Main Factions Balance

Discussion of all aspects of multiplayer development: unit balancing, map development, server development, and so forth.

Moderators: Forum Moderators, Developers

Post Reply
User avatar
ghype
Posts: 800
Joined: December 13th, 2016, 4:43 pm
Location: Berlin, Germany

Re: Main Factions Balance

Post by ghype » June 27th, 2019, 10:15 am

sergey wrote:
June 27th, 2019, 9:26 am
Expert opinions are much more important.
Exactly. It is just foolish to assume that just because of Cackfiend being the top of the Ladder that he is THE expert.
post sprites & stuff every day: ghype's Daily Art
destroying enemy castle here: Orcish Castle - Destroyed - work in progress

User avatar
sergey
Posts: 348
Joined: January 9th, 2015, 9:25 pm

Re: Main Factions Balance

Post by sergey » June 27th, 2019, 10:56 am

ghype wrote:
June 27th, 2019, 10:15 am
sergey wrote:
June 27th, 2019, 9:26 am
Expert opinions are much more important.
Exactly. It is just foolish to assume that just because of Cackfiend being the top of the Ladder that he is THE expert.
:hmm: Do you have a better way to determine if someone is an expert or not? Earlier in this thread both Hejnewar and Cackfiend have estimated an Elo rating which may be considered as an indicator of a skilled player. It is not good to separate players based on the Elo rating, such approach is harmful for the community. However, I think it is justified to assume that all players with high Elo are experts.

P.S. I find it funny that we are discussing who is an expert and who is not. Let's stop the flood?
Author of SP scenario Dragon Fight and SP campaign Captured by a Nightmare

User avatar
sergey
Posts: 348
Joined: January 9th, 2015, 9:25 pm

Re: Main Factions Balance

Post by sergey » June 28th, 2019, 8:57 am

Hardwood wrote:
June 24th, 2019, 10:09 pm
Cackfiend wrote:
June 23rd, 2019, 4:36 pm
I think the main reason UD have a lower win ratio is because of how unbalanced the matchup is vs Knalgans in Default Era
What is this claim based on? Such a claim should not be made without strong proof to substantiate it; the burden of proof lies with he who claims such imbalances exist. A large sample size of statistical data featuring strong play is required to demonstrate such a perceived imbalance. Said data should be further linked to detailed in-depth analysis. I have previously requested statistics and analysis in the Wesnoth Ladder Discord server, and I would like to repeat this request once more, if any conclusions are to be drawn about game balance.
In this topic Velensk (link) and Krogen (link) posted their opinions about Undead-Knalgans imbalance. Taking into account Cackfiend's point of view, we have at least 3 documented expert opinions. Thus, the Undead-Knalgans imbalance is an accepted fact.

Someone who wants to dispute the well-known fact must provide "a large sample size of statistical data featuring strong play" and "detailed in-depth analysis" as an evidence that the fact is incorrect.
Author of SP scenario Dragon Fight and SP campaign Captured by a Nightmare

User avatar
Krogen
Posts: 177
Joined: January 1st, 2013, 3:43 pm

Re: Main Factions Balance

Post by Krogen » June 29th, 2019, 1:50 am

That opinion of mine is 4 years old. Right now, i don't agree with myself at all in most points, matchup in question included.
I just started to discover high level 1v1 back then and i didn't have half the skill or knowledge i have now.
"A lion doesn't concern himself with the opinions of the sheep." - Tywin Lannister

Hardwood
Posts: 11
Joined: April 18th, 2019, 6:50 am

Re: Main Factions Balance

Post by Hardwood » June 29th, 2019, 5:08 am

sergey wrote:
June 28th, 2019, 8:57 am
we have at least 3 documented expert opinions. Thus, the Undead-Knalgans imbalance is an accepted fact.
This is a disingenuous argument.

1. There is no data in the thread to support the arguments made there. The arguments that were made consist chiefly of opinions.
2. A fact is an objectively correct statement about reality that can be verified and proven to be true. There may be an agreement (consensus) that a balance issue exists, among several players who may or may not be experts at the game, and they may even accept it to be a fact, but it has neither been demonstrated nor proven to be a fact in the thread you linked. There is not even a single replay to reference their opinions to actual in-game gameplay.
3. The posts that you have linked were made in 2015; the opinions they contain could be outdated. Standards of play may have changed since then, commonly held opinions may have changed since then, even the posters themselves could have changed their opinions; any number of things could have happened.
4. The criteria for "experts" has not been clearly defined, nor is it easy to establish a set of them that works reliably. Such a set of criteria is required for a statement regarding a consensus of experts - and then the players must be demonstrated to be experts. Having a high rating on ladder is insufficient as highly rated players are still capable of making big mistakes. For example, Cackfiend himself, despite being the most highly rated active player on the ladder, is still capable of making such big mistakes in his games - including conceptual mistakes that demonstrate a lack of understanding of certain matchups or maps.

