Running with Xert for a PB in Parkrun

Hello everyone.

To start, please be understanding if there are any mistakes, as English is my second language. Since there are practically no discussions here (at least I don’t see any) related to running, I decided to write a bit about my adventure with Xert, which has only been going on for three weeks. However, I have been running with Stryd since 2016. For various reasons, I only have input data from October, with a long break for the holidays, so Xert doesn’t have much possibility to create the “ideal plan for me.” However, what I already see, I really like.

In March 2025, I will turn 55 years old. I am not, nor have I ever been, a professional runner, but I have been running recreationally with various breaks since 1999. I haven’t trained in any particularly structured way, sometimes more, sometimes less. In 2022, I was running around 39:50 min/10 km and 19:15/5 km. Two years ago, I think I experienced a mid-life crisis because I decided to achieve a solid time for 5 km, and last year I ran 18:57/5 km. Unfortunately, this year I had a big break from January to June; I did some running, but there was no real training involved.

As I mentioned, I’ve been running with Stryd since 2016, also with breaks from power-based training, mainly because no one offered any sensible solutions. I looked everywhere, as I see my Xert account was created in 2017 :slight_smile:

I also have a WKO5 and am well-acquainted with Golden Cheetah, but in reality, these are all data analysis programs, and you can draw conclusions from them. However, I’ve always had a tendency to create overly intense plans for myself.

I apologize for writing so much, but firstly, I enjoy writing, and secondly, I’d like to present a picture of my current situation. Perhaps someone will find this thread interesting, which I somewhat hope for. Therefore, I believe that in this case, the more initial data, the better.

Three weeks ago, I was already determined to follow Stryd’s plans until I remembered that Xert still existed and decided to check if there had been any changes. (I hadn’t checked in about a year?) It turns out there were significant changes. I bought a monthly subscription, and once I did, I started playing with the settings, reading the forum, FAQ, etc. I found it satisfying enough that a few days later, I purchased an annual subscription, and for the past three weeks, I’ve been systematically running according to the plans generated by Xert.

I run five times a week with the goal of participating in a 5km parkrun on March 29, aiming to do my best. I don’t have any specific expectations; I just want to conduct solid training throughout the autumn and winter, play with the program, gather data, and experiment. In Stryd (and here), I have my weight entered as 70 kg, and I stick to that, although currently, it’s 75 kg. As I mentioned, practically no running in spring resulted in a higher weight. By spring, I should be able to lose those 3-4 kg without any problem (I managed to weigh 66 kg in 2020 with a height of 177 cm). At the moment, after a few “breakthroughs,” I have a Threshold Power (TP) of 292 watts.

At this point, I’ll wrap up this lengthy entry. In future posts, I’ll present my training assumptions and share my thoughts on the generated workouts and how I generally see everything. Comments are most welcome, and I hope this writing makes some sense. If anyone is interested in checking out my workouts, here’s the link to my Strava profile.

Best regards,
Robert

2 Likes

Hi all,
Where I lived for the last 9 years, I have a really tough parkrun to run.
There are 2x 500 meters of solid uphill (it always blows because it’s close to the sea - Liverpool area) and on top of that, a whole bunch of turns over the 2 loops, with 3 complete 180-degree turnarounds.
Last year, at the same time of year, being at the same stage of preparation, just a year younger, I ran in better weather conditions in 21:15.
Today, the wind was blowing around 30/kph.
Xert predicted around 302 watts for me.
I ended up with an average of 307 watts and a time of 20:26.
I also got a bronze breakthrough and an LT update.
It wasn’t a target race; I just thought a week ago it might be worth seeing “where I am” with my current form.
I’m satisfied, and more details will be in the next post.

I would like to clearly state that Xert’s running training definitely works.
form me of course :slight_smile:

3 Likes

Have you tried mixing run and bike?

Disappointed with Xert – Likely Not Renewing Subscription

I’ve been hesitating for a long time before writing this post. I really wanted to give Xert a fair chance, but it’s looking like 99% certain I won’t renew my subscription in October. I’m writing this with genuine sadness because I had high hopes for Xert and was thrilled with it at the start. Unfortunately, the deeper I got, the worse things became.

I’ve been using Xert for running since last October, training with power using a Stryd for my workouts. I generally run on flat terrain, focusing on 5K-specific training. I’ve used Xert’s goal-oriented plans, setting targets for power improvement from 15-minute to 5-minute efforts. But these are just details. The issue is that I’m increasingly getting completely absurd data. Yes, my fitness has improved, but only because I stopped following Xert’s training recommendations for the last 8 weeks. If I had stuck to Xert’s guidance, I’d likely be overtrained or injured by now.

Here’s why I say this: while I don’t have major issues with the variety or structure of the training plans (I actually like their diversity), the threshold power calculations are wildly inflated. It’s a disaster. This hasn’t happened once or twice—it’s after most hard workouts. I get breakthroughs with unrealistically high power numbers. For example, at the end of March, I ran a 5K flat-out on flat terrain, giving 100%, with an average power of 317 watts (70kg body weight) and a time of 19:13. Now, I’ve gotten a new breakthrough showing an average power of 333 watts for 30 minutes? Meanwhile, I’m doing 6x1min/1min intervals at 340 watts, and I’m completely spent after that kind of session. How am I supposed to train at these ranges when they’re totally unrealistic? My current sweet spot is basically my 5K pace!

