Ranked by Organ Robber

[graceful pier]

Welcome @bold warren! Any feedback from the community will show up here.
Please note that this thread discussion is not shared with SMG.
It is solely for the RISK Community to discuss ideas & provide constructive feedback!

[mint turret]

tldr

But I think this a long way of saying "skill based mm pls". Someone tell me if there's a concrete solution in there somewhere.

I'm all for skill based matchmaking, paired with an automatch feature.

Ah, I'm seeing he mentions that.
Definitely an opvote from me!

[timber mauve]

Correct if I'm wrong, but like in chess, Risk uses an Elo rating system, which means that if you win against beginners, you get less points than if you win against experts. Likewise, if you lose against beginners, you lose more points than against experts. So you could theoretically get to 2.6k Elo in chess.com by just playing against beginners, but that would just take you ages.

I guess one difference is that in chess.com, they automatch you with a similarly ranked player. Though, you can still chose to play with whomever you want (if they accept your challenge), which I think is a good feature to have. Similarly, in Risk you can filter your lobby based on rank. For example I have been filtering my EAPC games this season to expert+. It takes me around 5-10mins to fill the lobby.

Like you, I do want to see this game grow in popularity and in competitiveness (at the higher level). However, I am just not sure what is your specific suggestion of a feature to add to Risk, since filtering by rank is something that can be done, like I explained above. I'll happily continue the discussion, @bold warren.

[bold warren]

As a high expert with 15,500 lp, winning against 5 beginners in free for all would award you 791 lp. Whereas winning against 5 players also at 15,500 lp would award you 1,360 lp. If you get last against 5 beginners when you are at 15,500 lp you lose 1,929 points.

Your lp gain for beating 5 beginners only drops to 500 lp when you reach 25,900 lp. Against 5 beginners as a 25,900 lp player if u get last you lose 2,200 points.

The player with 25,900 LP should not still be able to get good lp gains and have similar max loss potential to a 15,500.

Beginners play in a very different way from expert and up players. Expert and up players tend to go for smart takedowns. In skilled only player lobbies players avoid conflict and all prioritize getting a bonus. Beginners will just select a player as a mortal enemy round 1 or they'll maniacally attack everyone. Pretty much none of the beginners stand a chance against you unless it's a talented risk player on a new account. Since beginners don't stand a chance VS a skilled player, the lp gains should not be so high and losses vs them should be more harmful. It's like putting a trained knight in a series of 1v1s against squires on their first day. I put 0 lp for the beginners lp, these are literally first time players and skilled players are somehow allowed to get tons of lp off them. Until beginners fight against beginners and skilled players fight against skilled players, the rank system will be horribly unbalanced.

Options:

1. Adjust LP system to be more sensitive to LP differences, so that high ranked players cannot make big lp gains VS beginners. If a expert player wins vs all beginners they should get around 250 lp. The issue is if expert players cannot play VS beginners and expert lobbies are still hard to find. Still something that should probably be done though.
2. Make a ranked system built around matching people with players of a similar ranking as I suggested in my original post. LP system wouldn't need adjusting if this system was adopted.

google sheet pinned in #risk_rank which I used for lp calculations:
https://docs.google.com/spreadsheets/d/1NBwtXu_DVJ_2LFbf6NJihaIqB7jUoV_4cZ5B7LUkRnw/edit?usp=sharing
@timber mauve

[timber mauve]

If option 1 is implemented, then it would automatically incentivize players who care about rank to play higher skilled lobbies. So they would self select an appropriate opponent rank for themselves. That is why I am not sure if option 2 is needed. Although, I am not against it being the default mechanism for joining lobbies (say with the automatch feature). For it to be the only mechanism on FFA and 1v1, I am also not against trying out, but I have not played games like LoL, that you mentioned, which have this second option implemented, so honestly don't know where is the active player base number cut off for this option to work.

However, I see option 1 having other unwanted side effects, like people playing more passively because they are scared to get 5th or 6th, which in turn might increase the frequency and or duration of stalemates. Especially if you play caps against high skilled players, it just becomes about who has more free time and less about skill.

Some people have suggested that you should also get bounty points in FFA, making the scoring system closer to what the tournaments have and decreasing the risk of stalemates. What would you think of that idea? So let's say you get 5 points for being 1st and 4 for being 2nd, etc, and for each kill you get +1. Then the total amount of points in a 6 player WD game are 5+4+3+2+1+0+5 = 20. Then you would sum up all the LPs of the players in that game and each one of them would get LPs based on the ratio of their score to 20 and the ratio of their LP to the overall LP sum.

[bold warren]

Either option 1 or 2 would make players play against people at their skill level more often. However, with option 2 what I'm talking about is the sorta automatch feature so high rank players can actually find lobbies in a reasonable time span. Risk on steam averages 2,000 players, this does not include the players from google play store of which there are over 10 million downloads and 341k reviews, I can't find downloads on apple store but it says risk has 171k reviews there. At any one time I see between 10-30 lobbies in ranked assuming I don't have any setting restrictions.

You mention a LoL like automatch system, The system of manually joining lobbies is intrinsically inefficient because at any one time: 15% of players are on the lobby screen searching for a lobby
20-40% of players are in lobbies waiting for them to fill up, many of these lobbies will never fill up and eventually get disbanded. meaning the players who waited entirely wasted their time.
65-45% are actually in quick lobbies which get into a game in less than 5 minutes. However, these are nearly always games with huge skill differences, like a master player VS a bunch of noobs.
As such the current system is DRASTICALLY slower. As mentioned in my initial post, a enforced automatch system for all ranked games would result in people getting quick lobbies with people at their skill level. This system would match people with players at their skill level, one rank below them or a rank above them (could only match them with people of their own skill level but that'd make finding games slower.)

Let's look at the 10-30 lobbies statistic from earlier. Assuming 80% of these are beginners and intermediates and in these lobbies there are an average of 3 players a lobby. That means if all the players went into automatch, 80% of those 30-90 people would find a game instantly. And since all the high rank players wouldn't be seperated into manual lobbies, that remaining 20% of the 30-90 people would quickly find a high ranking match. Most online games have an automatch feature of sorts, pretty much all battle royale games do. Automatch with risk would be easier because the games only take six players to fill up. LoL needs 10 players to fill a lobby. Fortnite and other battle royales need 100 players to fill a lobby.

data:
https://steamcharts.com/app/1128810#3m
https://play.google.com/store/apps/details?id=com.hasbro.riskbigscreen&hl=en_US&pli=1
https://apps.apple.com/us/app/risk-global-domination/id1051334048

[bold warren]

In your second paragraph you bring up a good point about if high rank players are in low rank lobbies they will be scared of getting a low placement. However, I'm already scared of getting a low placement in low rank lobbies because the LP losses are huge. Either way, expert and higher players shouldn't be getting matched with first time players and low rank players.

Last night my final game was on dino world (basically earth map but with 2 africa bonuses). I eliminated every player whilst black sat on their bonuses the whole game and ignored or didn't do what I asked the whole game I kept asking him to help. I got second place. Black was a master player.

You bring up a great idea with the bounty system which I think you should make as a separate suggestion on its own in #risk_ideas in case this LP system doesn't get passed. Currently a lot of high rank players are just passive players who sit there and accumulate troops, which definetly is not skill and it creates boring games. Your bounty system is truly amazing and I wish I had it in my game last night lol.