Jump to content
Sign in to follow this  
malladin.ben

Data Analysis - National Comparisons (or WTF?!?!)

Recommended Posts

I think there's been a growing recognition that this game is played differently by the different regions and nations, but it wasn't until I started to look at the data that I realised how stark the differences are.

I've tried to group the data together in tiers where the differences between guilds are small, and I've excluded Blacksmiths as there's really not much data available for them currently. I am also using the date of the last errata as the cut off point - there's been plenty of games played since this to enable patterns to start to show in the data.

Firstly, a quick definition of tiers, always a contentious issue, but if I define what I'm talking about you can probably see what I mean, whether you agree or disagree with 

Tier 0: Is usually reserved for a single guild that stands out above the rest of the pack, the rest of the meta is defined in response to it

Tier 1: are guilds that are strong and would expect to be seen competing at the top tables regularly

Tier 2: are guilds that a good player can compete with, but regular appearance on the top tables would be less common.

Tier 3: are guilds that you would not expect to be competitive at all and tournament wins from them would be rare indeed.

EU:

Tier 0 (56%+): Fishermen

Tier 1 (51-55%): Union, Engineers, Butchers

Tier 2: (47.5%-50%): Hunters, Morticians, Masons

Tier 3 (<47.5%): Alchemists, Farmers, Brewers

UK:

Tier 0 (58%+): Alchemists

Tier 1 (58-53%): Engineers, Farmers, Fishermen, Union

Tier 2: (48%-52%): Butchers, Morticians, Masons

Tier 3 (<48%): Hunters, Brewers

US:

Tier 0 (58%+): Masons, Union

Tier 1 (58-53%): FIshermen, Farmers, Alchemists

Tier 2: (48%-52%): Butchers, Morticians

Tier 3 (<48%): Engineers, Hunters, Brewers

 

So, it's a really crazy situation, where a team that dominates one meta can be tier 3 in other metas. If anything this just tells me that the game is balanced and it's the choices we make as players (or possibly the choices made by the best players) that determine the relative strength of the guilds over anything else.

I would be open to hearing from anyone who can make any sense of this beyond that.

Bonkers.

:wacko:

Share this post


Link to post
Share on other sites

Technical question - why the different cutoffs for EU compared to the rest? What happens if all the cutoffs are equalised?

Could it be partly down to a sort of “national gaming temperament” (now that’s a management buzzword phrase for the ages :P ), with the gaming background and community types skewing people’s playstyle a little. That might lead to a favouring of certain guilds as they fit the broader meta more, but I’m not convinced it explains the alchemists!

At least the morticians are consistent though :D

Share this post


Link to post
Share on other sites
25 minutes ago, Mako said:

Technical question - why the different cutoffs for EU compared to the rest? What happens if all the cutoffs are equalised?

Could it be partly down to a sort of “national gaming temperament” (now that’s a management buzzword phrase for the ages :P ), with the gaming background and community types skewing people’s playstyle a little. That might lead to a favouring of certain guilds as they fit the broader meta more, but I’m not convinced it explains the alchemists!

At least the morticians are consistent though :D

I understood very little of what you said. From what I took from it, it means Ozzies all play Hunters because they're from the "outback" and have all wrestled a crocodile at least once, right? 

Share this post


Link to post
Share on other sites

For me it only confirms two things:

Game is new, fresh and fluent in a manner that we are yet to unlock a full potentials in all guilds.

We all, have extremely good/bad rolls, that more often than not decide the fate of the game.

Share this post


Link to post
Share on other sites
22 minutes ago, ForestRambo said:

I understood very little of what you said. From what I took from it, it means Ozzies all play Hunters because they're from the "outback" and have all wrestled a crocodile at least once, right? 

Not exactly (though they do all wrestle crocodiles, right?) :D

More that there might be a preference for faster guilds, or fighting guilds, or guilds that do a lot of tricks, that shows up across an entire gaming zone/country. If the majority in the states prefer a certain type of game, while Europe prefers a different type (because that’s their previous experience and what their communities played), that could lead to an affinity for certain guilds as those fit the game style better. 

