I agree. But I'm usually ready to move on after 200-400 hours of playing one character anyway. Ready to try another build and put on another skin.

Okay, then which is it? Is this a game which centered around difficult challenge provided by other people in which losses had long-term, lasting repercussions – or game in which you could regularly play 200 400 hours of a single character in a single build?

Because these two things don't describe the same game.

And yes, I know the inevitable "I was just that good" line is about to deploy, but that seems pretty meaningless if we are talking about game design in general. Maybe you were, but you are taking the position that the system that you held up as an exemplar of the things that you want clearly didn't deliver those things. Either you've misremembered, or gameplay for the average user was very, very different than your presenting it.

All of these are for level 30+ out of 51, so none directly after creation. There isn't a limit of 1 char per player, so most are likely due to more interest in another character. (AUTO) means the character was abandoned for an extended period of time. You can click on any name and see how many hours the character had been played.

That's a list of characters which are all level 30+. That doesn't tell me anything about what retention was at lower levels. And I can tell you pretty comfortably, having administrated on more than a few LPs, player retention from the early game could be quite brutal, especially in games where PK was life.

There are a lot of abandoned accounts here, though. Given that all of them are level 30+, most of them have a good number of our sunk into them. That people just abandoned them (to do something else, or to take up a new character) isn't a great sign of health for a game. You would think if the ongoing gameplay is compelling, if nothing else you would want your character to go out fighting, picking things way out of your reach just to make it interesting.

By more competition, I mean depth of strategy and depth of tactics. Other players provide the best competition, in my opinion.

Yes, but those words don't actually mean anything in terms of game design – which is pretty much one of the great problems of trying to get people to tell you what kind of games that they want. There are excellent for describing games that already exist, as a method classification. They are absolute crap for describing what you want to see going forward.

What kind of competition? Direct clashes? Economic competition, where one group attempts to out produce the other and thus do better in some marketplace to take away from them? What level of strategy? Grand strategy, where you give direction at the highest levels and must trust that your subordinates are capable within their parameters? Lower level strategy, where you control the battlefield, pick your places of engagement, and trust your subordinates to do their jobs?

(You may come away with the idea that strategy and tactics do not get on well came mechanically – and you would be right. In real-time strategy games, while micromanagement of individual tactical engagements as become "the norm" in games as designed, most of the time the strategy is of the most rudimentary sort.)

There are reasons for these questions.

The top four most popular online games right now are all direct competition, and all have more active users than World of Warcraft: Dota 2, Overwatch, Hearthstone, and League of Legends.

And none of the named games have anything but the most flimsy veneer of role-playing over the top of mechanics which are completely and utterly devoted to resolving direct conflict. There's a reason for that. They have characters, but they have no roles in the sense of a narrative. They use the word roles when talking about their tactical disposition on the battlefield, but it's not in any way the same use of the word roles as found in role-playing games.

People don't want role-playing in their conflict games because being attached to the character gets in the way of engaging in conflict – and the conflict, the direct, physical, high loss/high gain conflicts are the meat of why people want to play those games.

We are definitely not going to pull people into role-playing by enticing them with competition, because we already have that – wargames exist. They have existed for far longer than RPGs, in a strict sense. And there is role-playing that occurs within them, very often. Even in as constricted a character space as "World War II Wargaming" there are a multitude of roles – they are just not roles fulfilled by single characters. They're roles that are fulfilled by squads, platoons, companies, Battalion, regiments, armies, task forces – they're roles that are individual and customizable by swapping out traits (in the person of individual figures), and they're roles that often involve individual characters of notable recognition (like Heroes in the latest 40k edition).

Check out something like Kingdom Death: Monster – which is apparently doing pretty good, despite the insane price tag. It's a game where conflict and strategy and tactics are all right there on the board, with individual characters engaging in squad-level conflict, in an ongoing manner.

People who want direct competition in their role-playing already have an entire historical era of gaming to cling to. The entirety of OSR is predicated on that style of gameplay, where it is the players versus the GM it in a specifically and clearly adversarial position.

An argument could be made that Traveller, run as it is traditionally depicted, is just as adversarial from the GM position. I would suggest that when it is run such, it is a lesser game.

People that want role-playing want that role-playing to be meaningful. They want those role-playing choices to make a difference. The problem for a designer is that there are a lot of direct conflict decisions to make moment to moment, and role-playing choices aren't nearly as dense – they require a lot more set up, a lot more development, and a lot more time for payoff. If you put them both in your game, side-by-side, then your perceptual profile of gameplay turns into a lot of conflict decisions, all of which – per your request – carry a lot of weight, and only a few role-playing decisions which may carry a lot of weight individually, but happen too rarely for that to feel like a part of the game that the people who are looking for role-playing can really get into.

For a videogame version of this conundrum in play, look at Mass Effect. The first in the series was a pretty good attempt at balancing the two sides, role-playing versus direct conflict, within a narrative. But as the games proceeded, it became obvious that the designers had the fact that narrative and direct conflict with meaningful result are at definite odds. They decided which side of their bread they wanted to be buttered, and it turned out to be the direct conflict side, letting the role-playing side wither away, but never entirely. Combined with crappy writing decisions, this made both sides unhappy, because neither felt that they were getting enough attention from the game design. And that led to Mass Effect: Andromeda, the less spoken of the better.

I've had some real success, surprisingly, getting wargamers into role-playing games – because more gamers are already playing roles; they're halfway there. Once you focus down to squad based/man-to-man wargamers, you're not only standing there with a foot in the door, you've simultaneously have crawled through their bedroom window.

Maybe you should take the time to go and see what's available in the wargaming space. It delivers largely what you want – direct conflict, often head-to-head with other people, the ability to take on a role, actual role-playing. It's got it all.