Shownotes

Sebastian’s background

00:45 -

• I am a professor of sports science at the University of Zaragoza in northern Spain, where I also work as a researcher. Alongside my academic role, I run my own coaching business, working primarily with road cyclists but also with triathletes and triathlon runners, across both professional and amateur levels.
• My research centres on performance in road cycling, and more specifically on everything that relates to power-based training. I also authored the book Cycling 2.0.

Top three specific findings from sports science

02:08 -

• Three types of studies have influenced me considerably. The first are retrospective longitudinal studies that track the performance evolution of top athletes over time.
• My colleague Fred Grappe from France, for instance, documented the career progression of Thibaut Pinot from his junior years through to a Tour de France podium finish.
• Similarly, Gabriele Gallo from Italy has published work on the training characteristics of five Grand Tour podium finishers. These studies are valuable because they show what the best athletes are actually doing, what is worth replicating, and equally, what we should not be rushing to imitate.
• The second type are head-to-head comparisons between interventions — for example, comparing on-bike strength training against off-the-bike strength training. These are particularly useful for understanding what works over shorter time frames.
• The third type, which does not get nearly enough attention in sports science, is studies with neutral or negative results. These matter because they map out the dead ends — the approaches that cost time and money and produce nothing. Knowing where not to go is just as useful as knowing where to go.
• A good example of a null result shaping my practice is a study I published around two years ago on core strength training for cyclists. The finding was clear: there is no performance benefit. That kind of result directly informs how I coach, because — especially with amateur athletes who have limited time — there is no point directing effort toward interventions that the evidence currently suggests do not work.
• Looking at the longitudinal work — Gabriele Gallo's study on Grand Tour podium finishers and Fred Grappe's tracking of Pinot — what stands out is that the athletes followed quite different approaches to training and periodization, yet arrived at the same level of performance. That tells me there is no single correct method.
• What it does demand, though, is that whatever programme you put in front of an athlete is as well suited to that specific person as possible. Getting there requires both experience as a coach and enough time working with the athlete to genuinely understand them.

Differences between pro and amateur athletes

06:57 -

• One of the most common mistakes I see, particularly among highly motivated amateur athletes, is trying to replicate the training characteristics of professionals — the volume, the intensity, the total load.
• The problem is that they cannot replicate the rest of the pro's life. The single biggest difference between coaching a professional and an amateur is monitoring training load outside of the bike. With a pro, when training is done, it is largely done.
• With an amateur, that same person then spends eight hours at the office, works in a factory, comes home to look after their children. That stress is real, it accumulates, and it is very hard to quantify.
• The athletes I work with typically have ten to fourteen hours a week to train. Yet I have seen forum discussions where people are chasing a CTL of 155 or 160 — the kind of numbers you see from WorldTour riders during the Tour de France — while holding down a full-time job and family life.
• They might get there, but the cost is not just physical. It spills into their relationships, their work, their whole social context. That is where burnout comes from.

Science vs. empiricism: what is actually worth following 

11:34 -

• There is an important distinction between the empirical world of professional coaching, where experimentation is constant, and the scientific world, which moves more conservatively but with greater certainty.
• Some people say coaching is ten years ahead of science. I disagree. Experimentation is not the same as effectiveness. From my own experience working with professionals, it is much closer to throwing things at a wall and seeing what sticks — and perhaps twenty or thirty percent of what we try actually ends up being implemented.
• Ketones, torque training on the bike — these were trends. Look five years on and most of them have not held up.
• For an athlete who wants to do fewer things but do them well, the scientific framework is far more reliable than chasing what the pros are seen doing on social media.
• It may lag behind by a few years, but that is not a problem. What it offers in return is a much higher degree of confidence that what you are doing actually works.
• One of the clearest examples is the tri-phasic model of training — three zones defined by two thresholds. It has been around for over fifty years, the thresholds are straightforward to measure in a laboratory, and the evidence shows it produces results at least as good as more elaborate zone systems.
• There is something worth saying plainly here: we do not currently know whether sitting just below zone 2, just above it, or somewhere in zone 3 produces meaningfully different adaptations. The precision that detailed zone models imply is not backed by evidence that the differences matter.
• In terms of long-term adaptations, as far as the evidence goes, no, there is no demonstrated difference.