Share this post


Link to post
Share on other sites
11 minutes ago, Mako said:

Not exactly (though they do all wrestle crocodiles, right?) :D

More that there might be a preference for faster guilds, or fighting guilds, or guilds that do a lot of tricks, that shows up across an entire gaming zone/country. If the majority in the states prefer a certain type of game, while Europe prefers a different type (because that’s their previous experience and what their communities played), that could lead to an affinity for certain guilds as those fit the game style better. 

Ahh thank you for explaining in Donkey terms, i got you now! That makes sense, if the U.S was dominated by fast paced teams, those that suffer would fall short and vice versa could be true for the U.K

Share this post


Link to post
Share on other sites
2 hours ago, Mako said:

Technical question - why the different cutoffs for EU compared to the rest? What happens if all the cutoffs are equalised?

The different cut offs are based on where there is a significant gap between guild performances. It's all pretty arbitrary, really, just designed to show a sort of spread of where different guild sit in the different metas. If you  equalise them the EU looks like this:

EU:

Tier 0 (58%+):(none)

Tier 1 (53-57%): Fishermen, Union (despite the fact that there's a 5% gap between them and a <1% gap between Engineers and Union)

Tier 2: (48%-52%): Engineers, Butchers, Hunters, Morticians, Masons

Tier 3 (<48%): Alchemists, Farmers, Brewers

Share this post


Link to post
Share on other sites

Makes sense, I’ve just spent far too long reading and checking research papers to not ask the question!

Still doesn’t get the alchemists out of being weird :D

Share this post


Link to post
Share on other sites
3 hours ago, Mootaz said:

Yes, Alchemists and Engineers are the most puzzling of the lot.

I can kind of get engineers. They're (reportedly) good vs Fishermen and Alchemists, so where one of both of them does well, the Engineers stock goes up in response. I personally find them quite easy to deal with with Butchers, so perhaps when you have other fighty teams dominating, like Masons or Union, they struggle. But why Alchemists continue to dominate in the UK I don't know, especially seeing as a lot of the top players who were all over them before the errata seem to have dropped them.

Share this post


Link to post
Share on other sites

As somebody who grades a lot of papers with students analyzing data/documents, my question to Steamforged's development team based on this would be "what's going on with your Brewers, there?"

As the only guild that is consistently, worldwide, in Tier 3, an accounting for their standing needs to be made.  Now, that accounting might simply be "people aren't playing them" or "they have a really bad matchup in _____________", so I'm not necessarily arguing for changes/errata for them, but a data point that is that consistent definitely needs to be explained.

Share this post


Link to post
Share on other sites

I thought brewers and masons may lose a bit of grade by being in kick off, as newer players will be more likely to use them and only have the starting 6. But that probably doesn’t get the brewers out of 3 and into 2, and the American Masons players are ruining that theory a bit too, curse them :P

Has anyone done a grid of each guild’s win rate against each other guild since the errata? I do like a good heat map in data analysis :)

Share this post


Link to post
Share on other sites
36 minutes ago, Mako said:

I thought brewers and masons may lose a bit of grade by being in kick off, as newer players will be more likely to use them and only have the starting 6. But that probably doesn’t get the brewers out of 3 and into 2, and the American Masons players are ruining that theory a bit too, curse them :P

Has anyone done a grid of each guild’s win rate against each other guild since the errata? I do like a good heat map in data analysis :)

http://longshanks.org/guilds/?type=all&date_from=2017-08-01

If I remember correctly errata was in July so from 2017-08-01 will be good

Share this post


Link to post
Share on other sites

You know... the numbers could be fixed.... you just take all the really good players just play against a scrub like me... have them play with the teams that are perceived as weaker... and I'll play the teams that are perceived as stronger. The really good players are still gonna to crush me unmercifully into the pitch... and then it all evens out. :P 

Now don't get me wrong... all games can use a tweak here and there and constantly battle for balance....but it seems to me like the best players like to win... and therefore gravitate toward the "better" teams... and that puts a slant on statistics that needs to be considered.

Which brings me to a question... how good are the opponents where you play locally. Are you the best player of your bunch? I struggle to determine my own strength as a player... many times doubting myself... but am I better than I seem to be because I am simply playing better players than I. I feel that is the case and that I get better because of strong matches with very good players... but is that just me trying to stay positive while never quite getting to the victory column?

