Rating discussion

[ufm]

In cooperative games, developers are responsible for rating system as the final score determines the rating gain/loss:

For cooperative games, by default players will earn as many rating points as the score (0 counts as a defeat).
But, assigning 0 to the losses prevents experienced players avoiding beginners to escape potential rating loss.

Assigning higher final scores for harder modes can compensate the player efforts and give some incentives for playing harder games.
In addition, assigning 1 to every win makes very hard to separate player level by rating, as 0 wins (the first play sets rating to 1) to 99 wins are all treated as the same category.

[Sarev]

Agree, I would love to have ELO system for this game expanded.

[ufm]

Agree, I would love to have ELO system for this game expanded.

If so, please upvote this suggestion as well: https://boardgamearena.com/bug?id=101116

[Kayvon]

At the time of writing, this is the most popular suggestion for Daybreak.

I'd be curious to hear how people think the scoring should be calculated across all variants of the game. No promises as to whether a change would be implemented, but knowing how to approach the scoring would make the problem more... approachable, I guess.

[ufm]

Applying multipliers based on the table options, probably:

- Player count (2 ~ 4)
- Used world powers
- Group challenge card types (+ card / - card / variety card)
- Individual challenge card types (+ card / - card) of each player

The values to be assigned are to be pretty much arbitrary though, as it's very hard to calculate 'appropriate amount'.
However, variety cards should be assigned with a multiplier less than 1.0 as they generally make the game easier (especially Our Own Plans).

Also, assigning 0 instead of a negative value for defeats also helps:

For cooperative games, by default players will earn as many rating points as the score (0 counts as a defeat).
But, assigning 0 to the losses prevents experienced players avoiding beginners to escape potential rating loss.

[Kayvon]

Those are good examples of the type of things that might matter point-wise. I guess I'm asking, specifically, how would you assign point values? The more standardized and simpler the system, the more likely it would be to get implemented.

As for the zero vs. negative, I've always used -1 to indicate a loss, 0 to indicate a tie (in games with an automa, for instance), and a positive value to indicate a win. Daybreak isn't likely to get a special case for it.

[ufm]

Those are good examples of the type of things that might matter point-wise. I guess I'm asking, specifically, how would you assign point values? The more standardized and simpler the system, the more likely it would be to get implemented.

For example, in Tranquility: the Ascent I implemented, the base score (2 + max(2, player count)) is multiplied by:
Green path: x 1.5
Goats: x 1.5
Panoramas (3 difficulty levels): x 1.5 / 2.0 / 2.5
Start anywhere (easier variant): x 0.5
Initial guide ropes (4 difficulty levels): x 1.0 (default) / 1.5 / 2.0 / 2.5

So, I'd propose:

Base score: (2 + max(2, player count)) x 2 (gave x2 to the base score due to easy world power + easy group card combination)
2p difficulty (easy/normal/hard/extreme): x 0.5 / 1.0 / 1.25 / 1.5
3p difficulty (easy/normal/hard): x 0.5 / 1.0 / 1.2 (setting 3p hard as 1.2 makes the base rating of 3p hard equivalent to 4p)
Group (easy/variety/hard): x 0.5 / 0.75 / 1.5
Individual (easy/hard): x (1.0 + (0.125 x (hard cards - easy cards)))

Again, I didn't calculate win rates while assigning these values, but I don't think it would matter much.

As for the zero vs. negative, I've always used -1 to indicate a loss, 0 to indicate a tie (in games with an automa, for instance), and a positive value to indicate a win. Daybreak isn't likely to get a special case for it.

The critical differences are:

Automa mode in multiplayer games
- Do not have any impact on ratings at all
- Ties are possible

Co-op games
- The final game score changes the rating of each player
- Ties are impossible (in almost all co-op games), you either win or lose as a team

[Jellby]

If there is no "elo" loss, any gain is pretty meaningless.

If you still want an always-increasing rating, I suggest a simple 10 points per win. It makes reaching the different skill levels (good, strong, expert...) possible in a reasonable time, for people who care about that (but they'd still be meaningless).

[ufm]

If there is no "elo" loss, any gain is pretty meaningless.

If you still want an always-increasing rating, I suggest a simple 10 points per win. It makes reaching the different skill levels (good, strong, expert...) possible in a reasonable time, for people who care about that (but they'd still be meaningless).

You said the same thing in 2 years ago, but co-op ratings have never been proper 'Elo' ratings: https://boardgamearena.com/forum/viewtopic.php?p=105659
Just look around co-op game ratings since then. ELO simulation simply doesn't work well as you expect.

[Kayvon]

Automa mode in multiplayer games
- Do not have any impact on ratings at all
- Ties are possible

Co-op games
- The final game score changes the rating of each player
- Ties are impossible (in almost all co-op games), you either win or lose as a team

I follow that. The most common "co-op" mode is solo mode, in both this and other games I've done. I've alway been consistent about a zero being a tie. While I hear you about ties being impossible in Daybreak, I'm still unlikely to make things inconsistent by carving out a special exception here.

Thanks for the scoring proposal. If your winning score would scale with the difficulty, would your losing score also scale? Or is it only upside for higher difficulty, but no real downside? It does feel pretty silly that your score only ever goes up, even if it has nothing to do with Elo.