New Campaign Proposal (Poll added)

[Necromancer]

from statistic point of view, basing statistic results on 1 balanced campaign means nothing.
you need at least few balanced campaigns to make a rule about balanced vs not balanced campaigns.

[mrBLUE9]

from statistic point of view, basing statistic results on 1 balanced campaign means nothing.
you need at least few balanced campaigns to make a rule about balanced vs not balanced campaigns.

I'm not basing it on 1 balanced campaign, I'm basing it on 3 unbalanced campaigns.

[Necromancer]

you are saying that a balanced campaign is about 50% win/loss after 4-weeks. one instance says nothing.
maybe 34% after 4 weeks would also result in a balanced campaign.

ELI5: i have 2 coins, one metal, one plastic. i threw the metal once, and it fell on heads. so it has 100% heads.
i flipped the plastic 6 times and it came out 33% heads. now i have a bunch of coins i can flip 4 times, and try and determine accurately if they are metal or plastic.
we don't know if the coins are even or not. either way you can't determine anything because the metal was only flipped once, and it says nothing about its characteristics. maybe it also has 33% heads, but the first time it fell on heads giving it 100% heads?

[mrBLUE9]

I'm not saying 50%, I'm saying higher than 35%. And yes, although unlikely, a 34% campaign can still be balanced after the 4 weeks. We can add an extra rule that if both armies are having fun and feel it's balanced, the campaign can then continue. I'm only trying to suggest something to alleviate the lack of fun that unbalanced campaigns have for one side. I know that if it's up to some members of a HC a campaign will go on till the very end, regardless of people in the army having fun or not. Which only contributes to GC losing more and more people. If I was in GC purely for gameplay and not the community, I would have left after C5.

[ZebraPeps]

...stuff...maybe 34% after 4 weeks would also result in a balanced campaign...stuff...

mrBLUE still have a valid point that, aside the statistical aspect of things, 4 battledays into the campaign can surely tell the signs of an imbalance campaign. A rule-asserted option of ending it immediately should be in place so the work of preparing next campaign can start asap, instead of dragging on for a few more battledays until the losing army decides to surrender.
This could also act as a motivator for each army to carefully organize & prepare for the campaign and avoid falling into the pit of "we hope we will get better during nextcoming battledays..."

[InsanityRocks]

random armies sounds like a terrible idea, reasons should be obvious.

I think it would be fair to the army leadership that they get to pick their armies and not have them picked for them.

IMO it best worked out when we did it the old way, each HC prepares alone who to draft, on the day of the draft 3 separate drafts are done - inf, armor, air. First pick determined by coin flip, each side pick one guy per turn untill draft list is empty.

How were the previous 3 drafts (BF4C1, BF3C6, BF3C5) done that were different from the 'old way' described? Admittedly I was involved, only, in the last one. From what you described, that's how that draft was performed.

I'm just suggesting Randomization as another method for selecting the Army. I don't expect it to be perfect, but then again, I don't believe the 'old way' is either.

...as for divisional selections... I didn't want to answer any and all questions/concerns. I was going for a straight numbers split, knowing one Army /may/ get more Armour or more Air. That's why I mentioned doing several shuffles until both HCs like the split or some end-event occurs.

I believe it would encourage creativity on the part of the HC to put together a plan of action for their Army (probably wishful thinking on my part).

And I'm in favor of creating some kind of 4-week rule that would end Campaigns with obvious imbalances.

[Bock]

How were the previous 3 drafts (BF4C1, BF3C6, BF3C5) done that were different from the 'old way' described?

BF3C5 was done by trying to engineer balance of the most highly skilled players and a couple of people who wished to play with each other. Then the rest of the players were picked one at a time. BF4C1 was done in a similar way, but with more attempts to accommodate people wanting to play with particular friends. BF4C1's draft also gave a lot more time and effort to get the "pre-draft" balancing of groups and high-skill players right. BF3C6 was done a bit differently. The two HCs basically sat down together and came up with pairings of players or groups that seemed balanced against each other, giving one player/group to one side and the other to the other. There was no back and forth picking of players.

Since we started trying to engineer balance in BF3C4, that first campaign was the only one that seemingly got it right, but I'd be willing to wager that that was mostly due to chance (and also to a much larger player pool).

When you try and engineer balance and it fails, a certain level of butthurt is know to occur. I'm with Hitman in that I think the best way is probably to just do the old-school draft.

