There's been a recent trend in large, popular titles that are sequels. And that is to make the game simpler than its previous incarnations, assumedly in order to broaden the potential audience of the titles. Examples include The Elderscrolls and Skyrim, which included the popular mechanic of regenerating health so "people don't have to worry as much".
The hypothesis is a common one. Outside of combat people are just going to go about and do something to regenerate their health anyway, so why not speed that up for them and get it over with? Make it so they "get back to the game" faster.
This assumes firstly that getting your health back isn't part of the game and isn't fun. And a lot of aspects in games are treated as such. It's "not fun" so get rid of it. But even more recently there's been a trend of taking these "not fun" aspects of a game and finding out how to make them fun. "Roguelikes" even go so far as to take death itself, a normally frustrating portion of games where one might just reload your previous save, and actually turn the entire process into something fun and enjoyable.
Other games have done as much with regenerating health. The Last of Us, critically acclaimed and with an excellent for a platform exlusive as well as new IP 4+ million sales, makes regenerating health tie into its theme of a desperate post apocalyptic world, where "health packs" are rare and valuable things, making combat all the more intense as not only do you not want to die, but encourages the player to take as little damage as possible.
So instead of just calling a portion of a game "frustrating" and cutting it, trying to make that portion fun can clearly have benefits for making the entirety of the game more "fun" as well.
But that's not all game developers have done in recent years. Another trend is to take complexity out of games, even if it is fun, just for the sake of making your game more "understandable".
For the sake of simplicity we'll go back to Skyrim and show its purely linear dungeons as a more simple, and easier to understand area than the series previous, less linear dungeons.
Now the hypothesis is that by making playing the game simpler it will appeal to more people. And I simply aim to show that this is, perhaps not nonsense, but completely unnessary if it comes down to a choice between more interesting games and a "broader appeal".
The first point to make is that complex games have a broad enough appeal to be bestselling games anyway. Minecraft has a near archaic and archane crafting sysyem, and yet is one of the most successful games of all time. League of Legends can take quite a while to learn, in fact learning all of it is half the game in itself, yet is undoubtedly one of the most successful games of all time.
The Sims needs no introduction to its sales numbers and popularity, and is massively complex. The key idea behind it though is that its complexity is something people understand already, you control small "people" and their everyday lives, and so any action you can take and its probable consequences are already understood from your own similar life experiences. It's an elegant solution to a problem which has the same ideal solution: that making your complex game understandable can by people is nigh as good a solution as making it less complex.
The other concept to note is that players being able to understand your game is only one factor limiting the breadth of its appeal, and this breadth may already be constrained more by these other factors. Constraints can and do include your games price, thematics, marketing, time input requirement (how much you need to play it to enjoy it), system availability (do you need to be on your tv with a new PS4, or can you play it in the bathroom on your old smartphone?), and other things. Just assuming making your game simpler will somehow allow it to reach a wider audience may be an inherently wrong idea, if you have other bottlenecks elsewhere you may not reach many more people at all.
And finally there's one key concept to make in all of this: 99.9% of the population of humanity could understand calculus given the right motivation. People you know personally may not seem capable, but they are. They are just not going to put the time and effort into it that such would take. Now your game is hopefully not calculus, but if it is fun, and you can show people that its worth their time and effort, they will put that in for you. People can understand a seemingly remarkable amount of complexity, but they will only choose to do so given the right motivation. So you don't need to make your game less complex necessarily, as long as you give your players the right opportunity, the right methods, and the right motivation to understand that complexity.