I know you can flag breakthroughs, but how many times can you keep flagging them? How am I supposed to trust Xert’s data when it’s consistently unrealistic? I’ve been running with a Stryd since 2016, so I know my optimal power zones very well. WKO5 shows my threshold at 299 watts, Stryd’s estimate based on my 5K race is 299 watts, and Xert is telling me 333 watts. Thanks, but no thanks for these numbers.





1 Like

It is a shame but if Xert doesn’t work for you best to move on.

I don’t run anymore and have never used running power. Your mention of ~300 watts for a ~20 min 5k does not seem to relate to cycling power where 300 watts for 20 mins is very impressive. A 20 min 5k does not seem as strong (to me at least).

1 Like

If you were a cyclist I’d say your peak power is way too low, which in turn is why HIE is low and threshold is too high. Advice would be to sprint to get a good peak power estimate, which should then give a more realistic signature overall (increase HiE, lower TP)… I’m no runner so don’t know what a good peak running power is, but principles may be similar?

I do sprints from time to time, and I understand their importance. However, for 5K training, they aren’t essential, and considering I’m over 55 years old, they can be quite injury-prone.

I don’t want my post to diminish Xert as a platform—I believe it’s excellent for cyclists. Unfortunately, for runners, the power calculations are completely unrealistic, and there doesn’t seem to be any progress being made on this issue.

Today, I ran a 5K race and finished with a time of 19:19, taking 1st place in the M55+ age category. My average power for the entire race was 325 watts (the course was hilly, up and down, which likely explains the higher power compared to my recent runs). I received a gold breakthrough, which is exciting, but my Threshold Power was also increased by 1 watt and is now at 329 watts, while my 30-minute power was calculated at 330 watts—quite miraculous, to say the least.

In contrast, WKO5 raised my power from 299 to 304 watts, which feels much more realistic.
I’d like to point out that when I purchased a year-long subscription upfront,
I was expecting the promised accuracy.
I really like Xert’s overall concept and the suggested workouts, but the power calculations for running (perhaps this is specific to me, I’m not sure) are far from realistic and, frankly, quite illogical.



Zrzut ekranu 2025-06-8 o 12.42.03

It’s only as accurate as the data you feed it, and if that peak power estimate is well below your true peak, everything else will be off

If you don’t want to or can’t sprint you can put an estimate in… go to your latest BT, put in say 900w for peak power, make HiE eg 16, and threshold around 300 then click extract to get a more realistic set of parameters which match your BT… then click save to lock the signature…it should then work from then on…

But I do sprints, so Xert gets a chance. I can’t go above 470 watts, and that’s what I entered.
Besides, Xert promises that LT is calculated based on current and up-to-date training data and determines the most realistic power, but the current calculations are way off.
As you can see, WKO5, which has the exact same 1:1 dataset, does it much more realistically.
Those 304 watts are absolutely realistic when it comes to LT, so sprints have nothing to do with it.
Leaving aside the general LT power calculations, how can a max effort race of about 20 minutes be lower than the 30-minute power calculated by Xert? And on top of that, I get a golden breakthrough. It’s like dividing pears by apples and ending up with gooseberries."

For what it’s worth, in my case, I followed a Stryd training program for a 10K race. After the race, Xert registered a breakthrough with a TP of 305W, which is very close to Stryd’s estimate of 304W—so the two are fairly well aligned. That said, back in April I had a golden breakthrough in Xert showing a TP of 314W, which seemed a bit high. I even flagged one breakthrough during the program because the output didn’t feel accurate. Overall, I tend to trust Stryd’s numbers more in this case.

1 Like

I appreciate what you guys are saying, but it has nothing to do with the topic of discussion. Xert is showing absurd numbers, and as I pointed out in my first post, while I understand flagging a trend, like for wind or excessively high power, I can’t be doing that every few workouts—what’s the point of that?

The last 8 weeks of training were workouts generated by Grok, and they didn’t differ qualitatively from those generated by Xert.

I really want to stick around here, but how am I supposed to trust data that tells me to push harder for 30 minutes than the power I managed in a 20-minute race today? That’s just ridiculous.

The screenshots you posted above show numbers much less than that… first one 394 and the second 419? And is 470 truly a max for eg 5 seconds, when fresh? That’s what you need for ‘peak’ power, and it’s unlikely to come at the end of a longer race

Anyway, just trying to help as you have a few months left on your subscription.

Not sure you wrote an email to the support team but if not, I would suggest that - they are super helpful

Hi,
I decided to listen to those of you who suggested doing short, intense sprints. So, I did 8x8-second max uphill sprints with power over 500 watts, and today I also did a 3-minute max and a 10-minute max to set my CP according to Stryd. As I mentioned earlier, both WKO5 Stryd and Xert have been using the same identical input data since October last year. Stryd calculated 306W, WKO5 calculated 307W, but Xert barely changed anything, despite another breakthrough today and after the sprints. 322 watts is my 5K race pace, not my LT. This will be my last post here, which is a shame because I considered Xert a very valuable program. Perhaps it is for cyclists, but unfortunately for me, as someone who only runs, it provides completely inaccurate data that would have easily led to overtraining if I had followed Xert’s parameters.