Arena season 0: the Final Sprint !

[sprockitz]

Note: the turn-based games that are not finished at this precise moment may continue, but will no longer count for Arena mode.

I think there is an issue regarding turn-based games for the final sprint of the season.
Some players might stop playing their on-going turn-based games when they know that they have already lost, and that their clock won't turn negative before the end of the season. It can be quite frustrating for the player who is about to win.

I don't have any good solution to solve this problem, since it would be too long to wait for all turn-based games to be finished before officially ending the season. Maybe for next seasons we could at least disable starting new turn-based games in arena mode for the last few days of the season ?

this is certainly possible for abuse. I intentionally wound down all arena turn based games prior to the end of the season, which worked out as other players did not drag games on to prevent them from finishing. In a high luck game it would be fairly easy to abuse the system by finishing your wins and holding out on your losses at the end of the season. Hopefully the end of season checks will sniff out this sort of behavior (someone wins a season but you see they have 10 games all 1-2 moves from finishing as losses and they are disqualified).

This can be impossible in some cases to predict how long a game will actually take (some higher player count games can take longer than an entire season, even at a 2 moves per day speed. You also don't want to wait months for turn based games to all be finalized in order to get a final ranking. I believe the turn based games need to end at a similar time to the real time games for the competition to be fairest. I'd promote disallowing starting turn based games in the latter parts of a season. This would be complex though as it'd have to be a function of player count and average moves per game, and you'd want to disallow different base playing speeds at different points (for most game 1 move per 2 days shouldn't even be an option for most arena mode games...and for a lot of games 2 moves per day is a more realistic standard people need to meet to finish a game than the baseline 1 move per day). The other option is to go with the tournament style timer for all these games, or at least for games started in the last month...give each player 1/n of the remaining time in the season, where n is the games player count, as each player's time for the game and nobody gets any time added...popup a warning with they click play now in turn based mode that you will have x total amount of time to complete all moves, failure to do so will result in you losing the game.

[dschingis27]

The other option is to go with the tournament style timer for all these games, or at least for games started in the last month...give each player 1/n of the remaining time in the season, where n is the games player count, as each player's time for the game and nobody gets any time added...popup a warning with they click play now in turn based mode that you will have x total amount of time to complete all moves, failure to do so will result in you losing the game.

That sounds actually like a fair solution.

[Sunmaster14]

Two questions.

1. Can you consider the dividing of real-time and turn-base ladder? It is bad to combine these modes.
2. What about players that got a good rating and did not play after this?

About number 2. Why would they do anything about someone who got a good rating then didn't play anymore? They had to earn their rating so it is that person's right not to play anymore. If they aren't playing anymore that gives you the chance to surpass their rating and it prevents that player from getting a higher rating. They've already played enough during the season to get a good rating so suggesting they should be somehow punished for not playing enough at the end of the season is ridiculous.

One could argue that in games where level of play increases as time goes on, skill of a player who attained 500 ELO in 2020 could be higher that the one who attained 800 ELO in 2016. ELO can function without a decay parameter for games such as chess which are old and strategies are mature, but for newer games much less so. It is also important to know that many of the BGA games were previously (or are still currently) played by dedicated players in separate different platforms and it took some time for them to migrate to BGA, which might mean that some of the early ELO gains were attained against lower skill player base (I do not remember 100% but I think one of the assumptions needed for ELO to work correctly is for the distribution of the skill level of playerbase not to change significantly over time).

[Sunmaster14]

The other option is to go with the tournament style timer for all these games, or at least for games started in the last month...give each player 1/n of the remaining time in the season, where n is the games player count, as each player's time for the game and nobody gets any time added...popup a warning with they click play now in turn based mode that you will have x total amount of time to complete all moves, failure to do so will result in you losing the game.

That sounds actually like a fair solution.

+1, this would be fair and sounds workable.

[tibe]

Te season 1 starts inmediately finished this one?

[sprockitz]

Two questions.

1. Can you consider the dividing of real-time and turn-base ladder? It is bad to combine these modes.
2. What about players that got a good rating and did not play after this?

