Board Game Arena

poichich wrote:I know you don't see the problem I'm trying to explain. Someone who plays a 7min game monthly is not a gamer.

So what is your problem exactly?

To get the non gamers out of the top ranking.

The 'non-gamers' you talk about, by definition have reached the top rank only because they have played and won more consistently than everyone else. If you want to join them, then it's easy, you need to play as well as they can, or better if you want to go higher.

If you play 10 games and each of them play 10 games against equal opponents, then if they have a higher ELO then it's because they won more.

Just play more and win ELO points just like they did.

The ELO system:
https://en.wikipedia.org/wiki/Elo_rating_system

You will find no reference to the frequency of play.

I'd mention this suggestion reasonable:

>> I can see some benefits in reducing 60 days to get unranked to, say, 30

As author mentioned they use to Play elo off - we have had side discussion pointing to games with elo off are not easy noticeable at table list level as we'd be happy if they were.

When I forget to check, I sometimes again step on the same rake playing with high elo and at the end of game noticing elo was off.

30 days could be enough on long games, but not short. And if you take some long vacations my system to decrease it a little bit each day or each week depending on the duration of a game is better. You don't loose everything, you can loose a bit each days if you are out.

Ricardorix you forgot to say that the average level is not the same, because people improve at games. So the more we are oon a game the harder it is to reach the top than when the first good players that arrived there. It's not about talent it's about logic.

I'd like to confirm and complete what Een said about what BGA is planning now to handle this situation:

At first, we are totally aware of the situation. Some players are really "camping" at the top of some games leaderboards, taking the minimum risks, and this is not what we want for the BGA competition system.

Then, we want to keep the ELO as it is. ELO is a very good measurement of the strength of your opponents, and this is very important to know the strength of your opponents. So we don't want to reduce the ELO every X day or something.

The very simple consequence of what I just said is that we have to change the way the leaderboards are working. So we are going to change the leaderboards.

We don't intend to be particularly innovative here. All the popular online games has leaderboard and ladder systems and they are very similar between each other because ... there is no 10 ways to achieve this. So the BGA ladder system is going to look like what you can find elsewhere.

To avoid confusions, here's are some hints on the future BGA ladder:
_ It's going to be based on seasons. This way, once you manage to win a top trophy for a given season, you can keep it forever as an achievement.
_ BGA players on top of the current leaderboard will be rewarded for their positions, so this is a good idea to continue to play with the current system
_ Current leaderboard system (based on ELO) will be kept as it is (but BGA will push forward another competition instead of this one).
_ We are probably not going to find the perfect solution immediately. The good news is that because it will be seasonal, we will be able to slightly change things from one season to another (without breaking the competition in progress).
_ To win the most important trophy for a game for a season, you will need to play more than with the previous system. HOWEVER, the very good players that plays not so much will be able to get some other trophies too.
_ We won't impose a "number of games to play each X". However, we will make sure that the winners of the most important trophies of a season are part of the players who are the most active during this season, which makes sense.


About the planning: you probably think that the current situation is unsatisfactory for way too many time, and that we should have fix this problem a long time ago. And you are probably right. However, you should know that we never stopped thinking about this problem and best ways to fix it, and we had very good reasons not to address this problem before. We do not want to give a date here, but this is definitely in our plan to address this problem in 2019.

Thanks for playing on BGA and for your inputs on this subject.

sourisdudesert wrote: At first, we are totally aware of the situation. Some players are really "camping" at the top of some games leaderboards, taking the minimum risks, and this is not what we want for the BGA competition system.

That's my all point, thank you for telling this. Because with the previous posts of the others answering me, i felt like i was the only one to notice this point. Now that it is on the line, i can wait.

End of the topic

sourisdudesert wrote:

Is 'forced' matchmaking planned in lieu of the seasonal ranking?

N_Faker wrote: Is 'forced' matchmaking planned in lieu of the seasonal ranking?

Nope.

Note that it does not make sense anyway: for many games on BGA, the number of games played each day is not sufficient to have a real "matchmaking": you just take the first opponent available.

sourisdudesert wrote:
About the planning: you probably think that the current situation is unsatisfactory for way too many time, and that we should have fix this problem a long time ago. And you are probably right. However, you should know that we never stopped thinking about this problem and best ways to fix it, and we had very good reasons not to address this problem before. We do not want to give a date here, but this is definitely in our plan to address this problem in 2019.

Thanks for playing on BGA and for your inputs on this subject.

To keep it simple, can you just reduce the time to become unranked? Or if you want a little more sophisticated, then 5 games in the last month or something rather than just 1 game. As you say ELO scoring system is to remain the same, if you see the leaderboards as broken, then this is the only way to solve that.
---
Two other potential strong candidates for being broken, both coming from 'prestige (you asked for input)':

1. Prestige given for the 1st place trophy is very disproportionate. I could win 1 game go from 5th spot to 1st spot and get 45k presige. The difference from 1st to 2nd is 25k prestige. It would be better if these numbers were closer. Say 1st place has x2 the prestige of 5th place. I'm not sure why such high regard is given to 1st place over 2nd.

As for example my profile, I have 100k prestige in total, 50k comes from just 1 games' 1st place with 100 games played. Given I've played 5000 games with a similar degree of success, how does this compute?
Also some former higher ELO player that me comes out of retirement to play a single game and instantly I go into 2nd place and 25k is taken of my profile. Did I suddenly become a whole lot less prestigious?

2. Gaining large amounts of prestige from tournaments, and then permanently having high prestige to get to the top of the overall leaderboards.
I give you the remarkably talented keyflower player: https://en.boardgamearena.com/#!player?id=84324852. Currently #4 in the overall prestige standings.
Looking at other similar high overall prestige players, being good in just 1 game is not nearly enough.
As well, leaderboard prestige rankings can dry up, you can get knocked off the top and you need to play every so often. Tournament prestige is permanent.
Also did the 2 player private tournaments with large prestige get fixed?

3. Lastly ELO calculation in joint placement results can be broken. Lewis & Clark for example, someone can lose and still get additional ELO, making a 'free bet' scenario - just join tables with a few people of ELO points about 100 greater than yourself.
https://en.boardgamearena.com/#!table?table=49188417 (look at Alex_Prague ELO calculation).

There is also talk of a similar thing happening in 2x2 Innovation.