[Bock]

I think a rule to end unbalanced campaigns at a certain time is unnecessary. It's up to an army and its HC to decide when it's no longer fun to play a given campaign and surrender.

[mrBLUE9]

I think a rule to end unbalanced campaigns at a certain time is unnecessary. It's up to an army and its HC to decide when it's no longer fun to play a given campaign and surrender.

Thing is, in the past that ending was always stretched. Both C5 and C6 should have ended sooner. HCs always like to prolong a campaign until the servers are half full.

EDIT: What we could do is set a mandatory poll after 4 weeks of each campaign, asking each army if they want to continue playing. I think that would be more fair, since it includes everyone's opinion.

ELI5: i have 2 coins, one metal, one plastic. i threw the metal once, and it fell on heads. so it has 100% heads.
i flipped the plastic 6 times and it came out 33% heads. now i have a bunch of coins i can flip 4 times, and try and determine accurately if they are metal or plastic.

That would be true if both coins were unrelated, they are not, if you have an unbalanced campaign it means you won't have a balanced one.

[A Docile Sloth]

ELI5: i have 2 coins, one metal, one plastic. i threw the metal once, and it fell on heads. so it has 100% heads.
i flipped the plastic 6 times and it came out 33% heads. now i have a bunch of coins i can flip 4 times, and try and determine accurately if they are metal or plastic.
we don't know if the coins are even or not. either way you can't determine anything because the metal was only flipped once, and it says nothing about its characteristics. maybe it also has 33% heads, but the first time it fell on heads giving it 100% heads?

You are correct, you cannot determine if the coin is metal. You can, however, determine if the coin is not plastic. If it is not 33% heads you know it is not plastic and therefore infer it is metal.

If the army gets ~ 30% win ratio, the campaign is unbalanced and therefore not balanced. mrBlue's assumption that 35% will be unbalanced is not bad in my eyes. It is unlikely that a campaign was at the end of the unbalanced distribution so 35% will probably be near the unbalanced-balanced crossover. Yes, a 34% campaign could end up being balanced but it is more likely that it won't be.

Things is, in the past that ending was always stretched. Both C5 and C6 should have ended sooner. HCs always like to prolong a campaign until the servers are half full.

I don't think the HCs wait until the servers are empty but wait until they feel there is no chance of winning. Which, admittedly, is when the servers are half full. It's the "we can turn it around next week" feeling that causes them to go on.

[mrBLUE9]

I don't think the HCs wait until the servers are empty but wait until they feel there is no chance of winning. Which, admittedly, is when the servers are half full. It's the "we can turn it around next week" feeling that causes them to go on.

I don't disagree at all with that! I know HCs have the best intentions in mind, specially since they are the ones that put the effort to organize everything, it's understandable that they will want to end the campaign only after trying everything they can. And I have no problem with that except when we are struggling with numbers, which is why I think we need to shorten any bad campaigns to the absolute minimum duration possible.

[Wi1D_K4rD]

You are correct, you cannot determine if the coin is metal. You can, however, determine if the coin is not plastic. If it is not 33% heads you know it is not plastic and therefore infer it is metal.

I may be wrong but assuming the coins are similar in size and depictions on each side with only real difference being the material being different. The probability of getting heads would be the exact same for both as the ratios of weights on each side of the coin would be the same. Which in case of most coins from what i understand is weighted just slightly in favor of heads. Very small but still. Plus having a 33% chance on a coin that "resets" its state every time it is flipped i.e. it doesn't remember its previous state last flip doesn't mean that after a set number of flips it has to be 33% heads.

The real message I would get out of this comparison is you can't use the results of one trial versus the results of multiple trials. So, now that I have provided nothing to the actual conversation involving balance as I have no meaningful suggestion I'm going to slink away now.

[elchino7]

land and repairing jets actually happened like 10 times a game in bf3
At some point Expandas came with the philosophy that its not worth your time, and that bailing over a backflag when your jet is almost dead is more efficient use of personell.

However not everyone agreed with that strategy, I know ID's strat usually is more conservative and just revolves around keeping air controll the entire game, in which case repairs happened a lot.


In bf4 im not entirely sure how the jet repairing fits in, because Ive hardly flown at all

Correct me if i´m wrong but since we played with 100% respawn time, jets spawn were on most maps at 40s±. Discount the 15s time to respawn and it wasnt too long the time you need to wait for a new jet.
I think its all about context and map. On a 5 or 7 flag big map, bailing to force a squad to go to a backflag is worthy.

