Proposed Changes to 1v1 (2 player) games

[ndrm31]

In that scenario, would player 1 still attack and reinforce?

I like it the way that it is, at least in tourney finals u can sspli the pot

 

[AuraCraft]

@periwinkle; it's probly too confusing if there are different rules for different maps

 

[EasyToKill]

I have no problem changing rules as an OPTION. But I agree that having different rules on different maps can be confusing.

 

[Cardinalsrule]

I agree with Easy. Making rule changes as an option for the person starting the game is fine, but I'm 100% diametrically opposed to a blanket change of ANY kind.

 

[AuraCraft]

Any such experiment requires significant figures to base conclusions on.
How much is that? a (few) thousand games? We won't get such figures in a a few months like Sheriff has in mind "if the person starting the game" needs to choose to participate.
Even a fantastic new map like Land Rush couldn't persuade ppl to play more than 337 beta games.
So if a choice would be offered, perhaps the default setting could be deploy only in the 1st turn, would that be a satisfactory solution for you?

The 100 games trial that I did showed that if opponents were not aware of the setting, that would be in my favor if I played a deploy only 1st turn... so personally I would be in favor of a "blanket cover"

A lot of real time players Ive spoken to told me they really wanted something done; my solution, Sheriff's solution, anything. A dozen said they'd leave the site coz of the 61/39 ratio.

Only the first turn is affected it's not a big deal imho

 

[LotsOFFLuck]

Any such experiment requires significant figures to base conclusions on.
How much is that? a (few) thousand games? We won't get such figures in a a few months like Sheriff has in mind "if the person starting the game" needs to choose to participate.
Even a fantastic new map like Land Rush couldn't persuade ppl to play more than 337 beta games.
So if a choice would be offered, perhaps the default setting could be deploy only in the 1st turn, would that be a satisfactory solution for you?

The 100 games trial that I did showed that if opponents were not aware of the setting, that would be in my favor if I played a deploy only 1st turn... so personally I would be in favor of a "blanket cover"

A lot of real time players Ive spoken to told me they really wanted something done; my solution, Sheriff's solution, anything. A dozen said they'd leave the site coz of the 61/39 ratio.

Only the first turn is affected it's not a big deal imho

If a "choice were offered", as in you can make it a setting when setting up a game, I'd be in favor of that. But a forced, blanket change (as cards called it) is unnecessary and a terrible idea.

 

[Cardinalsrule]

Any such experiment requires significant figures to base conclusions on.
How much is that? a (few) thousand games? We won't get such figures in a a few months like Sheriff has in mind "if the person starting the game" needs to choose to participate.
Even a fantastic new map like Land Rush couldn't persuade ppl to play more than 337 beta games.
So if a choice would be offered, perhaps the default setting could be deploy only in the 1st turn, would that be a satisfactory solution for you?

The 100 games trial that I did showed that if opponents were not aware of the setting, that would be in my favor if I played a deploy only 1st turn... so personally I would be in favor of a "blanket cover"

A lot of real time players Ive spoken to told me they really wanted something done; my solution, Sheriff's solution, anything. A dozen said they'd leave the site coz of the 61/39 ratio.

Only the first turn is affected it's not a big deal imho

Default setting being first turn only deploys and can't attack doesn't work for me. As an OPTION that can be used in setting up the game, yes. But I'm against changing the default settings.

 

[haWD96lz]

Any such experiment requires significant figures to base conclusions on.
How much is that? a (few) thousand games? We won't get such figures in a a few months like Sheriff has in mind "if the person starting the game" needs to choose to participate.
Even a fantastic new map like Land Rush couldn't persuade ppl to play more than 337 beta games.
So if a choice would be offered, perhaps the default setting could be deploy only in the 1st turn, would that be a satisfactory solution for you?

The 100 games trial that I did showed that if opponents were not aware of the setting, that would be in my favor if I played a deploy only 1st turn... so personally I would be in favor of a "blanket cover"

A lot of real time players Ive spoken to told me they really wanted something done; my solution, Sheriff's solution, anything. A dozen said they'd leave the site coz of the 61/39 ratio.

