I think that using distrowatch's analytics is a particularly unreliable method of estimating popularity. For example, I can't even remember the last time I visited distrowatch--and I would guess that a vast majority of desktop users fall into this category.

I'd guess is that distrowatch is visited more frequently by people who are learning about linux rather than actually using it. This is known as statistical skew.

In other words, this would be like estimating total breakdown of most popular foods worldwide by basing it exclusively on restaurant sales in one random college town in the US and extrapolating it out.

As far as whether you're in the right lane, I pick a lane based on my requirements and experience, not viral fad trends that get popular because some influencers told a bunch of noobs about a particular distro.

Mint is a good example of this: noobs think Mint is easier, because it is designed to appeal to noobs migrating from Windows and is marketed and branded as such. But Mint is also more difficult for more experienced Linux users migrating from KDE or Gnome, or people migrating from macOS. But really, what is the difference between Mint and Ubuntu? It's essentially the default DE, default themes, a handful of default packages, and an additional partial repository (which itself is becoming less relevant as time goes by). But most people will develop a DE preference over time (gnome/kde/cinnamon/etc), change their desktop background or colors, and install/uninstall the apps they want, including via universal packages like flatpaks that don't rely on your distro's repos.

Over time, Linux distros are getting more diverse in the little default things; but they are converging in the big things. Picking a distro today matters significantly less than it did 10-20 years ago. A few reasons:

• Apps have trended toward universal package managers like flatpaks, appimages, snaps, etc. Package management is the entire fundamental purpose of a distro; and if people use universal packages, the distro doesn't matter. So now, the only packages a distro manages is mainly back-end system stuff that most people don't notice or care about. The actual apps people use aren't managed by the distro.
• The kernel has gotten more flexible. Example: Where you previously needed a distro to provide a separately tuned low-latency kernel, low-latency is now a flag in the standard kernel (and the low-latency kernel is on its way to becoming deprecated). Another example: many kernel tunings can be done while logged in at runtime rather than complex configuration / reboots like previously. So how your distro tunes the kernel matters less too.
• There are more and better semantic standardization layers today than previously. For example, back in the day you had pulseaudio server for desktop apps and jack audio server for music production apps--and this required complex setup, so we had distros like Ubuntu Studio. Today, we have pipewire, which is preinstalled on most distros and handles both pulseaudio and jack apps. The display server is in a similar boat with Wayland vs X11. As are some system file locations.
• The ecosystem and critical mass of users has matured quite a bit. So somewhere, a user has ran into what you are trying to do and has made an app or a script. Want to install Davinci Resolve (which is designed for RHEL / Rocky Linux) on Debian? Simple: makeresolvedeb. Want to install it on Fedora? Easy: DaVinci Helper. Arch? No problem. Most popular distros today are some derivative of one of these; and these will work for those.

As trends like this continue, the differentiation between distros becomes more trivialized.