Is the Crossbow upgrade too cheap?

[AvihooI]

Yeah I said the same thing about melee pathing. Fixing melee pathing has nothing to do with the archer line.

[MatCauthon3]

@AvihooI

The bottom 5 are more stark. All civs can utilize archers to some extent. I don’t see how any of this is evidence that archers are OP or that crossbow upgrade is too cheap, seriously. What I’m trying to show is that the evidence we do have suggests otherwise:

the bottom 5 include 4 civs that focus on navy, meaning their land is weaker and one that focuses on siege weapons.
they also include 2 civs that many agree need some love (koreans and portuguese).
they also include not a single civ who has been played more then 90 games.

@AvihooI

• Archer civs do worse right now.

they aren’t though. you don’t have enough data to say they do worse and i just showed you that 5 of the top 10 civs make extensive use of archers, to the point you could consider them the core of their army.
the top 4 played civs are all civs that make EXTREME use of archers in their army.

@AvihooI

1. No one in the competitive community complained about archers or the archer upgrade cost being too cheap.

i agree with you.

@AvihooI

I’d really like the OP to explain what other than his personal anecdotal and limited experience has led him to think archer upgrade is cheap and OP.

the op didn’t say that. what the op said was it was brought up in another thread. i brought it up myself in the other thread to prove a point. the OP had argued in the other thread that upgrades that costed more meant that they were more useful then other upgrades. and i had pointed out that this wasn’t true, because crossbow upgrade was cheaper then long swords upgrade (both from the same age, but xbow is more useful, cheaper, and faster). and that’s what prompted him to bring it up here.

@AvihooI

Also, if you think the entire population size of matches for 2 weeks (even though the site also takes into account previous patches and you can also similarly look up stats from before) is insufficient then please explain what makes it insufficient. Would 4 weeks suffice? 4 months? 20 years?

honestly? i feel the dev team should wait 2-3 months between balance patches. what i do know is this. i have never seen any game make balance changes based on two weeks data unless it was quite obvious something was wrong (see Cumans at launch, everyone knew they were going to be broken).

@AvihooI

Truth of the matter is there’s no real indication of archers being OP and plenty suggesting the opposite.

i don’t think many people believe archers are op or that the xbow upgrade is to cheap. the problem with archers right now isn’t even archers. its melee path-finding.

@AvihooI

Yeah I said the same thing about melee pathing. Fixing melee pathing has nothing to do with the archer line.

except if you fix melee pathing, they can better deal with archers because they don’t bug out, and thus archers aren’t nearly as efficient.

[AvihooI]

What I meant is that they shouldn’t make the archer upgrade more expensive simply because melee pathing is broken right now… they should simply fix melee pathing.

As for the longsword upgrade for comparison… archers are a lot more expensive than the militia line (over twice the gold cost ) while also requiring a critical mass to be useful. There’s more gold going down the drain with archers. In fact, what makes infantry so potent (especially post supplies) is the ability to make a ton of them while booming for farms and going for moderate gold income. Unlike archers, infantry can actually bring an entire base down. Archers give you map control, but infantry can potentially close out early.

Comparing just the upgrade costs is simply not a fair analysis.

[MatCauthon3]

@AvihooI

What I meant is that they shouldn’t make the archer upgrade more expensive simply because melee pathing is broken right now… they should simply fix melee pathing.

and i agree with you their.

@AvihooI

As for the longsword upgrade for comparison… archers are a lot more expensive than the militia line (over twice the gold cost ) while also requiring a critical mass to be useful.

except i can make archers while still saving food for workers and teching up to the next age.

@AvihooI

In fact, what makes infantry so potent (especially post supplies) is the ability to make a ton of them while booming for farms and going for moderate gold income.

the militia line is weak, and terrible for trying to boom with because the food cost of the militia line takes away from food you could be using to make workers, research eco upgrades, and teching to the next age.

[WoodsierCorn696]

@AvihooI

1. No one in the competitive community complained about archers or the archer upgrade cost being too cheap.

I give you the example of Hera. Probably most high level players agree with this. Ask most high level players and they will tell you the meta changed to more archer focused thanks to DE.

@AvihooI

Yeah I said the same thing about melee pathing. Fixing melee pathing has nothing to do with the archer line.

Fixing melee pathing is an indirect nerf to ranged units. So yes, it has something to do with the archer line.