Only the first turn is affected it's not a big deal imho

The way to figure out if a change made a difference is to run enough experiments. To understand that you can use the binomial distribution. To be simple we can assume a 50/50 proposition since its close enough to the numbers are are talking about. What we need is sigma in terms of a percentage for both the current data set sheriff has and any new data set that is created. 100 games gets you sigma = ~5%, 1000 games sigma = ~2% and 10,000 games gets sigma = 0.7%.

How many new setting games we need depends on how big an impact the settings changes have and the exact mean and sigma of the current data set. My guess from what sheriff has shown is the current data set has a mean of about 63% and probably have 10,000+ games. Lets assume sigma is 0.5 for it. So that means there is a 95% chance the real mean is 62-64%. There is a 99.7% chance its between 61.5 and 64.5%.

Now the new data set needs to have a low enough chance of its mean overlapping with the current mean. If the setting changes get us to a true mean of 55% first turn wins then probably 1000 games are more than enough since the range we would land in is likely 51-59% though 100 is not since the range is 45-65% and that overlaps with the other true mean, but if the true mean is 60% its going to take a ton of games to make sure the changes are real.

I did a few quick and dirty things here since I used a two direction distribution and we can probably use a one direction, which should lower the number of games needed somewhat. Happy to sharpen my pencil and help.

Though do note since each map has a different 1 v 1 chance it might be best to try any experiment on a single map, but it makes the stats harder.

 

[brianstheman]

Aura, you and I can probably do fifty 1v1s ourselves

That's 1/20th of Hawd's proposed 1000 game sample size.

 

[MNatt]

For what it's worth:
1. If you're going to change the rules, best to do it for all 2v2 games imo.
2. The idea of having a reduced number of reinforcements sounds the preferable option to me.
3. Can't you introduce it for a trial period - say a month. By the end of the month players would be able to form an informed view on whether they like it or not.

 

[AuraCraft]

@Brian; thanks for the offer
Can I save it until we can try my solution? I'd gladly play 100 in 2 months then
=
@Hawdz; Impressive reply. I forgot about that stuff after 2 decades How many games would you reckon would it take to see if Sheriff's solution would get us close enough to 50/50? Coz if after 200-400? 1st turn deploy only games (classic evolved) it would still be 58/42 instead of 60/40 we could safely assume it won't go from there to 50/50 and conclude it's time to try my solution.
=
@cardinals; that's exaclty what I meant
"A default, in computer science, refers to the preexisting value of a user-configurable setting that is assigned to a software application, computer program or device. Such settings are also called presets or factory presets, especially for electronic devices."
So there are 3 options
1 "blanket cover" as you call it
2 default setting = the experiment, in this case 1st turn deploy only, but a player who starts a game could alter the setting and choose the regular setting instead. If he doesn't change the default setting, it's first turn deploy only.
3 no default setting; a player who starts a game has to choose either regular OR first turn deploy only.
However I'm not sure how much additional effort it would be for the developers to implement a settings choice. Maybe a "blanket cover" requires less progamming effort compared to adding choices.

 

[clarkenfeld]

I'd be up for experimenting with Aura's suggestion. Not smart enough to engage deeply with the statistical ins and outs but something would be done. I now never play Euro massive duels - or Classic massive or Balkans - any more because the first player advantage makes it a silly contest.

 

[riskyone]

first turn in

I just wanted to add that if the first player advantage, or the second player advantage seams to high. An easy option would be to make that player wait till 5 cards on first turn in. Wow, neat. Sounds good.:stoned::dontknow::deal::smile::top::shot::listen::heeeelllllooooo::driver:

 

[riskyone]

I just wanted to add that if the first player advantage, or the second player advantage seams to high. An easy option would be to make that player wait till 5 cards on first turn in. Wow, neat. Sounds good.:stoned::dontknow::deal::smile::top::shot::listen::heeeelllllooooo::driver:

Could be implimented after any trial period. If there is one needed. If not don't try to rush it. Give peace a chance. I mean the new system. Peace doesn't belong here.

