Stats by player count

[Salmur]

In most games, the stats are heavily influenced by player count.
Therefore, it often becomes meaningless looking at them without knowing the player count of the games.
My simple suggestion is to add a "player count" filter that allows you to see the stats referred only to the games of a specific player count.
For instance, if you select "3 players" you are able to see the stats of only the games played with 3 players.

[FrankJones]

Being able to filter by expansions would be a nice feature too.

[Kayvon]

This is a really cool suggestion. I'd love to see the capability of doing that. BGA actually has this data in their database already, but I'm not sure what the workload would be to implement the capability.

If you file an official suggestion, add the link to this thread and I'll vote for it. BGA developers don't act on suggestion thread discussions.

[FrankJones]

I think it depends on what the developer decides to include for any individual game.

For some of my games, it has number of turns and win percentage by player order. Other games do not. Some games have no stats.

[Kayvon]

Frank, you're misunderstanding the suggestion. I am a developer. As you said, I can choose which stats to include in my game. What I can't easily do is make stats that work contingent on player count without adding a slew of redundant stats. For instance, a "did 2nd player win" stat would need to be duplicated based on player count for it to work.

Meanwhile, the data already exists in the system to sort existing stats by number of players, but BGA would need to implement a feature to allow that data to be sorted as such. That's what the suggestion is.

[FrankJones]

I understand. I think.

Let's take a game that can be played with 2-4 players.

Some stats would not matter relative to play count. Those could just be listed.

Then, perhaps something like this:

2-player games:
player 1 win rate: x %
player 1 trigger end game rate: Y %
Average player 1 score: Z points
Average player 2 score: Q points.

3-player games:
player 1 win rate: x %
player 2 win rate: J %
player 1 trigger end game rate: Y %
player 1 trigger end game rate: G %
Average player 1 score: Z points
Average player 2 score: Q points.
Player 3 average score: R points.

4-player games:
player 1 win rate: x %
player 2 win rate: J %
player 3 win rate: P %
player 1 trigger end game rate: Y %
player 1 trigger end game rate: G %
player 1 trigger end game rate: W %
Average player 1 score: Z points
Average player 2 score: Q points.
Player 3 average score: R points.
Player 3 average score: T points.

Is that feasible and workable within the existing framework?

[Kayvon]

Possible? Yes. But it doesn't scale well. You'd need to make them player statistics and those are multiplied by the number of possible players, making a big list of statistics.

It's possible and I've done it for a couple games I've worked on (they are hidden statistics because most of them won't apply to the current game and players will get confused). It's not a great solution.

BGA already has this data in their database, however. They could produce these automatically if they dedicated the time to it. They may have bigger fish to fry.

[FrankJones]

Possible? Yes. But it doesn't scale well. You'd need to make them player statistics and those are multiplied by the number of possible players, making a big list of statistics.

It's possible and I've done it for a couple games I've worked on (they are hidden statistics because most of them won't apply to the current game and players will get confused). It's not a great solution.

BGA already has this data in their database, however. They could produce these automatically if they dedicated the time to it. They may have bigger fish to fry.

The person to ask would be the developer for the specific game right ?

For example, "Pente" already lists stats for player 1 wins versus player 2 wins.

So, for a game "Aquatica", I could ask the developer for that game?

[Kayvon]

The person to ask would be the developer for the specific game right ?

At the risk of repeating what I've already said in this thread... No. I am a developer and if someone were to ask I would refer them to this thread to explain why it should be a BGA feature. I'm not signing myself up for even more false bug reports when there's player confusion over a ton of unused stats that don't apply to each game.

[FrankJones]

The person to ask would be the developer for the specific game right ?

At the risk of repeating what I've already said in this thread... No. I am a developer and if someone were to ask I would refer them to this thread to explain why it should be a BGA feature. I'm not signing myself up for even more false bug reports when there's player confusion over a ton of unused stats that don't apply to each game.

I apologize if I am causing you to repeat yourself.

Maybe I can illustrate my point with a specific example:

If someone clicks on state for the game "Pente", dozens of stats are displayed, including:

First player win % (swap disabled) 52.94 100
Second player win % (swap disabled) 47.06 100

First player win % (no swap) 100 50 100
Second player win % (no swap) 50 100

Tentative first player win % (swap) 100 60 100
Tentative second player win % (swap) 100 40 100

First player win % (no swap in swap2) 100 49.43 100
Second player win % (no swap in swap2) 50.57 100

Tentative first player win % (single swap in swap2) 53.98 100
Tentative second player win % (single swap in swap2) 100 46.02 100

Tentative first player win % (double swap) 76.47 100
Tentative second player win % (double swap)


So, if this is a site-wide issue or limitation, I'm not sure I understand how one game, Pente. and possibly other games, have exactly the stats I am suggesting while other games do not.