@AvihooI

Also, if you think the entire population size of matches for 2 weeks (even though the site also takes into account previous patches and you can also similarly look up stats from before) is insufficient then please explain what makes it insufficient. Would 4 weeks suffice? 4 months? 20 years? Why yes, why not? Have you done any statistical analysis on the data to determine what the right sample size (mind you, this is not even a sample, but the entire data set for recent matches) should be?

There are many statistical method to determine if the number of observations is big enough. I think about confidence intervals for example.

One of the questions to ask is for example: Do we need to have a look at all games? Or do we only have a look at high level players? Balancing civs is mostly around high level players. So you can only have a look at for example the 1650+ category. This will greatly decrease the number of observations you can use.

[AvihooI]

Confidence intervals are for determining whether or not a sample size is likely to be representative of the population. However, what we’re dealing with here is not a sample but the entire population of recent ranked matches (and you can find more of the same for HD and Voobly historically speaking).

So, I would argue that by definition when you see non-archer focused civs being top 5 whereas archer civs being bottom 5 it literally means, considering the current meta (which is basically how players play the game at this very moment), that archer civs and by extension, archers, are not OP.

This of course, does not account for players learning a new meta and adapting to it. But if that is the case then I should remind you the game is quite old and the players are quite experienced. It’s not a new game where we should expect massive learning curve differences.

If the meta was set to be archer-favored then we’d be seeing archer civs both being picked and having a high win-rate. This is simply not the case. I’d say there’s a pretty nice balance which right now, despite melee pathing problems, favors non-archer civs.

If you want to see real pain in the ■■■ power spikes then consider Spanish Conquistadors or Burmese Arambai (and to some extent Camel Archers and Mangudai). Now those are a real pain to deal with in many maps. With crossbows I know I can deal with a bunch of towers or some easy counter units + mangonels. This is not the case with either Arambai or Conqs.

I can’t help but think that the OP simply had a couple of bad games involving archers defeating him (perhaps as they were upgraded) and made his conclusion based off of that.

But again, I stress… the good old rule of “if you don’t like it - mod it” applies.

[Parthnan]

@AvihooI

If the meta was set to be archer-favored then we’d be seeing archer civs both being picked and having a high win-rate.

You’re again confusing the prevalence of the Archer unit-line in the metagame with predicted prevalence of ‘archer civilizations’.

The symptom of the archer problem is not that ‘Archer civs are on the top’, but that ‘ALL civs, not just Archer civs are relying heavily on only the Archer line, especially in Feudal and Castle Age’

For god’s sake, this thread is about balancing the Archer , Crossbowman and Arbalester line , not about balancing civilizations that are supposed to be specialized in archery.

[AvihooI]

And yet you’ve shown no real indication whatsoever that the archer line is OP. The closest we have is looking at the civs. I’d at least expect a correlation there.

Sure, making archers and upgrading them to crossbowmen is something almost all civs can do - but you’re far less inclined to do so if you lack bracer, arbalester, thumb ring and ring archer armor (or even leather archer armor - Burmese). If that is your original game plan then you might as well go for an archer civ to begin with. This is simply not happening.

Show any real data or analysis other than your gut feeling and your recent game experience and I might take you seriously. If we’re just going for personal anecdotal experience then I have no problem dealing with archers nor do I see myself making them often. Here, opinion stated, we can move on.

[Parthnan]

@AvihooI

The closest we have is looking at the civs. I’d at least expect a correlation there.

You will not find a correlation there, because there are so many hundreds of other variables and factors, when it comes to civ winrates.

And no, civilization winrate data is not the closest you can get to discussing the Archer, Crossbowman, Arbalester line.

If you make civilization winrates the basis of your arguments, then you will end up balancing civilizations that are supposed to be specialized in archery with other civilizations but NOT the Archer unit-line with other unit-lines, and the LATTER is the goal of this thread, NOT the former.

[AvihooI]

Ok there’s no correlation between the OPness of the archer line and the winrate of archer civs

I’m done, seriously.

[Parthnan]

@AvihooI

Ok there’s no correlation between the OPness of the archer line and the winrate of archer civs

There are so many hundreds of other variables and factors , when it comes to civ winrates.
To extract a multivariate correlation on this complexity is not as simple as you think. No Human eye can spot Multivariate correlation above 4-5 variables let alone hundreds of in-game choice variables.

And that’s before you consider that this is just 2 weeks of data with a civ-matchup graph that is barely filled(not even 3 matches per possible 800+ civ combinations in 1v1), but I digress.

And then there is random noise, arising from player personalities, player perceptions etcetera.