About number 2. Why would they do anything about someone who got a good rating then didn't play anymore? They had to earn their rating so it is that person's right not to play anymore. If they aren't playing anymore that gives you the chance to surpass their rating and it prevents that player from getting a higher rating. They've already played enough during the season to get a good rating so suggesting they should be somehow punished for not playing enough at the end of the season is ridiculous.

One could argue that in games where level of play increases as time goes on, skill of a player who attained 500 ELO in 2020 could be higher that the one who attained 800 ELO in 2016. ELO can function without a decay parameter for games such as chess which are old and strategies are mature, but for newer games much less so. It is also important to know that many of the BGA games were previously (or are still currently) played by dedicated players in separate different platforms and it took some time for them to migrate to BGA, which might mean that some of the early ELO gains were attained against lower skill player base (I do not remember 100% but I think one of the assumptions needed for ELO to work correctly is for the distribution of the skill level of playerbase not to change significantly over time).

Yes, this is true for Elo which persists for years, but for a single season the arena score had to be earned in the past 3 months, so I don't believe the same holds true for arena score.

And it definitely holds true even moreso for Elo score because they changed how the Elo scale by 200 points about 3 years ago so there are a lot of games that have people with higher Elo's than they deserve simply b/c of this...I have at least a handful of games I don't play much anymore where my Elo is still inflated because of this fact (Jump Gate, Bombay come to mind).

[Sunmaster14]

About number 2. Why would they do anything about someone who got a good rating then didn't play anymore? They had to earn their rating so it is that person's right not to play anymore. If they aren't playing anymore that gives you the chance to surpass their rating and it prevents that player from getting a higher rating. They've already played enough during the season to get a good rating so suggesting they should be somehow punished for not playing enough at the end of the season is ridiculous.

One could argue that in games where level of play increases as time goes on, skill of a player who attained 500 ELO in 2020 could be higher that the one who attained 800 ELO in 2016. ELO can function without a decay parameter for games such as chess which are old and strategies are mature, but for newer games much less so. It is also important to know that many of the BGA games were previously (or are still currently) played by dedicated players in separate different platforms and it took some time for them to migrate to BGA, which might mean that some of the early ELO gains were attained against lower skill player base (I do not remember 100% but I think one of the assumptions needed for ELO to work correctly is for the distribution of the skill level of playerbase not to change significantly over time).

Yes, this is true for Elo which persists for years, but for a single season the arena score had to be earned in the past 3 months, so I don't believe the same holds true for arena score.

And it definitely holds true even moreso for Elo score because they changed how the Elo scale by 200 points about 3 years ago so there are a lot of games that have people with higher Elo's than they deserve simply b/c of this...I have at least a handful of games I don't play much anymore where my Elo is still inflated because of this fact (Jump Gate, Bombay come to mind).

Thanks, I misunderstood the original question, I thought it was for the main ELO ranking in this site ("all-time"), not the Arena one. I agree that this is much smaller issue in Arena (if I understand correctly, Arena ELO gains/losses are calculated based on long term ELO (not "Arena ELO") - otherwise it would be problematic, as players of diverse skill levels will start next season in the same level). I came here quite recently and did not know about the historic ELO scale change - that then might also be a part of explanation why in Through the Ages (old version) I sometimes see previously unknown names popping up in 700+ ELO range who then fail to consistently win against sub-500 ELO players (surprising, as there is low variance in this game)!

[Rogue_Meeple]

Something to consider for future Arenas:

Allow for selection of expansions. I went to play Race for the Galaxy in Arena. To my surprise and dismay I ended up playing in a game with all the expansions, none of which I new.

If there was already a way to selectively turn on/off expansions that I am not aware, I apologize.

Thanks.

[sprockitz]

It seems as if I am missing a trophy for 4th in Clans of Caledonia. Not sure if others are missing trophies as well or why this is the case. It seems I have most (if not all) of my other arena mode trophies.

[cream]

Just want to make sure of this before Season 1: what is the tie breaker between players with exactly the same arena point at the end of the season? I know this doesn't happen so often but it did happen on Tokaido at season 0, both top players tied at the same arena point, and one is champion and another is vice champion. Is there a decimal arena point that is not shown?