A plea to the developers of PA
 

(cliffnotes at the end if you want to skip all this...)

To the creators/maintainers/developers/etc,

Round 31 has been, and will probably be remembered as, a gigantic flop of epic proportions. The amount of time and effort that went into balancing the races and ship classes was pretty much wasted. This round we saw gigantic FI/CO attack fleets that spent a lot of their time pecking away at tiny targets throughout the Uni. Alliances who sought to defend their members against these attacks would be forced to send 5+ fleet slots just to discourage the attacker so they could recall and attack again only two ticks later. Below are a few statistics taken at Tick 769 of Round 31:


What is incredibly discouraging is the laughable amount of CR/BS fleets as compared to FR/DE and especially to the FI/CO fleets. Looking at the race statistics, we see that hardly anyone was drawn to either the Etd or Zik races this round. 50% of the top 100 players (43.5% of the top 200) were Cat, and 32% of the top 100 players and top 200 players were Xan. Two races accounted for >80% of the top players in the Universe. Lo and behold, the primary attack fleets used by both of those races this round were FI/CO.

In an ideal game environment, you should STRIVE to achieve a BALANCED game where players are drawn to all races and all ship classes equally as they all have relatively equal pro's and con's. Instead, we saw Round 31 heavily favoring the Cat and Xan races and the use of FI or CO fleets. To encourage new player growth, and minimize player turnover between rounds you should be bending over backwards in order to achieve a balanced Universe. Extensive amounts of unit testing should be conducted with automated scripts to benchmark each race and their fleets in an attempt to give all players equal footing. A new player should not need to have a large Alliance at their fingertips in order to crunch all the numbers and figure out which race and ship class will be the power house. Experienced players will always have the advanced knowledge of tactics and teamwork that will be able to provide them with an advantage over other players. Number crunching should give more insight into tactics, but should not reveal such a ridiculous hole in gameplay mechanics.

Personally, I hope I'm just reiterating the major and obvious issues of this round and that the developers have already revamped their methods for creating ship stats. If that's not the case, then I'm hoping that something I've said will be recognized by the PA staff and they'll recognize the changes they need to make in order for Round 32 to be successful. Creating a better GUI for the game is great and I think everyone appreciates the work that's gone into making the website more user friendly, but all that is for nothing if the basic mechanics of gameplay are completely screwed.

----

Cliffnotes: Developers, spend a LOT more time running extensive tests on proposed ship and race stats before choosing the final numbers for a round. Make better unit testing scripts that can run thorough and varied battlecalcs to generate outcomes. Verify that fleets of relatively equal value are relatively equal in abilities. Make all races and ship classes a viable choice to a successful round. We don't need another experience like this again.

Re: A plea to the developers of PA
 

PA team should dedicate absolutely 0 time to improving shipstats and/or coding tools to make good shipstats.

Should be up to the community to do that (I thought they had been for the last bunch of shipstats tbh :() - there's far more important things for pa team to work on.*

edit: * bad shipstats won't make people leave the game, and they sure as hell won't help to draw new people into the game (new players couldn't care less about that aspect of the game mechanics..). Also, pa team don't play pa... so will always be better for members of the community to step up and create the stats (maybe have a stats team? :p)

Re: A plea to the developers of PA
 

An actual competent, dedicated stats team isn't a bad idea - the problem is finding people willing to be in charge of the thing. Stats are the cause of such contention that nobody wants the hassle anymore.

Re: A plea to the developers of PA
 

members of the community have been working on stats for the last few rounds. Cochese will be working with members of the community to create stats, race bonuses, salvage, etc
analysing race stats before the end of the round is not helpful nor accurate. Percentages reflect user opinion on the stats and may not be always useful for analysing stats.

Re: A plea to the developers of PA
 

The outcome of the round obviously depends massively on player choice.


The more players that go FI/CO, the better FI/CO is. The more players go CR/BS, the better CR/BS is. FR/DE is generally self regulating (since FR/DE fleets often target FI/CO/CR/BS leaving an anti FR/DE defence fleet)

What I think I've learned from this round is that there needs to be a very good anti FI/CO ship, particularly something that'll deal with xands. A defensive killship that fires at the same init (albeit maybe less efficient than last round's harpy), or non-xand FR/DE that shoot first. (The beetle's t2 was also a disastrous idea)

This help reduce public perception of the power of FI/CO and should improve balance.

