Removing new game barriers

[prioli]

This issue is important, but less trivial.
It may not seem like an important issue unless you put some thought into it

How do you reach a situation where there's always a game to join? Always someone to play with?
Well obviously you need players for that. But how many?

Well - if you're a beginner, then it's relatively easy.
You can either always play a bot, or always craft the UI (for example in the app) to do the following:
- Join a random open Ladder game, or create a new Ladder Maladroit ParaFC if there is none.
with this logic you don't need many beginners to always have a game to play.

But what do you do about players with Ladder >5 ?
I claim that currently there's a negative feedback loop. Less public games --> less games to join --> less players sticking around --> less public games


You can open a Ladder game and wait for someone to join, but there aren't many players to join in your +/- 5 ladder
You can open a Friendly, but if your elo is above 1866 you'd be losing points even if you win against a 1500, and even at elo above 1650 or 1700 this doesn't really make sense.
So, the better choice would be to open a private game, or a VF.

Private games are well... private. So less open games to join, and VFs are nice but competition is a major driver in games and retention.
so you get: Less public games --> less games to join --> less players sticking around --> less public games

Solution?
A simple solution might be to change ladder games so that you can always join a ladder game if [your elo > (other elo - 100)]
And the actual ladder score logic wouldn't change - so you get a ladder point only if the ladder score diff is <=5

Why does it solve the issue?
You can always open a Ladder game without being worried that low elo would join - so no barrier for opening
Everyone that's about your level or higher may join your game - so more games to join
...and it doesn't hurt the ladder scoring, because you get a ladder point only if the diff is <=5.

In fact, friendly games wouldn't be needed after that. You'd only have VFs and Scored games, where in scored you'd get elo and Ladder if applicable.

How can it be used to grow the number of games and players?
It would then be possible to change the UI, for example in the app, to simply have a "start game" button,
that by default lets you join any open ladder game, or create a new one if there is none.

Any thoughts?

[BioLogIn]

I agree, that the current system is quite cumbersome and does, in fact, dissuade players from sticking around.

But as for actions you suggest, at the same time I am not sure if it would be enough and I am not sure if RB would be willing to implement it.

If RB would not be willing to implement it, then the discussion is moot. If RB is willing to implement a change in ratings, then I think we can go in a slightly different direction. I think it is the ladder system which should be ditched, and the ELO system kept:
- ladder points should be retired, possibly converted to something of achievement kind of thing;
- when creating a rating game, by default it would be created with 100 ELO range (around current ELO or the creator); but with options to create rating games with 200, 300, 500 range should be there as well for those who need it.

In general, having two visible separate ratings is vastly confusing to players, especially if a certain player is top1 in one rating system and bottom1 in the other. This kinda invalidates both systems for the playerbase.

[prioli]

BioLogIn

I agree that having 2 systems is confusing.

I didn't quite understand the problem you were trying to solve, or the insight behind the solution.
The problem I was how to create an incentive for players to open public (non-private) ranked games.

I agree that elo is much more interesting for advanced users, but it's not friendly at all for beginners
I think that this was the thinking behind RB's plans.

My feeling is that right now it is better to find a KISS solution (Keep-it-simple-stupid) with very little options,
and aim to find a way to create a working funnel from newbie --> playing, understanding, enjoying --> advanced

[BioLogIn]

Sorry, but I find it hard to agree that ladder is newbie-friendly; RB ladder (specifically losing all points after one death) gerenally discourages people from playing ladder; only the most hardcore players are motivated to climb it again and again. Please refer to basically any modern duelling videogame, they all use either pure elo-like rating systems (mostly RTS), or they use ladder (mostly CCG), but even then the points that are subtracted after a loss are always equal or less to the points that get added for a win.

I understand that you are trying to solve one problem at a time. But I don't think that things such as matchmaking system can be viewed or improved in isolation from rating system. I think that only holistic approach works in such cases, and that solving one small problem just bring us closer to the larger problem. In other words, removing some barriers on the entrance to the system while the system is flawed is the same as reconstructing a section of a road while leaving the next part of the road narrow - all you get is you move the traffic jam one intersection further.

Also you can keep the interface as simple with my proposal as with yours; the default "new game" option will join/create a game with 100 elo range in both scenarios.

[prioli]

BioLogIn

So, it seems that we agree that:
1. It would be better to have one good system
2. Ladder could have been good if it were proper ladder (no death to zero, equal points for win lose, maybe death = +/- 2 for win/lose)
3. I don't object to a good elo, but today's elo system means that as a beginner you have no progress for a while, and I didn't fully understand the rest of your suggestion for good elo.


Maybe we could go with:
A. New players get 10 points for every victory until 1600 points, and can't lose points if their score is below 1600.
B. You can join/create an elo game if your score is no more/less than X from the player you joined
C. In this case if we only go with elo then X=100 is probably a little low, especially now that there are very few player around. Maybe X = 150.

The place where I don't feel I agree with you is about strategy for solving.
You say there's no point in solving one bottle neck as long as there are other bottle necks.

I think you agree that we're very few players now playing the game, with very little resources and power to improve the situation.
So are you saying there's no path that can take us to making this game live again?

[BioLogIn]

prioli

If we are looking for a "good" ELO system, I suggest considering Glicko-2. It offers more modern approach than classic ELO implementations. It is a bit tricky "under hood", but it works nicely in terms of converging on player's rating faster. Although I understand that simpler systems have their advantages, especially with population as low as we have in warlocks.

In terms of approach to solving the problem of warlocks population, I say that I see no path I am willing to invest efforts into (and I am generally quite willing to invest into this game, and I think I've proven this some years ago). But I am not saying that this decision should apply to everyone. Others might have a different approach or a different perspective, and I certainly don't want to dissuade anyone, and that's why I did not comment here before you kinda asked me to ingame =)

[prioli]

BioLogIn
Fair enough

This is not the immediate bottle neck.
If we reach this bottle neck, I hope we'll find an agreed solution that's simple enough for RB to implement,
and I hope you won't consider adding your name to the request a big investment

[BioLogIn]

Agreed =)