However if it's 5%, it probably is - that's a lot in endurance sports.
• Consider what is the best use of your time and energy.
• As we talked earlier, you need to go deeper and assess the context, the demographics, the protocol etc.
• E.g. if it's heat acclimation they usually have very specific protocols, can you replicate this at home?
• Be very critical of the source of evidence.
• The best thing is to read the whole research study directly.
• If you're reading it in a magazine or a newspaper, it doesn't mean it's wrong, but you need to have a filter on.
• It's easy to misinterpret research.
• There are some great authors such as Matt Fitzgerald who writes for Triathlete magazine, and he knows his stuff.
• If I read one of his articles I tend to trust it, but if it's an author who I don't know I don't trust them as easily.
• If the authors can make mistakes interpreting research, anybody can, including journalists writing about it in more mainstream media.
• Be aware of your bias.
• It's easy to find one study which supports your view, but try to play the devils advocate to kill your own idea which will give you a more objective basis.

• If you're coached, talk to your coach and ask them questions.
• If you're self-coached, talk to other coaches and other athletes to get ideas and feedback.
• The better questions you ask the better feedback you get. Ask "what do you think is good about this idea, and what is bad about it" or "what are the potential benefits and disadvantages"

Links and resources mentioned

• The Enemy of a Good Plan is the Dream of a Perfect Plan with Dean Golich | EP#104
• "Should your hydration strategy change as you get older?" Article by Andy from Precision Hydration

Connect with host Mikael Eriksson

Hi! I'm your host Mikael,

I am a full-time triathlon coach and an ambitious age-group triathlete. My goal is podium at the Finnish national championships within the next few years.

I first started the website Scientific Triathlon in autumn 2015 as a passion project to share my learnings with a larger triathlon audience. Later on, in early 2017 I started the podcast That Triathlon Show. 

I sincerely want you to contact me to

• Send me feedback
• Give constructive critic​ism 
• Request topics and guests for the podcast
• Send me your triathlon-related questions 
• Tell me that you've rated and reviewed That Triathlon Show so I can give you a shout-out on the show and tell you how much it means to me!