Re: A plea to the developers of PA
 

Actually having the whole universe go fi/co made Cr/Bs very strong as there wasn't much going around so nobody built anti Cr/Bs. But now everyone is building Cr/Bs.
Fr/De wasn't as strong as there are quite a few Fr/De planets (WAFHH), and people's fi/co fleets target Fr/De.

Re: A plea to the developers of PA
 

There should be a tool to run tests. I've read someone post about it somewhere.

Basically, after drafting the stats' targetting specifications. Making it as fun and exciting as possible - like a certain race is strong on a particular ship class, plus taking into account the race' advantage in res/con/security/prod time.

Then, that's when the tool comes in. After inputting the stats' in the tool, it checks the general efficiency of ships. Simulating various attack combos against various defense combos. To make sure certain ship/s' properties isn't overpowered (emp res, initiative, guns, armour), or underpowered.

To me, beta testing is a place to test how exciting, playable the stats is, and if there are very obvious holes. The balance part of it, we need a computer or a tool (i.e. excel speadsheet) for it.

Re: A plea to the developers of PA
 

I'd say a tool like that would be just as simple for one of the stats-ppl to make as it is pa-team (assuming they have some basic programming knowledge). Just make a stats team and make sure a semi-decent techie is in it (christ even I would suffice and I'm pretty shite) to automate some tests. Of course pa team would need to provide some nice bcalc ... stuff.

Re: A plea to the developers of PA
 

No. Because the players choice obviously depends on what ship class/ race is better "based on the stats' " - taking into account time travel, efficiency, race bonuses, combo variations.

It's not like they have to draw straws what they'll choose - then FI/CO is the straw that 80% of them picked.

Re: A plea to the developers of PA
 

I really doubt that coding a program that could accurately analyse the stats for balance pre-round is possible. There are a staggering number of variables and a lot of feedback. I mean a huge % of the universe went xan but I don't think they were actually that good this round. This is primarily due to just how good cath are though and the fact that a higher number of people (higher than in previous rounds) went for them, thereby making xan relatively worse. It's hard to figure out how to factor in something like the importance of salvage into this round, which has really been a bigger change than everything else put together.

Re: A plea to the developers of PA
 

It's my experience that 'popular opinion' on stats is often wrong.

Eg round 27 etd CR/BS 'sucked'. Last round terran FR sucked and CR was the way to go.


I thought of writing a little program to test races out, but then I couldn't be bothered (also the only programming language I know atm is fortran...). A working one would be very useful but probably a bitch to code properly. Maybe MrLobster could let us use PAOL coding?

Re: A plea to the developers of PA
 

People aren't often wrong about the reasons they pick their race/classes for. It's just that they often incorrectly evaluate the importance of these reasons. I mean terran cr was able to own xans pretty hard as people had thought. The problem was it was owned even harder by etd bs. That, and the length of the round, meant that people gravitated towards fr due to this and its natural advantages. The fact that it's generally better to be difficult to roid than good at roiding in pax helps out here as well.

Re: A plea to the developers of PA
 

I highly doubt any automated tool (beyond maybe doing 10 calcs automatically instead of manually here and there) will be of much use. A competent stats team to find any glaring mistakes, and then betas is about your best bet.

Something like, you give an attack fleet (eg, a xan fi fleet), and the tool will automatically go through every anti fi ship seeing how efficient they are etc is quite easy to do... but I'd like to think a competent stats team can grasp stuff like that by looking at the stats themselves :up:

So what exactly would you want this code to do? Baring in mind artificial intelligence in scripts isn't something many people will be willing to code for you ;p

Re: A plea to the developers of PA
 

There is an actual stats efficiency page (linked at the bottom of the stats page), if that's all you're really talking about. It's basically the only way to do emp heh.

Re: A plea to the developers of PA
 

I think one way to remedy the fi/co fleet imbalance is to hard code the game for no out of ally/out of gal/out of cluster defense as for anti fi and co it cant be done without it being prelaunched. I.E. it doesn't go red until eta 7 with full eta researched and that can only be done with intag ally def.

