In my work as a cycling YouTuber, I focus on using an evidence-based approach to answer cycling and sports science-related questions. My viewers have become accustomed to seeing me cite many studies on a given topic and then expressing my opinion on where the balance of evidence lies at the end of the video. In this article, I break down my research process.
In the bodybuilding and weightlifting world, the term “broscience” is commonly thrown around and describes the kind of advice you might get in the locker room of a gym from a person who has done very little research, but hey…they have huge muscles, so obviously they must know what they’re talking about. Many cyclists and other endurance athletes are also familiar with the term, and it’s safe to say that this phenomenon is prevalent in endurance sports as well.
I’ve found many instances where common wisdom, tradition, what feels suitable, broscience, or whatever you want to call it simply doesn’t hold up when you look at the science. A quick example of this is stretching. Most athletes and coaches alike assume that stretching is in some way beneficial. But when you look at what the research has to say about stretching, you quickly find little evidence that stretching improves performance, speeds recovery, or even prevents injuries. This, of course, doesn’t mean that there are no circumstances under which a person should stretch. Though it does fly in the face of what we’ve all been told and stretching is far from the only example of this.
This is what I dedicate the majority of my YouTube channel content to—getting to the bottom of cycling training topics by looking at the science research. If you watch one of my videos, what you’ll likely find is me citing many studies, often with conflicting findings, and trying to parse out which ones have a better study design and hold more weight. Then, at the end, take a step back and try to determine where the balance of evidence lies. When I edit it down into a 10-to-15-minute video highlighting the most relevant parts of each study, it may seem as though I did some quick Googling and then regurgitate it on video, but it can be a rather long and tedious process. Let’s get into the approach I take to reading and understanding research.
Keep an Open Mind
Before I even start my research on a given topic, I try my best to go in with an open mind. Of course, being completely unbiased is impossible, but it is always the goal. Most topics will have many conflicting studies, some of which come to opposite conclusions. If you go into your research with an agenda, it is very easy for confirmation bias to take over and ignore studies that go against your ideas and focus on those that support them. You can see this play out when someone scrutinizes every inch of a study that goes against what they believe, but when a study confirms their belief, they quickly accept it without question. Again, we all do this, but the goal should not be to find research confirming your bias. It should be to find out which direction the research is pointing.
How I Find Studies
I start my search for relevant articles with PubMed and Google Scholar searches. I’ll copy any study or review relevant to the topic to a separate word document. If I find a review or meta-analysis on a topic, looking at the references for other related articles can also be helpful. During this step, I will usually skim the abstract to make sure that the article pertains to what I’m searching for. Going back to my first point about keeping an open mind, I make sure to copy over all relevant research onto the Word document regardless of the conclusion.
Reading Research
Research articles will start with an abstract which is essentially a summary of the article. As I said, by reading the abstract, I can see whether the article is relevant to what I’m looking for. As I continue to read through the article, I make sure to pay attention to the methods section, which tells you how the research was conducted. This is important because it isn’t uncommon to come across questionable study methods that likely skewed the results.
For example, some research on post-workout fueling concludes that carbs plus protein are better for recovery than carbs alone. When you take a quick look at the methods, though, you find that they had a carb group and then a group that consumed the same amount of carbs plus additional protein, meaning that group consumed significantly more calories post-workout. It’s not hard to see how this could be a confounding factor and would affect the results. Similar research has been done where the study had a carb-only group and carb-plus protein group, but the calories consumed were equal between the two groups. When this method is used, consuming carbs plus protein no longer provides an advantage over carbs alone.
Other things to look for are whether the study had a control group or used a placebo or had a large enough group of subjects or that there weren’t other confounding factors that the researchers didn’t account for.
Once I’ve read over the methods, I’ll move on to the results and discussion. Exercise-related studies will often test a long list of physiological markers. It can be important to look at each marker and not just take the improved one as evidence that the tested idea worked. For example, maybe a new supplement was found to reduce heart rate for a 40km TT but digging deeper, we find that VO2 max, TT time, power at LT, etc., remained unchanged. It would be quite a stretch to say that the supplement improved performance in this case, but I have seen supplement companies do this often.
Finally, at the end of the article, I’ll look for any mentions of potential sources of error, conflicts of interest, how the study relates to other research in the field, and what if any conclusions can be drawn.
Know the Value of Different Types of Research
Unsurprisingly, not all research holds the same weight. If I come across a case study of one or even a few people, it’s not nearly as valuable as a randomized controlled trial on the same topic. In this case, I would be much more likely to draw a conclusion using the latter. Generally, observational studies like cohort and cross-sectional studies have a lower weight of evidence than interventional studies. A systematic review or meta-analysis is of value because it examines many studies to come to a conclusion. Still, I would put in the caveat that even different reviews can come to different conclusions, and it can be important to look at which studies were included and which weren’t.
Take A Step Back and Look at the Whole Picture
This is perhaps the most critical step. After I’ve looked at the research, I try to step back and see where the evidence is pointing. Again, I’ll give more weight to the reviews and meta-analyses because this is essentially what those papers are doing. If there are a lot of conflicting results, then I have no problem concluding that I don’t know the answer and that more research needs to be done, which is a common ending to my videos. Even if the evidence points in a particular direction, I’m often cautious about talking in absolutes. I generally use words like “may” or “perhaps” very intentionally to convey that we haven’t quite gotten to the bottom of this yet.
There are certain cases where I will speak with much more confidence about where the evidence lies, for example, regarding the question of whether weightlifting improves cycling performance (spoiler, it does). In this case, almost every study on how weightlifting affects cycling performance comes to the same conclusion. And this is across a wide range of different study designs. I don’t mind talking with more certainty when this is the case. That being said, new research is published all the time, and if a study came out tomorrow saying that lifting didn’t improve cycling performance, I wouldn’t ignore it. I’d evaluate it and reshape my opinion based on how much weight (see what I did there) I think that study holds.