The funding problem in sports science 

15:06 -

• In medicine, pharmaceutical companies can fund multi-year trials with hundreds of subjects. In sports science, we fund our own research. I have paid for most of my own studies.
• My ideal would be to take three hundred subjects, train them across different intensity distributions for a year, and measure what happens at the muscular, cardiac and pulmonary level. That study would be genuinely illuminating.
• It is also not financially viable. Sports science is not considered a first-necessity field, so the money simply is not there. Whatever gets done in this field gets done out of genuine interest, not commercial investment.

Misreading what research actually says 

16:27 -

• More people are reading research than ever, which is broadly a good thing. The problem is that reading a study and understanding it are not the same thing, particularly when it comes to statistics.
• A direct, literal interpretation of findings can lead people badly astray.
• A good example is a study published last year from a Spanish lab comparing on-bike resistance training to off-bike resistance training. The title and abstract suggest the two approaches are equivalent.
• An amateur reads that, concludes he does not need the gym, and carries on doing torque intervals on the bike. But read further and you find the comparison was between five or six squats and five or six pedalling actions during a climb.
• Nobody goes to the gym and does five squats and goes home. That is simply what the methodology required for a controlled head-to-head comparison.
• The performance outcomes may have been similar in that narrow context, but that does not mean the adaptations are equivalent across the board.
• Off-bike strength training produces benefits for bone density that cycling alone does not — which matters considerably if cycling is the only load-bearing activity in your life.
• It also works the antagonist muscles, which has real implications for injury prevention. None of that shows up in a performance metric from a short study.

The carbohydrate intake example 

19:20 -

• Perhaps the most visible example in recent years is the trend toward consuming very high amounts of carbohydrates during exercise, driven largely by what professionals are seen doing.
• If you are a WorldTour rider producing 300 watts in zone 2 and burning around 1,300 calories per hour — the majority of which comes from carbohydrates — and you have specifically trained your gut to handle that intake, then yes, that level of fuelling makes sense.
• But the amateur attempting to replicate that is often riding at 180 watts and burning perhaps 700 calories per hour. It is an entirely different physiological situation, yet the behaviour is being copied wholesale from a completely different population.

Myths about endurance training

21:41 -

• The importance of body weight is significantly overestimated. Some context is necessary — if you are ten kilograms overweight due to excess body fat, yes, you will be slower, in cycling or running. That is straightforward. But the conversation needs to go further than that.
• In cycling, particularly around 2015 to 2017, the pursuit of leanness reached a damaging extreme. Riders were not just lean — they were combining four percent body fat with severe muscle wasting. That is a very different thing.
• Two or three kilograms of additional body weight means very little if that weight is muscle contributing to your power output.
• More absolute power matters even when climbing. A heavier rider with greater muscle mass and higher absolute power output actually needs a lower watts-per-kilo figure to travel at the same speed uphill as a lighter rider, because the added power also helps overcome aerodynamic resistance — which remains a meaningful force even on climbs.
• The physics works in their favour more than most people assume.
• The practical message is straightforward: stop obsessing over weight. If you are going to focus on something, focus on building the muscle mass that allows you to produce your best power. If that means weighing three kilograms more, that is not a problem — it is likely an advantage.

Periodization: no clear winner

24:39 -

• As far as the evidence shows, no single periodization model consistently produces better outcomes than another. Part of the reason is methodological — most of the evidence we have is retrospective, which makes controlled comparisons difficult.
• A paper I find particularly illustrative looked at two rowing gold medallists four years apart. Their training distributions were assessed and found to be completely different from one another, yet both reached the top of the podium. That is about as graphic an example as you can find.
• One was predominantly pyramidal, the other was largely high-intensity focused. But when you look across periodization models more broadly — ATR blocks, conventional periodization, reverse periodization as Team Sky attempted for several years — the long-term picture tends to look similar regardless.
• Most of the time is spent at easy intensities, with specific blocks of high intensity placed at appropriate moments in the season. The labels differ more than the reality does.
• The Gabriele Gallo paper on Tour and Giro podium finishers threw up some genuinely surprising details. Of the five riders assessed, three never did gym-based strength training at all, and the other two only did it in the pre-season.
• That alone challenges a lot of assumptions. Perhaps even more striking was the average weekly training volume across the year — it came in at under seventeen hours. For athletes performing at that level, that figure is lower than most people would expect.