I have seen alot of eta 8 banshee/nixes def.
and eta 9 Bws and eta 9 xan def from planets that have no bs/cr.
(thats rules out any cr ships being sent)

Re: A plea to the developers of PA
 

I'm not sure how you're justifying your point?
You're saying that alliances are so pushed for defence that they're getting prelaunched out of alliance defence to cover themselves? Removing this defence just leaves the alliance weaker to fi/co. The alliance has to then mass-produce anti fi/co and / or pull all fi/co attack fleets to help defend. This hurts the alliance but doesn't seem to solve the problem imo?

I think that there's too many fi/co fleets, so there's no ship that can be built purely for 0 loss fi/co defence at alliance eta. Even a rogue-type ship (bs -> fi/co) would seriously redress the balance. I'd rate these as higher than looking at out of galaxy/alliance defence.

Re: A plea to the developers of PA
 

I'll go into more detail.
I am not saying there is alot of prelaunching def against fi and co fleets and predicting prelaunching via jumpgate probing yourself a few minutes before tick.
I am saing that there is a tactical disavantage with frig/de/cr/bs fleets.
whilst there is out of ally/cluster defending.

Presume full eta research has been done.
lets say if you have a cath fi or a xan co fleet (eta 8 in base goes "red" at eta 7)
there are only going to be 3 types of def against it. 1st ingal 2nd same cluster and 3rd ally....def it isnt possible for any types of def to do the eta (hence a tactical advantage compared to fr/de/cr/bs)

If you have any other types of roiding fleets then it can be def against out of tag / out of cluster.

predominately banshees and nixies are the def ship of choice v frig and de fleets bombers and Black widows. that have a 1 tick window where regardless if it is in ally/cluster or not.

even if there was a bs or a cr class anti fi/co ship that is 0 loss def it still wouldnt fundamentally correct this....

without naming alliances... they do use out of tag out of cluster def

another thing with xans it is always tricky with the old fi class anti frig (banshee) and frig class anti fi ships (ghosts) is which one should have the better init. I would have personally would have swaped the inits around.
if memory serves me correctly it was something similar back in r15 when this was done there was alot of support planets/out of ally def then again if it was hard coded to disable out of tag def. then this problem wouldnt have been an issue.

Re: A plea to the developers of PA
 

De is REALLY strong this round. Problem is that with the sick amounts of FI u are just doomed to be FCed in the end :)

Re: A plea to the developers of PA
 

MrLobster has PAOL which might work.


Alternatively, I might be able to make a statistical mechanics model but I don't have the time right now. That could be low on coding...

How many of us are physicists/mathematicians?

The init was an awful, awful thing & I avoid xand FR targeting xand FI first with lower init as a gut reaction. With multitargeting there's plenty of ways around that though, that was just my laziness.

Re: A plea to the developers of PA
 

I bet if there was fi or co class anti bs/cr then the frig/dest count would be even lower

Re: A plea to the developers of PA
 

Can we go back to actually setting up races to their traditional strengths? Don't give xans CR pods, the very notion of a terran having a CO pod is absurd. If you give all the races the option to build fast roiding ships, then of course people will go down that road as everything will go to lowest common denominator fleet compositions. By limiting choice you actually add depth because if certain races rely on high class ships and others on low class ones, you actually force people into the challenge of balancing their fleets so they can actually defend.

Re: A plea to the developers of PA
 

I'm in complete agreement with Lokken.

Re: A plea to the developers of PA
 

what's the problem with Xan CR or BS fleets? and why not Ter CO?
Tbh, although Ter had a co fleet, it's very inefficient compared to Ter BS.
But, really, change the whole concepts we had this round.
Ter should have high init high empres ships.
And, when calcing efficiency, take into account the init: Ter/Zik might have decent efficient ships, but when matched against lower init ships, like Cath/Xan, it's a whole different thing and get owned pretty easily, thus reducing their efficiency against these races.
That's why last round harpy rocked against Xan Fi roid fleet, it was the Xan nightmare, with same init as Xan anti-fi/co FI (don't remember the name).
If keeping this salvage %, thus lowering the stealers defense ratio, at least get some decent anti-fi/co FI/CO, which ever race it is, with a decent to high empres and low init, matching the Xan anti-fi/co FI/CO and able to make some harm to cath FI/CO fleets.
Edit: like this round, cath FI had bigger empres than Ter CO. Why? Because of spiders low init high efficiency?
Edit: another thing: make any FI/CO roid fleet efficiency a bit lower, it already has the advantage of speed

Re: A plea to the developers of PA
 

I wholeheartedly agree with you lokken.