Ultimately, analyzing replays tends to produce evaluations that may be more subjective than objective - to reach any conclusion that is absolutely objective, the only method to accomplish this would be to calculate all the possible branches of play in a game, decide on the absolutely best moves, and calculate the chances of winning from the very start, i.e. to solve a game of Wesnoth. This is unfeasible. However, analyzing replays and forming arguments based on said analysis constitutes a much more meaningful attempt to reach a truthful conclusion about balance than simply asking for opinions. Opinions that rest on the strength of players' reputations hold no water in any discussions that aim to reach such conclusions.
sergey wrote:
June 27th, 2019, 9:26 am
However, in one of his casts (Krogen vs Diggy) he said that Krogen's village grabbing with a leader looks like a rookie mistake. Krogen commented that it is a meta (common approach) for that map.
I later examined the map, and I will stand by what I said. I don't think I was mistaken. Even for a player with p2 chaotic faction, it should be possible to punish the village grabbing pattern. I think Diggy's attack would have been much stronger had he kept his leader around to recruit, and rushed on the other side, the side where Krogen's leader went to grab villages; there would have been two villages to attack instead of one.

Meta is not always optimal, nor is it complete. It grows and develops over time as players optimize their gameplay and find better solutions to issues that they face. A blind belief in the meta hinders its growth and causes it to stagnate.

You raise a valid point that there can be mistakes in analysis. As a matter of fact, I often make mistakes when talking in my casts. However, analysis provided in a running commentary with the screen controlled by a partner is very different from analysis that has been carefully considered after viewing a replay several times and corroborating with others. The accuracy of statements drawn from the second type of analysis is much higher. Discussions of balance involve the second type, rather than the first.

User avatar
sergey
Posts: 348
Joined: January 9th, 2015, 9:25 pm

Re: Main Factions Balance

Post by sergey » June 29th, 2019, 1:34 pm

Regarding my "disingenuous argument". This is not an argument, this is what is called a well-known or accepted fact. If there is no evidence that the statement is true or false, but most experts believe that this statement is true, then it is called an accepted fact.

You can watch a thousand replays, analyze them and say that you have a data that proves something. That data proves nothing. Another analyst could make different conclusion from that data and say that they also have data that proves their opinion.

Data analysis can't prove your opinion. Data analysis may result in hypothesis. When you have a hypothesis, the next step is to perform experiments that show that the hypothesis is always reproducible under the given conditions. If it is not always reproducible, it means that there is something you didn't take into account and your hypothesis is not proven. That is an essential characteristic of a proof - under the same conditions, given the same input, you always get the expected output.

Is that (practically) possible in regards to Wesnoth multiplayer? No. There are too many variable factors. Stop using the term "proof" please. Stop saying that with a tons of data and in-depth analysis you can prove something. The best you can do is to achieve a consensus of expert opinions.
Author of SP scenario Dragon Fight and SP campaign Captured by a Nightmare

Hardwood
Posts: 11
Joined: April 18th, 2019, 6:50 am

Re: Main Factions Balance

Post by Hardwood » June 29th, 2019, 3:21 pm

sergey wrote:
June 29th, 2019, 1:34 pm
Regarding my "disingenuous argument". This is not an argument, this is what is called a well-known or accepted fact. If there is no evidence that the statement is true or false, but most experts believe that this statement is true, then it is called an accepted fact.

You can watch a thousand replays, analyze them and say that you have a data that proves something. That data proves nothing. Another analyst could make different conclusion from that data and say that they also have data that proves their opinion.

Data analysis can't prove your opinion. Data analysis may result in hypothesis. When you have a hypothesis, the next step is to perform experiments that show that the hypothesis is always reproducible under the given conditions. If it is not always reproducible, it means that there is something you didn't take into account and your hypothesis is not proven. That is an essential characteristic of a proof - under the same conditions, given the same input, you always get the expected output.

Is that (practically) possible in regards to Wesnoth multiplayer? No. There are too many variable factors. Stop using the term "proof" please. Stop saying that with a tons of data and in-depth analysis you can prove something. The best you can do is to achieve a consensus of expert opinions.
You state that because three expert players shared the same opinion, it was an accepted fact. Common knowledge is not necessarily factual. Common knowledge changes over time. Three opinions do not make a fact. A statement is either a fact or not a fact. Whether it is accepted as a fact by some number of people, or not accepted as one, has no bearing on whether it actually is a fact or not. Even the basis of your statement - that the players were three experts - is unclear. To begin with, a consensus between three players is unrepresentative of "most experts". Krogen himself has posted in this thread that his ideas were outdated.

