It seems like when you restrict the time period of the data it also restricts the data to which you’re comparing. That is, my absolute max watts are lower for 2025 than 2024 (no surprise, only 2 months into 2025 and no racing yet.) But my percentiles for 2025 are higher, suggesting it’s comparing my 2025 to everyone (or not everyone, just men 40-50 or 40-45 or ??) else’s 2025, not to the same denominator used for the “all time” percentiles.
Anyway, cool tool but it’d be nice if it were clearer exactly what it’s showing us.
Maybe I’m in the minority, but I’d rather have something released in beta (or alpha) vs have a company sit on something until they feel it’s totally baked and ready for release. Because it’s never really baked until you release it to the world and get some feedback/criticism and then adjust. Give me half baked and buggy new stuff all day long as long as you aren’t breaking the core stuff that already works.
Criticism is a positive normal thing and I bet the dashboards will improve based on the feedback. Companies that get overly concerned or defensive about new product hiccups tend to be very risk averse and never innovate. Gotta break some eggs and all that. The funniest thing is seeing people bash on SpaceX every time they blow up a rocket like it’s some kind of failure when that rapid iterative approach is the cornerstone of their success.
@Nate_Pearson I’m curious and it doesn’t specify on the page itself – who is this comparing you to? TR users with the same gender and/or age bracket? It’s telling me I’m in the >90th percentiles across the board. Is that just because the dataset for women is smaller?
I am curious what this means though: “…other athletes like yourself — real TrainerRoad athletes striving to get faster.” This could mean everything from literally everyone to those who are similar in age, weight, height, w/kg, FTP, how long you have been on TR, what country or state are you in… (ok you get the point).
Interesting new feature. It definitely reminds me that I need to see what my 1, 5, and 10 minute efforts can be. I’ve only really tested myself on 15/20 minute, and 3+ hour efforts. My 5 minute PR is within 3 watts of my 15 minute PR, and my 45 minute PR is only 4 watts more than my 3 hour PR.
I find it interesting that my evaluation categorizes me as a climber. I understand that my lighter weight (71 kg) probably helps contribute to that, but my w/kg (3.8) isn’t anything too wild. The funny part is that I live in Houston, where my typical 65 mile ride might have around 1200’ of climbing. My training involves almost no actual climbing. Regardless of the logic behind the classification, I do really really love climbing. Last year I went to Colorado and over Christmas, New Mexico, and my perfect ride is just grinding it out straight up a steep mountain. I’m planning on the Triple Bypass in July, which is around 11k feet of climbing over 118 miles.
I was surprised by a large discrepancy between last year and all time. Yes 2mths of 2025 against all time would be apples and oranges, but 2024 vs all time should be roughly comparable.
Example: My 20min all time is 317 (85.2%) vs. last year was 309 (95.2%). Same across other power durations, higher power in my all time, but corresponding lower ranking. And in 2024, lower power but higher ranking. What happened in 2024 that wiped out everyone else?
309 is not much lower than 317. So for you, 2024 was close to your best ever. Maybe on average, people’s 2024 was further from their best ever.
You might be relatively young (or young in training terms, or at least not in a trough)? On average, at any given point in time, people will be further from their all time best than that.
I noticed the same thing. I suspect 2 possible reasons -
The 1 year comparison is against a different slice of the population compared to all time
The TR general population is getting slower. Just to be clear, I’m not saying individuals are getting slower, but if TR is growing and adding a bunch of subscribers, that would naturally slow down the overall population as there are more beginners in the mix pulling averages down (and thereby raising the percentiles of more experienced athletes). Especially since TR has added a lot of features that make it more attractive to less experienced athletes.
Ohhhh that’s another idea/improvement when you have the bar chart sent to all time best it would be nice to say what year each interval had that all time best.
I feel this would be way more useful if the denominator were held constant. That is, regardless of which of my own seasons I pick, I’d like it to compare that to “all time” for other users. That way you could actually have a comparison between your own seasons, and percentiles would mean something.
As currently set up, since it compares my 2025 numbers to others’ 2025 numbers, that sample is small and the year is young so it makes no sense in comparison to 2024 or 2023. Plus if you belong to a group with fewer users (e.g. an older female) the data will get really noisy if using only current year as a comparator.
All that said it’s a cool, free feature and I don’t mean to slag it off! Just suggestions for improvement.
I completely agree with this. There’s no obvious way to make this really meaningful, after all - comparing myself to a large pool of others is ultimately not that important, I see this as a nice fun feature. With that in mind, I think it’s less confusing if the demographic filters remain (but are made known/visible), but that each of my years’ performance levels are compared against the “all time” values of others, or some kind of “average per year” performance of others to make it less pessimistic.
Pretty cool! Sort of akin to Wahoo’s 4DP without the testing! Just looked and mine was pretty accurate relative to what Wahoo has told me in the past. Will be curious if they extend this out to your “weakness” to see if that also correlates to what Wahoo has told me before. I don’t have SYSTM anymore so won’t be able to compare real time.
