Gauging Interest: Drop-in/Drop-out Ranked-Forums format

Yomi
1

One of the reasons that I keep running tournaments here on the forums is that I want to keep playing Yomi, and the tournaments are the best way I’ve found so far to get at least one game scheduled. However, they can be intimidating to new players, and are either a pretty heavy time commitment (for IYL or a swiss format) or potentially a small number of games (for players that run 0-2, for instance). Inspired by the latest game design podcast from @Sirlin, I’m considering trying to run basically a forum-based ranked mode.

The goal would be to have a pool of players who are interested in playing, and each week you’d be assigned an opponent. The mechanisms for ranking/match making are TBD, but ideally the format would be such that you basically don’t get heavily penalized for stepping away for a week or two (it’s an opportunity cost, but that’s it). Also, like any other ranked mode, rather than having a definite end-point, you’d instead have a running list of who the top-ranked forum players are.

I’d be willing to put in the time to streamline this, if there’s a body of players who would be interested and would likely sign up to be scheduled for a weekly game. So, takers? Questions?

9 Likes
2

This is essentially a ladder, yes? They tried to run one a long time ago and it sort of fizzled, but perhaps it’s time for another try.

1 Like
3

I think the intent is much the same as a ladder, but the mechanics would probably be different (although that could certainly be up for discussion).

1 Like
4

I would be interested.

But wouldn’t this fill much the same role as the round-robin part of the IYL?

With the obvious difference that the IYL only runs for a limited time and ends with a playoff, of course.

1 Like
5

Yeah, my main goal for this is to make it easy for new players to jump in at any time. It’s rough to have a player show up right after IYL starts and have them be told “Welp, come back in 3 months when the next tournament starts”.

3 Likes
6

Ah, I see. Didn’t catch that.

Great idea!

2 Likes
7

I am very interested. I was told 2 years ago “don’t waste your time sitting in QM, nobody uses it”. I’ve been practicing with @variable since Jun '17 weekly and I really enjoy it. Wouldn’t it essentially replace the leaderboards? There are so many corpses.

If Yomi was able to retain the people who expressed an interest, but left because they couldn’t figure out how to get matches without hitting QM, then we could increase the player base up to at least @mysticjuicer 's YouTube subscription numbers.

Also, I would like to request a hack showing how to change the background image in the Yomi client. I’m tired of the Dojo… whatever… that image is.

3 Likes
8

As one who will likely go 0-2 in tournaments, i’m interested. :slight_smile:

1 Like
9

I’m very interested in this! It also seems like a good way to appeal to newer players, or players with more time commitments! :smiley_cat:

Would the ladder rank be determined through points, @vengefulpickle? Let’s say that each player would be given one match every two weeks. Matches could be played as a best-of-seven format in which the winning player in each game earns 5 points & the loser earns zero.

(Long aside: Maybe perfects could be worth more? 7 points, or so? Also, maybe we should use the matchup chart to determine the points the winner receives, instead, Lum’s Long Odds-style! It could be trickier to reward players for perfects then, but maybe a perfect could be worth extra points? Like, 2-3 more than usual or something?)

Anyway, if a :grave: fought a :chibigloria: in the “Five Points to the Winner!” version…

:chibigrave::psfist::pschip::gloria:: 5 points for Grave.
:chibigrave::stuck_out_tongue_winking_eye:: :pschip::gloria:: 7 points for Grave due to a PERFECT!
:grave::pschip::psfist::chibigloria:: 5 points for Gloria.
:grave::pschip::psfist::chibigloria:: 5 points for Gloria.
:grave::pschip::psfist::chibigloria:: 5 points for Gloria.
:grave::pschip::psfist::chibigloria:: 5 points for Gloria!

(Note written in the margin: Gloria’s probably ((very)) slightly disadvantaged if perfects are worth more than just getting out of there with a win…)

So, in the end, :chibigloria: wins the set over :grave: 20-12! Since Gloria won the set in game 6 there won’t be a game 7, but Grave still has those 12 points to his name! Those results would then get added to their current points earned!

Players can only add to their ladder total twice per month, which allows players who can’t play as often a train that’s harder to fall off of! :train2: I was thinking that the remaining players who were interested would be seeded based on their rank in the ladder at the end of the year & that all players’ points would be reset to zero at that time! I’m not sure how exactly to arrange matches at the start though, because I don’t know the first thing about TO-ing & because I haven’t thought about it… AT ALL!! :scream_cat:

This is just an idea… Maybe it’s a bad idea, though! If it is, we should rethink it & come up with something better! It might be better if players weren’t assigned matches, and could just play against another player on the ladder at their convenience. I foresee issues with this setup, though! :scream_cat: I wouldn’t want a player who plays poorly in 5 matches to outrank a player who plays very well in 2, for example! There are probably other issues with that, as well…

However, I do very much look forward to this format being tuned & tinkered with for a bit & then existing!! Let’s figure that out!!! :techlab::smiley_cat::heart:

10

My plan was to crib from a style of ranked play that Sirlin laid out recently (and which is going to be used by FantasyStrike).