When I asked Appocomaster about why terrans had stealers his argument was: everyone loves stealers.

Re: A plea to the developers of PA
 

playing r14 terran was abolsutely fun with having wyvern as a bs class anti de stealing ship

edit - come to think of it probily the best round in terms of stats I have played.

Re: A plea to the developers of PA
 

I'm not really opposed to ships that steal for all races - but they should be off pod class to ensure that zik retains their advantage.

Re: A plea to the developers of PA
 

r14 stats were quite fun idd. Might not have been balanced fully, but fun :p

Re: A plea to the developers of PA
 

That explains why the only 3 terrans in the t100 are Co based... (only one has BS at all, and it's not very big)

Re: A plea to the developers of PA
 

This is just skipping around the larger question. Regardless of how strong etd/zik/ter are this round they have to have a lot of their fleet value in anti-fi/co in order to combat the fact that something like 60% of the value of the entire universe is in cath/xan (and the majority of that in fi/co). Planetarion stats are very much a matter of tipping points, once a few people think it's a good idea to go fi/co (ie above the amount you'd expect if everyone picked their strategy at random) it becomes very heavily weighted in favour of adjusting your own strategy to compensate for this change.

Re: A plea to the developers of PA
 

It is a pity alot of the universe went FI/CO as CR/BS are not bad at all. A basic Lancer + a few BS setup for Etd is perfectly fine for example.

I agree with the posts above to make the races more traditional so more thought goes in to the race pick.

Re: A plea to the developers of PA
 

It's not necessarily that cr/bs is that bad. It's more that cath fi are that ****ing good. And so many people went xan and fi/co is xan's most viable fleet.

Re: A plea to the developers of PA
 

Agreed.

And imo it is the cath that should be tweaked. People always tend to whine about Xan, but I find them quite poor this round. Cath is waaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaay overpowered... and this was also the case last round iirc. Also. I think there should be a discussion on the removal of etd. I think that may make balancing the stats a bit easier. Note: I haven't made stats ever, so if this observation is horribly wrong please correct me. :)

Re: A plea to the developers of PA
 

I hope you're not a doctor... "your leg hurt ? let's cut it"

Re: A plea to the developers of PA
 

But you do remove the appendix if it is sore :-)

Re: A plea to the developers of PA
 

Exactly. Only 4 Ter on t100.
Tbh, Ter BS are way more efficient than Ter CO. And, at least, have a decent target, seeing that Xan went CO with bomber as anti-BS/CR, and that gets owned by Wyvern due to ... low init
Problem is the 2 ticks defense and the higher FC probability.
Only usefulness of Ter CO is armour flak, being it on attack or defense.
Seen that universe stats, i've seen more 500 Ter on round start. Now, only 370. Meaning, some have reset their planet or simply deleted it, since Ter stats suxxxx bad this round.
So, if you want to go high amp scanner and try to play it a bit, what to do? Go Xan???
I hope someone tries to change the things for scanners. Atm, they are kicked from gals, set apart from attacks and looked like the "DEAD WEIGHT" from alliance. Because, not even going Ter is a choice, atm.

Re: A plea to the developers of PA
 

I have a large planet.
And i have like 400k spiders 80k scarabs and 30k black widow as an attack fleet.

Still when my attacks get coverd its being coverd by fi/co class. Only in the beginning of the round i had some attacks coverd with other fr/de ships. Never had ingal cr/bs covering my attacks.

And when i have incs its still fi/co while any att fleet should be fr/de/cr/bs.
There is just to much fi/co around.

All races just have to many pod classes.