It may be unfeasible to arrive at provable, verifiable, factual conclusions about balance. It may be better for the purposes of the discussion to abandon the term.

What I have called for are arguments, supported by data, to back up any conclusions drawn from common knowledge. The link you showed lacks this. Simply taking common knowledge for granted can only lead to biased results, and could potentially waste a lot of players' time in the testing phase.

User avatar
sergey
Posts: 348
Joined: January 9th, 2015, 9:25 pm

Re: Main Factions Balance

Post by sergey » June 29th, 2019, 5:33 pm

I am not going to continue this discussion. The only thing I ask - stop using term "proof" improperly. Thanks.
Author of SP scenario Dragon Fight and SP campaign Captured by a Nightmare

Hardwood
Posts: 11
Joined: April 18th, 2019, 6:50 am

Re: Main Factions Balance

Post by Hardwood » June 29th, 2019, 8:41 pm

sergey wrote:
June 29th, 2019, 5:33 pm
The only thing I ask - stop using term "proof" improperly. Thanks.
I have agreed to abandon the term. I agree again. You're welcome, and thanks to you as well.

Mawmoocn
Posts: 95
Joined: March 16th, 2019, 3:54 pm

Re: Main Factions Balance

Post by Mawmoocn » June 30th, 2019, 3:37 pm

I’m reluctant to reply to your statement, as many days have passed.

I hope that you may address this issue, with prudence, in the future:
sergey wrote:
June 12th, 2019, 7:32 am
Unfortunately, he does not always understand what he is saying.
I was hoping for a professional argument, I wasn’t advocating for creation of a new AI, it was a direct reply to another person and, it was about reference on how AI was using a similar data structure to improve it’s battle results like Master data management, Master data and Reference data.

I was trying to create a potential framework using a script to analyse mass data and use excel to validate information.

I hope you tone down your assumptions and slightly aggressive response on replying to other people, in the possible future.

I don’t want you to use personal attacks to make your point, rather I prefer you use sound statements, to make your point across. In which, we can both agree or disagree, after making our own views with the valid arguments, that you may have. It helps both people on making discussions, on how issues can be potentially addressed.

If my statements are aggressive, I apologize, I try to improve my statements, whenever I can.

Thank you for your time.

name
Posts: 396
Joined: January 6th, 2008, 3:32 am

Re: Main Factions Balance

Post by name » July 1st, 2019, 1:48 pm

Krogen wrote:
June 29th, 2019, 1:50 am
That opinion of mine is 4 years old. Right now, i don't agree with myself at all in most points, matchup in question included.
I just started to discover high level 1v1 back then and i didn't have half the skill or knowledge i have now.
Presently, what possible balance issues do you see with default, if any?

As a top ten player, your opinion could be quite enlightening.

User avatar
sergey
Posts: 348
Joined: January 9th, 2015, 9:25 pm

Re: Main Factions Balance

Post by sergey » July 1st, 2019, 3:59 pm

I asked the same question here viewtopic.php?f=15&t=50007#p644076

Krogen said that he will reply later, because gathering his thoughts takes time that he doesn't have now.
Author of SP scenario Dragon Fight and SP campaign Captured by a Nightmare

User avatar
sergey
Posts: 348
Joined: January 9th, 2015, 9:25 pm

Re: Main Factions Balance

Post by sergey » July 2nd, 2019, 1:42 pm

I found a very interesting balancing topic viewtopic.php?f=15&t=44782
Many highly skilled players posted their thoughts there. Great thanks to ForPeace for creating that topic! It's so good to have all that information documented, considering that many of those top players are not currently active. Enjoy reading.

Update.
Another interesting and very useful thread, amir analyzed ladder games data and gathered statistics viewtopic.php?p=644418#p644414
Last edited by sergey on July 9th, 2019, 5:42 am, edited 1 time in total.
Author of SP scenario Dragon Fight and SP campaign Captured by a Nightmare

User avatar
Pentarctagon
Forum Administrator
Posts: 3953
Joined: March 22nd, 2009, 10:50 pm
Location: Earth (occasionally)

Re: Main Factions Balance

Post by Pentarctagon » July 6th, 2019, 9:53 pm

As far as analysing replays goes, this post and the next one might help make that easier, if anyone attempts to automate it.
99 little bugs in the code, 99 little bugs
take one down, patch it around
-2,147,483,648 little bugs in the code

Post Reply