Periodization in practice

27:48 -

• The starting point is always the athlete's objectives. If you have one, two or three major targets in a year, conventional periodization — large blocks of low-intensity, high-volume work building into each block — can work very well.
• It is also well suited to athletes who are newer to the sport or currently at a low fitness level, where the lower injury risk of high-volume, low-intensity work is a real advantage.
• The difficulty is that most athletes today, even amateurs, compete fifteen, twenty or more times a year. With that kind of calendar, conventional periodization becomes hard to implement cleanly — you constantly have races interrupting training blocks that were never designed for that competition stimulus.
• In those cases, and particularly with already fit athletes, I prefer block periodization, even though I am confident that success in endurance sport is achievable with either approach.
• Two factors guide the implementation. The first is how specific a given quality is to the demands of competition. The second is how long adaptations from that quality persist once training stops — what is sometimes called the detraining timeline.
• Gym-based strength training is a good illustration. The evidence for its benefits in cyclists is solid, but it is not specific to the pedalling motion.
• More importantly, strength adaptations are slow to fade — stop the gym entirely for a month and most of what you built is still there.
• That tells you something practical: you do not need to be doing gym work in the final weeks before a race, and you can stop it well in advance without losing the benefit. It belongs in the accumulation phase, not the realization phase.
• A typical block runs eight to ten weeks. The first month or so is accumulation — non-specific work, such as higher volume at lower intensities, which produces adaptations that also persist reasonably well over time.
• This is followed by a transformation phase with more demanding and more specific work. The realization phase at the end is focused almost entirely on competition-specific intensity, with generous recovery built in — essentially a structured taper leading into the race.

Training ultra-endurance athletes

33:42 -

• With ultra cyclists, ultra runners and Ironman athletes, the approach that has consistently given me the best results is keeping roughly 98% of the work below the first threshold.
• I have tried various strategies with this population and that one — the most straightforward, the most logical on its face — is the one that works best.
• The 2% of work above threshold that I do include is almost always VO2 max work. I avoid threshold training with these athletes almost entirely. But the wider point is that in ultra-endurance events, you can arrive at the same performance outcome through very different routes.
• Last week I was speaking with a Spanish marathon runner running 2h06. His VO2 max is 66 or 67 — a number that, on paper, should not produce that performance.
• What explains it is a gross efficiency of 26%, which is as rare as a VO2 max above 90. Making an athlete as efficient as possible can get them to the same place as having a more powerful engine.
• There is also a selection bias worth acknowledging. The athletes who end up doing twelve or twenty-four-hour cycling events are generally not the ones who excelled in shorter, more explosive races.
• They tend to have modest VO2 max figures and modest second thresholds — qualities that limit them in two or four-hour races with hard intervals, but which matter far less when the task is sustaining a steady pace for ten or twelve hours.
• In ultra-endurance events, there are factors that high-intensity training simply does not prepare you for well, and psychological resilience is chief among them.
• Riding at a relatively modest intensity for eight, ten or twelve hours demands a specific kind of mental durability. Accumulating high volumes of low-intensity training prepares you for that in a way that interval sessions do not.
• In triathlon particularly, I have seen many athletes who are fitter than they realise fail not because of their physiology, but because they were not prepared for what that duration actually demands mentally.

Threshold training

39:01 -

• Threshold intensity is something I use, but it is not a distribution I favour. The metabolic cost is high — athletes need substantial recovery time after threshold sessions, and from the feedback I consistently get, it leaves them too fatigued to execute even easy sessions properly in the days that follow.
• From a scientific standpoint, there is no clear evidence that spending more time at threshold produces better outcomes than other intensity distributions. The one consistent finding is that top performers spend most of their time at very easy intensities. Beyond that, personal preference and empirical observation guide the decision.
• My own preference, rather than leaning on threshold, is to use a small amount of it alongside more work in the severe and extreme intensity domains — the glycolytic efforts, zone 6 and zone 7.
• I am aware that colleagues would answer this question differently, and I do not claim to be the most correct position here.

Profiling athletes through critical power testing 

41:49 -

• Athlete phenotype is relevant to how threshold training lands — riders who are more glycolytic, with a higher proportion of fast-twitch fibres, tend to respond differently.
• This connects to how you test and interpret critical power and W'. The number you get for W' depends heavily on how you structure the test.
• Most people use two or three efforts to calculate critical power, but that approach does not capture the right-hand side of the power-duration curve.
• Incorporate that and the absolute kilojoule value above critical power can change substantially — enough to categorise a rider in an entirely different way depending on which method you chose.
• In practice, getting to know an athlete well often tells you as much as any test. How they perform across different session types, their personal bests at various durations on the power-duration curve — that profile emerges naturally over time and is at least as informative as a formal W' calculation.