Atm its:
Terr: CO/FR/BS pods
Cath: FI/DE/CR pods
Xan: CO/FR/CR pods
Zik: FR/BS pods
Etd: FR/DE/BS pods

It should be like:
Terr: DE/BS pods
Cath: CO/CR pods
Xan: FI/FR pods
Zik: FR/CR pods
Etd: CO/DE (1) or DE/BS (2) pods (Or CO/BS. But then Etd will lose their traditional terran link)

When on galstatus.
This adds up in (1):
3 eta 7 roidfleets
4 eta 8 roidfleets
3 eta 9 roidfleets

This adds up in (2):
2 eta 7 roidfleets
4 eta 8 roidfleets
4 eta 9 roidfleets

This adds up in (3):
3 eta 7 roidfleets
3 eta 8 roidfleets
4 eta 9 roidfleets

More high eta fleets will make better combo's possible but have a disadvantage of being covered more easely cause it takes longer to reach the gal.

Re: A plea to the developers of PA
 

You made it look like I decided that was the best option. I merely proposed to discuss the merits and faults of such an action (in a new thread).

Re: A plea to the developers of PA
 

Back to basics, like some rounds ago.
Why not?
And why not back to single targetting?
Easier to play, easier to defend, harder to selfcover, more type of ships needed (no longer Xan FI/CO with bomber or Ter FR/DE or Cath DE fleet)
But, yet, would it help to new players? or game overall?

Re: A plea to the developers of PA
 

i would love a round with single targeting :O:O:O:O

but im quite sure they wont bring it back :S

Re: A plea to the developers of PA
 

tbh i also think single targeting might help new players in some way.. i mean its simplier for em to understand stats.. ive seen ppl playing this game that got absolutly no clue how to read stats etc.. t2, t3.. etc..

i might not remember it correctly but imo when single targeting was engaged then the gameplay was more offencive.. which this game needs imo :)

Re: A plea to the developers of PA
 

would certainly make stats creating easier

Re: A plea to the developers of PA
 

QFT!

<3 single targetting :)

Re: A plea to the developers of PA
 

I think T3 should be deleted totally. I think just 50% should have a T2 target.

And i think it helps new players because there being bashed by alot of combo's. Its easyer to understand for them if they just see 1 race incomming.

Re: A plea to the developers of PA
 

A ROUND WITHOUT DOWNTIME WOULD PROBABLY BE A GOOD THING DONT YOU THINK?
WHAT YOU DONT AGREE? **** YOU THEN






lowercaselowercaselowerecaselowerecveaselowewcaselo0wecaselowecercaselowecere

Re: A plea to the developers of PA
 

You are hopelessly oversimplyfing (and you're wrong, too).

Re: A plea to the developers of PA
 

A better solution would be to give everyone access to required intelligence + helper tools from start.

Even top players don't have incomming scans till like tick 500 (just a guess there, didn't count or something), how are new players supposed to know what's comming at them. Single targetting is not going to solve this.

Re: A plea to the developers of PA
 

Nope. And tbh, there isn't one, except remove completely the scan tree.
New players will need always some guidance, even for own incs.
And even with multi-targetting. But, with single targetting, roid fleets usually target only 2 kind of ships, not 4~6 (like the Cath DE fleet of this round) making harder the choice of what ship to build against certain incoming.

Re: A plea to the developers of PA
 

Am i?
It's easier to defend: you have the ships to stop it or not. You can't stop the incoming? get defense from alliance/galaxy.
With multi, you might end with less efficient defense ships targetting inc at T2.
FR/DE inc from 1 planet and can't stop all FR, so leaving the DE free to roid you? That won't happen with single-targetting, no flak possible. Unless, ofc, on teamups.
Easier to attack: get the specific race with less efficient ships against your roiding fleet and throw it all in.
Hard to self-cover: all races have, at least, one known flaw, so you know you'll get incs from that kind of roiding fleet.
But i have to agree: multi-targetting is fun and easy to build your fleet: in some cases you just need 4 kind of ships; with single-targetting, you need at least 7, 1 hit to each kind of ships and 1 for roiding fleet.

Re: A plea to the developers of PA
 

I'm for a T3 ALL for every ship :p

Re: A plea to the developers of PA
 

It's harder to defend. You end up with more different ships and less value targetting each class.