[MatCauthon3]

@AvihooI

If the meta was set to be archer-favored then we’d be seeing archer civs both being picked and having a high win-rate

the meta most assuredly favors archer units. that’s why 50% of the current top 10 use them extensively. and if you think archers aren’t meta, go watch NAC3 and HC3 and tell us what you see?

yeah celts, vikings, japanese, and aztecs get picked a bit, but do they focus on infantry units? no. they go xbows into arbs (or water).

[AvihooI]

Where’s YOUR data suggesting the crossbow upgrade is cheap other than YOUR personality, perception, experience?

Oh right, for the dozenth time in this thread none provided.

[Player965976697]

@AvihooI

As for the longsword upgrade for comparison… archers are a lot more expensive than the militia line (over twice the gold cost ) while also requiring a critical mass to be useful. There’s more gold going down the drain with archers. In fact, what makes infantry so potent (especially post supplies) is the ability to make a ton of them while booming for farms and going for moderate gold income. Unlike archers, infantry can actually bring an entire base down. Archers give you map control, but infantry can potentially close out early.

Havent seen someone being more wrong lately. Your first posts make it seem like your a higher level player that understands the game. 24 how wrong I was in that assumption

[Parthnan]

@AvihooI

Where’s YOUR data suggesting the crossbow upgrade is cheap other than YOUR personality, perception, experience?

Oh right, for the dozenth time in this thread none provided.

Look, I have always added the ‘I think’ / ‘I believe’ / ‘It seems’ / ‘In my opinion’ to every comment of mine on the Crossbows. So yes, I do not claim my opinions are objective. They ARE subjective.

Now let’s return to constructive discussion on the topic, shall we?

[Player965976697]

Where’s YOUR data suggesting the crossbow upgrade is cheap other than YOUR personality, perception, experience?

4->5 range, 4->5 attack, 30-35 HP, 80->85% accuracy. For 200ressources (125f 75g). Yes that is very cheap.

[MatCauthon3]

@Player965976697

4->5 range, 4->5 attack, 30-35 HP, 80->85% accuracy. For 200ressources (125f 75g). Yes that is very cheap.

you forgot the faster creation time (-8 seconds). that said i think the crossbow upgrade is fine.

[AvihooI]

If it’s about subjective experience then why the constant pestering about me bringing actual data? Why the constant objections to logical fallacies?

There you go then: in my opinion archers are easy enough to deal with, crossbow upgrade isn’t cheap. I don’t see myself winning with crossbows often nor losing to them often. There are far more frustrating units to deal with (e.g. knights wrecking my eco simply because my pocket was a tad too late making his own cavalry or a cheap wall-to-fast-castle arambai/conq push).

If you just want to play a game that has less archers in it, then mod it and play with your friends. It’s very easy to mod.

That is not to say there can exist a different balanced and interesting meta out there that involves less archers. It just so happens that this hypothetical meta is not the official age data set. Mind you, a data set forged through loads of experience. If that amount of experience is wrong, then make your own mod/game/whatever.

[WoodsierCorn696]

@AvihooI

Confidence intervals are for determining whether or not a sample size is likely to be representative of the population. However, what we’re dealing with here is not a sample but the entire population of recent ranked matches (and you can find more of the same for HD and Voobly historically speaking).

You are wrong. The played games are the sample. We can calculate a winrate based on the sample. And we can use CI to find a range for the true winrate, a value we dont know and so we estimate this value by the sample.

Example 1:
Lets pick Mayans, one of the most popular picks (8%) at high level (1650+)

Wins: 156
Picked: 301
Current winrate: 51.83%

Based on a 95% confidence interval we can say there true winrate is between 46% and 57%.

Example 2:
Lets pick Indians, since the pick rate is pretty low (0,67%).
Wins: 13
Picked: 24
Current winrate: 54,17%

Based on a 95% confidence interval we can say there true winrate is between 35% and 75%.

This statistics show us that we just have too few data to really compare the current winrates of the pros. The CI are just too big with this amount of data.

Yes, you can have a look at all games. But i dont think you must look at all data by balancing the games. Noobs like me dont really know how to make use of all civ bonusses. There is a reason why Chinese was top civ for pros at voobly, but bottom tier for noobs on the same platform. It was all about there different start. Pros could use this to there advantages, for noobs this was a disadvantage. Balancing is for me mostly for high level players, so we need to look at their data only.

[AvihooI]

Dude what are you on about? The data takes into account ALL OF THE RANKED games. It is not a sample, it’s the entire data. How can there be data that is more real than the actual population?