Fibre type categorisation and its limits 

44:14 -

• The problem with categorising athletes by fibre type is that we do not fully understand either whether training can genuinely shift fibre type composition, or what actually happens at the level of muscle recruitment under fatigue.
• You can label someone as fast-twitch, but if you took a muscle biopsy after four hours of hard effort — which nobody does — the recruitment pattern might surprise you considerably.
• I acknowledge my own uncertainty here, but the evidence as it stands does not give me enough confidence to build training decisions around that kind of categorisation.

Power-duration curves and race-specific testing

46:05 -

• For an initial assessment, looking at the power-duration curve is useful. But once I am working with an athlete toward a specific goal, what I am really interested in is performance under conditions that resemble the race itself.
• A strong five-minute power in a fresh, rested state tells me relatively little when the objective is Liège–Bastogne–Liège. Improving that number is not unwelcome, but it is not what I am optimising for.
• My preference is always to test under racing conditions — fatigued, in context — because that is the only data that directly reflects what we are actually trying to achieve. So no, I would not use an initial power-duration profile to determine an athlete's training intensity distribution.

Opportunity cost: spending your time and money wisely

47:30 -

• I want to highlight one that is less about training methodology and more about how athletes make decisions: opportunity cost.
• Every endurance athlete faces constant choices about where to spend their time and money, and the key characteristic of those choices is that they are irreversible.
• If you spend this season focused heavily on one training approach, you cannot go back six months later and undo that. If you spend two thousand euros on a new set of wheels, that money is gone — it cannot be redirected.
• This matters far more for amateurs than for professionals. The professional has, broadly speaking, unlimited time and financial resource for sport-related decisions.
• The amateur typically has meaningful constraints on both. A practical example: if someone comes to me having just bought their first bike, my advice is not to spend their first two hundred euros on coaching. Spend it on a professional bike fit.
• Ride comfortably, enjoy it, accumulate kilometres, and in four to six months they will be considerably stronger than if they had paid for coaching sessions from week one. That is the logic of opportunity cost in practice.
• The chapter in the book works through this for each stage of an athlete's sporting career — what to prioritise at which moment to get the most from the time and money available.
• Strength training is a good example of opportunity cost in action, particularly for Ironman athletes with limited training time. The two hours spent in the gym are two hours not spent on the bike or running.
• The benefit of the strength work has to be weighed directly against that lost sport-specific time. There is no free lunch — every hour allocated to one thing is an hour taken from something else, and with constrained training time, that trade-off deserves careful thought.

Ideal PhD program

51:41 -

• Two important questions remain unanswered in the science. The first is whether slightly different training intensities produce meaningfully different adaptations over long time periods. The second follows directly from it.
• Getting a sample of thirty to forty subjects in sports science is already considered extraordinary. Asking three hundred people to train in a specific, possibly suboptimal way for an entire season is not realistic.
• But if I had unlimited subjects, time and money, I would test training at intensities just below and just above the first ventilatory threshold — zone 2 and zone 3 in power terms — and compare those against work around critical power and above, zone 5.
• Zone 1 I would exclude entirely; we already know enough about active recovery to set that aside. I would control for internal load using perceived effort, heart rate and power, relativised to each athlete's body weight rather than using absolute kilojoule figures, which become misleading when comparing athletes of very different sizes.
• My honest theory is that after eight or nine months, the results across those groups would be quite similar. There might be statistical significance somewhere, but the effect sizes would likely be modest.
• Even the difference between riding at 180 watts and 210 watts — which feels substantial over several hours — I still believe would not translate into dramatically different performance outcomes.
• There is a meaningful difference between subjective recovery and objective recovery.
• We have studies showing that athletes who receive a massage or do an active recovery session after training report feeling less fatigued the following day — but when they are actually tested, their power output and heart rate responses are identical to those who did neither. The feeling of recovery and the reality of recovery are not the same thing.
• Applying that to zone 1 riding: if an athlete wants to go out for an easy hour because it makes them feel better, that is their choice, just as paying for a massage is their choice.
• But they should be clear about what it is. From what the evidence shows, zone 1 training produces no meaningful adaptation and no objective recovery benefit. It costs time, and in the framework of opportunity cost, that matters.