Such an amazingly good game.

Share this post


Link to post
Share on other sites

Well, as someone who has played a lot of Brewers vs Blacksmiths and Brewers vs Farmers:

The Farmers  can heal-tank (healing a fairly large number of boxes on their models per turn), while dealing more consistent damage (and AoE knockdowns), and having plenty of buffs for their own people. The Blacksmiths can armor-tank (a lot of their models have high ARM) not to mention the debuffs, the AoE Rowdy on everyone (will hurt other teams more than it does them). Both of them also seem to either be faster or have more consistent movement tricks.

The Brewers don't stack up well to either of them, and feel particularly poorly suited for this new meta. The Brewer's still do stuff (and are definitely pieces in the game), but the Blacksmiths and Farmers seem to do the Brewer's "job" better.

So no, the Brewer's placement as dead last everywhere doesn't surprise me all that much.

Share this post


Link to post
Share on other sites

Brewers just don't have answers like what a lot of other teams do.  It felt like originally they could adapt and do whatever the other team wasn't good at (E.G., Kick against Butchers, fight against Alchemists) but when kicking got so popular under Shark and Midas, the Brewers really just struggled to take out faster than the other team scored.

Siren1,  Shark, Midas all being nerfed helped, but at the end of the day, it still feels like the Brewers always start on the back foot.  They want to buff players up and that requires order of activation issues and heroics.  GIC helped some, and that has yet to settle out.  I think the biggest kicker was Tapper lost the ability to close a model out on his own with the change to Old Jakes.  Now, you want CA up early, but it's hard to get it up, and kill the model you want to kill first activation.  The rest of the team hinges on that CA.  Esters has more answers overall but I think more than anything, no one knows what the Brewers gameplan should be in many cases.

Share this post


Link to post
Share on other sites

I don't consider myself a great player, but I do have a winning record with my Brewers and have even won a small tournament with them. I do agree that they don't have a sharp "gimmick" like other teams. However, I'm rather surprised to see them ranking in the bottom percentage everywhere in the world.

Share this post


Link to post
Share on other sites

This is actually rather natural in all games, name a game and if it balanced correctly you will see tier lists like this. If all the areas have the exact same tier list then something is actual unbalanced. For example, the Brewers, but hey, no one can balance something perfectly. 

It depends on how the culture of the area defines the meta, I used to spend a lot of time watching DOTA 2 esports, and I picked up on how different the game styles were. US loved their team fights, China tried to create a powerful carry, Russia is aggressive and ratty, and South Korea was strange. These different game styles affect how a game is played in each region and what characters are more powerful. As well, Guild Ball does not have a community that crafts theories and strategy like many esports, preventing homogenization. 

Share this post


Link to post
Share on other sites
6 hours ago, Wildcard said:

I don't consider myself a great player, but I do have a winning record with my Brewers and have even won a small tournament with them. I do agree that they don't have a sharp "gimmick" like other teams. However, I'm rather surprised to see them ranking in the bottom percentage everywhere in the world.

This is totally dependent on the meta. Within my own local meta, the players with the best and the worst track record are both usually playing Brewers. Might be important to say that the meta is basically Brewers, Butchers, a Masons player and a Fish player.

The 100% best way of getting perfect statistics would be for a highly skilled player to play against a clone of himself, and having at least 10 matches per match-up played. Team match-up is one thing, but skill match-up is usually more important in my experience.

Share this post


Link to post
Share on other sites

Brewers are a decent team.

Other teams do what they do slightly easier but they definitely have game.

I've played 2 tournaments with Brewers came 2nd in both 1 to another Brewer player and 1 to a very good butcher player so ive won 5 out of 7 games played. Also played 1 tournament with each captain so think both are viable.

I played against one of my buddies who plays engineers and he was surprised how quick my team was as I mostly used spigot for tooled up and times called then buffed speed with esters too. 

I think Brewers can be succesful they do just have one or two really bad matchups so the most competetive players have stayed away from them meaning their win% is lower than the rest.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×