Start every second time 2-player games

[Lotus Blossom]

-1

This is extra information to keep track of that provides little benefit beyond the random start. The number of data points to keep track of increases close to the square of the number of players, so it could grow quite large for popular games. And, while the history of matches between players is a searchable function already, note how long that search takes; it ties up a lot of resources.

Additionally, functionality like this could be used to skew tournaments. If there is a tournament coming up, a player can track down the participants, and play one or two games against them beforehand to get the preferred starting position.

Then, there are the ELO boosters. Knowing which games would be starting in the favorable and unfavorable positions, they could play games in the favorable position with ELO on, and switch it off for the next game.

I disagree that there is little benefit. As OP stated, in some games there is an advantage in starting first. This is why the suggestion seems popular. Whether or not the benefit is worth it for the extra resources required to retain this information is another matter.

As for playing the system for tournament or ELO advantage, I can't help but laugh at your suggestions. For that to even work you'd have to get your specific opponents to agree to play with you in those circumstances in the first place. Highly improbable in my opinion.

[ollyfish2002]

+1
IMHO, BGA needs to keep tracks only for last game which is already stored in the history for each game. You can replay the game and view who starts. Therefore, new game between 2 players can start not randomly quite easily.
my 2p
EDIT : I even think this could be displayed for each game and added to the statistics : wins starting first 75%, wins starting second 10%, wins starting third and so on. This will rule out (or show) the bias in some games.

[N_Faker]

If it is feasibly possible, why not.

The idea, as proposed, seems far too prone to bugs and unintended consequences.
Maybe tie it in with the "Rematch" function to prevent the 1st player from being 1st again.

[Liallan]

-1

This is extra information to keep track of that provides little benefit beyond the random start. The number of data points to keep track of increases close to the square of the number of players, so it could grow quite large for popular games. And, while the history of matches between players is a searchable function already, note how long that search takes; it ties up a lot of resources.

Additionally, functionality like this could be used to skew tournaments. If there is a tournament coming up, a player can track down the participants, and play one or two games against them beforehand to get the preferred starting position.

Then, there are the ELO boosters. Knowing which games would be starting in the favorable and unfavorable positions, they could play games in the favorable position with ELO on, and switch it off for the next game.

You write against anything others post here. It's a pattern. You do it all the time. What is your problem???

Perhaps he simply likes to think through things, how difficult it might be to implement, whether it could end up causing problems, etc. Lots of people present what they think are cool or fun ideas, which might very well be good and useful ideas, but without giving any thought to the fact that someone has to code it, it takes time, takes resources, as though one just has to snap one's fingers to get what they want and it's done. Sometimes we need the voice of reason, which is often times unpopular but needed. (Not necessarily meaning I agree with everything he said. I'm just trying to make a point.)

The idea, as proposed, seems far too prone to bugs and unintended consequences.
Maybe tie it in with the "Rematch" function to prevent the 1st player from being 1st again.

That actually sounds like a reasonable way to handle it.

[CB Droege]

The idea, as proposed, seems far too prone to bugs and unintended consequences.
Maybe tie it in with the "Rematch" function to prevent the 1st player from being 1st again.

That actually sounds like a reasonable way to handle it.

I agree. If implemented, it should only be tied to the rematch button, so that people who play right in a row can get this feature. When I sit down to play a game with a friend, and we only play twice, we each get to start once. That makes sense.

Otherwise, just letting the system flip a coin each time is easier and just as good. When I sit down to play with a friend for just one game, we don't try to remember who went first when we played last 6 months ago. We just flip a coin fresh.

[Daggerheart]

I play in elimination tournaments and several times end up facing the same good players vs the end. Played only three games vs a guy (skilled player) and he had the benefit ALL three times. It was the reason for making the post in the first place. Still even a "rematch button with alternate start" would be a big improvement from today.

I still think when the same players meet in the same game the start should alternate all time in 2 player games...

[Jest Phulin]

I play in elimination tournaments and several times ...

Played only three games vs a guy (skilled player) and he had the benefit ALL three times.

So, several is only three? He had the advantage one more time than random chance might indicate.
Also, it seems strange. You criticized me for suggesting that this could affect tournaments, and now you say that you want it in place to affect tournaments?

As to the "switch starting players on rematch" suggestion, I think that this does have merit. However, just as immediate access to final game state is only saved for a limited time, defining start player should also only be kept for a limited time.

[Cedroc92]

+1

[CraWleR]

+1 to the original suggestion, however I also feel like this is only really relevant in Competitions/Tournaments and should be maintained per Competition/Tournament basis. Also, some games actually have their own mechanics to decide who goes first. Quantum is a good example. Therefore - definitely wouldn't apply to all games.

So, several is only three? He had the advantage one more time than random chance might indicate.
Also, it seems strange. You criticized me for suggesting that this could affect tournaments, and now you say that you want it in place to affect tournaments?

Did you read who actually criticized your suggestion? Says it's "Lotus Blossom", not "Daggerheart", so I'd recommend apologizing, because you look silly right now.

Also, if it's a 50/50 chance that someone will go first - going first 3/3 times is 12.5 % probability. In other words, he had the advantage 1.5 times more than random chance might indicate. Rounding up or down by so much to make an argument is a bit weird.

Overall - it's not rocket science. They're using MySQL anyway, so assuming the databases are setup properly, you can get an answer with 1 or 2 queries. Bigger problem could be making sure it would be used by games, however if someone came up with a list of games, I estimate 3-4 hours of work + X number of hours to change the games in that list.

Sadly, the site has bigger problems, so this would probably go into the lowest priority queue. Unfortunately as well, the more games they keep adding, the more time they would have to invest to do this sort of thing in the future.

[N_Faker]

Also, some games actually have their own mechanics to decide who goes first. Quantum is a good example.

The 1st player in Quantum isn't random?