 

[AuraCraft]

good idea risky1
1st player can't cash early 3/4 card sets.
although if player 2 gets hot dice, why not?
so I'd add some conditions;

1st player can't cash early 3/4 card sets.
1 unless player 2 has more units than player 1

1st player can't cash early 3/4 card sets.
2 only if player 1 doesn't have a card disvantage (so if player one skipped a card, player two has more cards so the problem doesn't occur?

I say disable all 3 card sets

 

[Sebrim]

Yes - this sounds like this:

 

[Cardinalsrule]

I say disable all 3 card sets

Now that's just ludicrous.

I disagree with almost all of this stuff; I can't get behind anything except making anything but the 'regular' setting the default and others options a setting that has to be specifically chosen.

 

[periwinkle]

Now that's just ludicrous.

I disagree with almost all of this stuff; I can't get behind anything except making anything but the 'regular' setting the default and others options a setting that has to be specifically chosen.

I have to agree you. Everyone knows there is a 1st person advantage, then you are joining a 1v1 knowing you have a 50% chance of being first ..isn't that already fair enough?

Also...if the fix was to just have the 1st person deploy then the 2nd person would have the advantage of early sets. Let's change 1 thing at a time.

Thirdly...if past 1v1 or 2v2 tourney games used this "1st person just deploy" as a fix...dont we have some data we can look at? And if people were happy with this fix...let's go with it? I like simple solutions.

But in reality...you will never get it fair and equal.....nothing is fair...the drop, the cards, the bolds...that's life. Work with what you have. The most successful people are the ones who have grit and just stick it through.

Plus let's work on the new UI....I want to see traffic increase here...I'm ready for some new blood

My 2 cents :lollypop:

 

[haWD96lz]

@arua I would probably go with th map where the low end of the 3 sigma is highest for first player. This will give you the best chance of seeing if either technique matters. Happy to calculate them or remind you how if Sherrif will provide number of games with the percentages.

 

[NewSheriffInTown]

Thanks for the effort of producing the stats, it will certainly help me choose maps and settings when I create future tournaments, particularly 1vs1.

That's great! I'm glad that someone will be using these stats to better improve the fun for players, rather then using them better improve their chances of winning!

@arua I would probably go with th map where the low end of the 3 sigma is highest for first player. This will give you the best chance of seeing if either technique matters. Happy to calculate them or remind you how if Sherrif will provide number of games with the percentages.

Thanks, but that's a bit too complicated at the moment. We're gonna just use various examples to test.

I agree with Easy. Making rule changes as an option for the person starting the game is fine.

Making all these an option in the settings is the goal, and they will be options to choose from in the new engine....

But for the old engine, we would need someone to work on the old messy front end code, which we don't have, so it's an all or nothing scenario if we were to make any changes in the old engine....

Having said that. Marcos did create a hack, so some predetermined games can be played with the new rules, while not affecting the current rules. Call it a test run with 200 or so 1v1 games for people to choose to join if they want.

ty
I suggest that in a 2p classic evolved game, the player who has the first turn starts with 3 instead of 6 units, then the game continues as usual. (so p2 gets 6 in turn 1, then p1 gets 6 in turn 2, so does p2 etc)
so the player whose turn it is has the advantage, not the player who went first. that means on average, the advantages even out.

or on a smaller map, when the (terts / 3) = 4, then p1 starts with 2 units instead of 4.
on a larger map, when terts/3 = 8. then p1 starts with 4.

Yeah, so I just encountered the first issue when trying to test 12 Domains. Since the starting number of troops is already 3, it won't go below that. Same thing with Classic Mini and all the other small maps.

So I don't think there is a general solution, unless all I can think of is:

• Lowering the initial troop count so it can go below 3
• Maybe doubling the second player's troops deploy instead
• Or use different rules for different maps that combine some of the things we are trying (like skipping attacking on 12 Domains)