Conceptually, players play in asynchronous/late-bound 8-man single-elim tournaments. You get 1 point for winning in round one of the tournament, 2 for round 2, and 3 for winning the entire tournament, at which point it resets. You lose one point for losing in a tournament. (An alternative formulation is you win one additional point for each consecutive win up to 3, and then it resets). Accumulating enough points gets you moved into the next league (Bronze, Silver, Gold, etc…)

The matchmaking would be done by pairing, in order of priority:

  1. People with the same number of wins (this gives you the virtual tournament setup)
  2. People in the same league (this will hopefully get people paired with folks who are mostly at the same level as they are)

In the forum based version, my intention was to run match-making once a week. Anyone who was currently in the pool would be matched up (as long as there are an even number of players).

Some things that could end up being bad about this format:

  1. Not enough people to sustain the win-counts. If there aren’t enough people in the pool, then the virtual tournaments get pretty messy, effectively generating a lot of byes, which wouldn’t be great. EDIT: I think you’d need at least 16 in each league, so that each week you could start a new 8-man with all of the people that lost the previous week.
  2. Too slow turnaround to allow effective stratification. If we only play one match per week, and you can get (on average), 2 points per week, but we require 25 points to move between leagues, then the league separation is probably too slow to be useful, and we end up pairing experienced and new player frequently (which results in frustrating noob-stomps, in all likelyhood).
  3. If someone ranks up, and then drops out of the pool, then they are effectively unassailable… But maybe that doesn’t matter, or maybe we would only list the active players in the current rankings.

Around allowing manually match arranging: one of the things that I actually want out of this is to take the assignment of players to matches out of players hands as much as possible. Just like jumping into QM, the barrier to entry is lower if you can just say “Give me a match” and you get one, rather than having to negotiate one for yourself.

4 Likes
11

I would enjoy playing a scheduled and competitive Bo5/7 set on a weekly or biweekly basis.

I don’t really like the idea of this being “internally ranked”, by which I mean I’d prefer a system based on each player’s lifetime ELO or win rate since we have those stats available.

An Idea: I personally like the idea of manual match arrangement, I think it would increase the likelihood of a match actually happening if one party deliberately reaches out. If we had a known pool of players potentially interested in a (bi)weekly competitive match, one player could post a particular time they’re available and others could respond if they can meet that time, with priority going to more even (based on ELO or win rate) matchups if multiple players can make that time. Each player is capped at one game per week or per two weeks, and gains a point for each win. Points reset every so often kinda like how QM does.

12

Around using Elo/win rate, I think there are two objections I have to that:

  1. Per Sirlin’s original description (which I think I agree with), it’s interesting and valuable to not always be paired up with people that are exactly at your level (that’s one of the things that makes tournaments fun). A system that just uses Elo or win rates is more likely to pair very close players all the time than the proposed system.
  2. The implementation is actually more complicated on my end, if I’m going to be automating this. In particular, doing a pure-Elo system that doesn’t tend to match you up with the same set of people every week would be hard (and I’m not sure that people move around enough in Elo that that wouldn’t happen). The proposed system, on the other hand, basically guarantees that you won’t face off against someone you just played (because you will be at different numbers of wins).

My concern with manual match arrangement is that I think it ends up with a more intimidating barrier to entry. I want this to be a forum-based replacement to QM (because the playing populace doesn’t support a synchronous QM experience right now). For that to work, I think you really need to just be able to opt-in to being matched, and then be given a match, rather than having to pro-actively seek out a player with the right ELO/time schedule. That said, taking into account schedules would be an interesting extension to the matchmaking criteria, in the case where there are multiple matches that meet the other criteria I listed.

4 Likes
13

I agree with not using ELO. a) I don’t want to play against mastrblastr every week; we’ve played enough. b) People get better by playing people better than them. If everybody is always swimming in their own ponds then player growth will be slower.

2 Likes
14

Can I get a show of hands (by way of Likes of this post) of how many people would be interested in at least trying out a format like I proposed above? It would involve some amount of upfront time investment on my part, so I want to make sure that there’s at least a chance of getting enough players to keep it running for a little while. (Keep in mind, too, that IYL is going to be starting up soon, and it would be great if this could run concurrently w/ IYL, in my mind).

11 Likes
15

If my anxiety curbed down id try this out at least.

3 Likes
16

There is a poll function on these forums, right? Seems like that would be useful here to give you more info, but I suppose your method works well enough if all you want is the number of people who would participate.

17

I had forgotten about that!

EDIT: Although, as I tried to build a poll, I couldn’t figure out any other sensible options to put in it…

1 Like
18

I like this idea a lot. I think an ongoing league where players can drop in and out would be a great addition to the Yomi scene. And I think the idea that players just have to sign up, rather than have to pick opponents and so on, makes sense.

Personally I would prefer if the match format wasn’t very long. I think best of 3 and 5 are okay, but Bo7 is too long.

On this point:

I’m thinking that it will be hard to get 16 players regularly. For a one off, normal tournament 16 players would be no problem. But if this is going to be an ongoing thing and especially if players can come and go, then I could see the numbers dropping below 16 very often.

The virtual tournament concept sounds great to me, but is there a tournament format that would be more robust when dealing with lower numbers?

1 Like
19

I’m not sure. I think I’m going to start with it as described, and see how it goes, and make adjustments once we are what the issues are.

3 Likes
20

Come back, JonnyD! I miss losing to ye Jaina! :crying_cat_face:

1 Like