Steamspy figures mostly show how many people own the game on Steam, but it doesn't really discern how many of those were sales at full price, how many sales at a discount, how many sales from bundles, and how many were free key activations from giveaways. It is a good indicator of a "general" idea, but it's not too precise.
I would confirm though that my older games did receive much more exposure and sold much better than the more recent releases. Steam market has become saturated, and also its new algorithm has made it even harder for indie devs to get noticed. As some people stated already above, you basically need to be popular already to get more exposure, so the popular games become even more popular and unpopular sink down even further. I think I saw year-by-year figures not long ago showing that roughly 90% of all sales on Steam every year are by top 20% of games. That means the lower 80% of games get the tail end of whatever is left over, and that would undoubtedly include 99% of indies.
I wouldn't go into specifics, but I would say that about 2-3 years ago, each new game of mine had noticeably more traction than these days. It will also strongly depend on the game itself of course. Just to give you some idea, two games I released last year: 1) Mercury, which is a cyberpunk RPG with an engrossing plot and choice & consequence mechanics, 2) Renegade Grounds, a short dungeon-crawler with bare-bones plot but more emphasis on strategic combat. Both had roughly similar budgets (about £600 or $700-800). They also both came out at roughly the same time (Mercury came out in mid August, and Renegade on the first day of September, so Mercury has a 3 week head start, but it's negligible by now). As of right now, Mercury has sold around twice as much as Renegade Grounds. It's already broken even. Renegade Grounds hasn't broken even yet. By current Steam standards, I consider Mercury to be a success. I know had it come out 2 years ago, it would've done as well as City of Chains (which to this day still manages to outsell most of the more recent releases each month). In the past I had a game break even in 2-3 months or so. Now it takes roughly half a year (in Mercury's case) or longer (in Renegade Grounds case).
I have released another project, just last month in fact. Snares of Ruin. I am still waiting for January royalties, but going from the numbers my publisher gave me after the first week of release, it did pretty well and already better than Renegade Grounds did in half a year. So, the game genre matters A LOT. I think also if you've been lucky enough to get a fair few positive reviews in the first week or two after release. If you haven't, then it will be an uphill struggle for that game (Renegade Grounds still sits at exactly 3 reviews last I checked, two of which were negative and there haven't been any new ones for months; by contrast, Snares of Ruin had about 7-8 reviews after just 1 month of being out and last I checked, they were all positive).
So I guess what I want to say is - you have to experiment a bit but don't put all your eggs in one basket. See what does well and what not, and then make more of that which does well. There is no guarantee your first project will do well. Or your second, or third. But if you then compare, you will get a good idea of where to go from there.