Zone 1 in context: when it might still have a role

1:01:41 -

• The counterargument worth taking seriously is around double sessions. If an athlete has done a hard run in the morning and then goes out for a two-hour ride in the evening, the external load may look like easy endurance — zone 1 in power terms — but because of the fatigue already accumulated from the first session, the internal load associated with that external output will be considerably higher.
• In that context, there may be something happening at the level of central adaptations that justifies the session, even if the watts look unimpressive.
• That is a meaningful distinction from pure active recovery, where my position remains the same: stay at home.
• But a deliberately low-intensity second session in a double-day structure is a different thing. The body is not in the same state, and the internal stimulus is not what the power numbers alone would suggest.
• On the question of very long, low-intensity rides for ultra-distance preparation — I ran experiments with this and the results were instructive.
• After seven hours at 150 watts, there was no cardiac drift worth noting. If there is no internal response of that kind after seven hours, it raises a serious question about whether that is a productive use of time, regardless of any adaptation it might theoretically produce.

Active Recovery: Activity or Recovery, Not Both

1:06:05 -

• To be clear on my position: if you have an hour and the goal is recovery, do not ride in zone 1. Stay at home. Passive recovery is better than active recovery. There is either activity or recovery — not both at the same time.

The replication problem in sports science

1:07:05 -

• Replication difficulties in sports science are real, but they are not unique to our field, and they do not mean the science should be discarded.
• Working with human subjects makes replication inherently harder than in engineering, where mathematical calculations produce the same result anywhere in the world.
• In human research, even the same researcher running the same experiment on the same sample twice can get meaningfully different results.
• Add a different researcher, different tools and a different population, and the variability compounds quickly. This happens in pharmaceutical research too — results with a given drug vary depending on the sample population.
• What this means in practice is that the conservative, well-established findings accumulated over several decades are solid and worth trusting.
• The ABC of training science is reliable. The difficulty comes when you try to apply that general knowledge to a specific individual — and that is where coaching experience becomes indispensable.
• You can produce good results with an athlete in your first year together, but the best work begins once you know them well enough to modulate what the science recommends to fit their specific life, schedule and context. No two athletes are the same in those respects.
• Producing research yourself also sharpens how you read it. Writing papers forces you to engage with the full body of literature, and that process teaches you how to interpret findings carefully and understand their limitations — as with the strength training study I mentioned earlier.

How to read and apply sports science 

1:10:16 -

• In my book, around twenty-five topics end with the conclusion that we simply do not know. I lay out what the evidence shows, explain the possibilities, and am honest that the answer is not yet there. Periodization is one example. That honesty matters.
• For coaches trying to stay current with the literature, the quality of the source is the first filter. A study published in a first-quartile journal carries very different weight from one published in a pay-to-publish outlet.
• Beyond that, there is a hierarchy of evidence — at the top sits a meta-analysis of meta-analyses, a synthesis of average results across multiple studies over time.
• When that level of evidence exists and points in a direction, you can be reasonably confident that your practical results will land somewhere nearby, even if not exactly there.
• I am not a particular admirer of Tim Noakes, but I think it is a good paper with a well-founded perspective. The truth in this area is not fixed, and it is probably somewhere between the mainstream view and what Noakes is proposing.
• Contrarian theories that are grounded in solid evidence deserve to be published and criticised openly. If we only publish work that confirms the dominant view, we stop learning.

Two practical recommendations most people overlook

1:16:45 -

• The first is vitamin D. It sounds strange when you picture endurance athletes — young, fit, outdoors for hours at a time. Yet what I see consistently is deficiency.
• Many athletes are riding or running for six hours in the mountains and still coming up short, particularly those in northern latitudes who do not get reliable sun exposure.
• Get tested. If you are deficient, supplement. The evidence for its role in recovery is solid, and its effect on mood is relevant both to general wellbeing and to performance. In opportunity cost terms, it is cheap, simple to implement, and the potential gains — especially if you are currently deficient — are real.
• The second, and I think the more important of the two, is this: stop trying to adapt your life to your training. It should work the other way around.
• Look honestly at your circumstances — your job, your family, your available time — and build your objectives and training plan from there.
• Choosing a twenty-four hour ultra-distance event when you have a full-time job and a family is not just ambitious, it is a mismatch that will eventually break down.
• The athletes who get this right — who match their goals to their actual life rather than forcing their life around their goals — are the ones who train consistently over many seasons, stay satisfied, and ultimately perform best in the long run.

LINKS AND RESOURCES

• Sebastian’s website and Instagram 

• Cycling 2.0: Evidence-based training for peak performance on the bike - Sebastian’s newly published book

• Sebastian Sitko, PhD | EP#376 - Sebastian’s first appearance on That Triathlon Show