On BF4 there is actually little to non repair places. Even worst, i think respawn time for air vehicles have increase.


Ontopic:

-I see how hard balancing armies and unless someone has a magic ball you wont know what will you be doing in the next 2-3 weeks.
But **I** feel that having a wider number of players should mitigate this. Depending on the same 2 o 3 guys flying/driving and an ace squad showing up shouldnt be so drastic.

-I see useful the "35% lose ratio" working if we have a smooth quick transition between campaigns. I think we all agree that no one wants to surrender before playing all it´s cards but sometimes its hard to draw the line between a definitive result and a possible comeback.

@MrBlue: this is just a suggestion (even if we dont have the numbers for previous campaigns).
** Instead of/adding to 100< ticket matches, why not %20± tickets left.
** What about looking into K/D or kill bleed in comparison to flag bleed.
(This is what i look for when pubbing and knowing if i should stick to a server without scramble).

_______________________________________________

IDEA:
1- We choose HC - Captains - SLs. Let´s try to balance this first.
2- "Day of the draft 3 separate drafts are done - inf, armor, air. First pick determined by coin flip, each side pick one guy per turn untill draft list is empty."
3- We play for 2 BD n BOs. If both teams agrees that teams are fine, we continue as usual.
4- If one of the teams feels like a "new draft" should be made on a certain division, we switch half team to the other side. Either by even or odd in order of choose (the first, third, fith or second, fourth...). Repeat step 3.
5- If neither of both teams agrees on keeping either of both team setups, we divide it on 3 different parts. Each divided in order of players being chosen. All this with the team setup which both HC agrees was the most balanced one. Then each team has the chance to swap with the equal number player of the other team. 1/2 swaps per team allowed on each category.


Notes: i think air is kind easier to balance and from then on armor and infantry. Only a not minor thing is that you have to consider how well some armor or air guys might play on heavy or pure infantry modes.

2cents

[mrBLUE9]

** Instead of/adding to 100< ticket matches, why not %20± tickets left.
** What about looking into K/D or kill bleed in comparison to flag bleed.

The sub-100 tickets rounds was a stat that I thought could be an indicative of balance, but it proved to not have a relation so far, specially since BF4 seems to be prone to large ticket difference rounds. So it's more of a trivia stat. I can add a sub-20 tickets rounds stat too, Chuko! Will do for next campaign.
Stats like K/D and flag bleed would need to be pulled out of the server directly, I think Cairdazar did that some time ago, but it seems awfully complicated. It would be easier if we had battle reports of each round.

[Divine-Sneaker]

Just a quick note.

If you want balancing to work, the people behind it need to actually know the people and their ability in the game for it to make any sense. From the limited amount of transparancy into the previous attempts, I can tell this hasn't always been the case.

Of course it's impossible to do with people completely new to the community, but that's another issue altogether. Either you try to get it as balanced as possible with the best available knowledge and considering known factors(known players), or you want as big of an influx of completely new and unknown people to randomly mix it up as much as possible.

Overpopulating seems to give the best result in my mind. Having people constantly rotate because of queues is good because it'll help balance out the player material. If you have an A team, in the spirit of the community you can't simply tell people who are waiting that there are better players currently playing.

Not only that, overpopulating also incentivices putting in an effort for each player personally. If you want to be ensured a spot, sign up! If you want to lessen the likelihood that you'll sit out, become a squad leader or specialize in a difficult niche role. It incentivices making a name for yourself in some way. Making yourself the guy always willing to rep the armor, the guy always willing to carry ammo or the guy who is willing to come up with mapstrats.

Additionally it helps alleviate attendance issues among the divisions, at least hopefully, which has been a huge imbalance point. As has already been mentioned by several people, attendance is fickle and rather impossible to predict. Personally I can never commit beforehand and all I can say is that I think I'll make it somewhat on time for a match. Planning anything further than that is impossible.

On a related topic, rating people on 283 different parameters with 2^4 other factors just doesn't work. No amount of data in the world will be able to quantify the overall ability of a player, no less in a setting like this. It's been tried and it's just pointless and either has to be kept practically secret and incredibly ambigous, or someone somewhere is gonna get butthurt as frak over some arbitrary number he had assigned. Let's just not.