Switch Theme:

Votann Nerfed Prematurely  [RSS] Share on facebook Share on Twitter Submit to Reddit
»
Author Message
Advert


Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you:
  • No adverts like this in the forums anymore.
  • Times and dates in your local timezone.
  • Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance.
  • Email notifications for threads you want to watch closely.
  • Being a part of the oldest wargaming community on the net.
If you are already a member then feel free to login now.




Made in de
Morally-Flexible Malleus Hearing Whispers




So now that we have a new "imperial faction" does GW make an equal but opposite new "Dark Votann" faction?
   
Made in gb
Gore-Drenched Khorne Chaos Lord




Asmodios wrote:

uhhhhh the only "basic math" that is flawed is your own. If there are 25 factions (just assuming all have the same popularity to keep it simple). That means that there are 24 players for every 1 votan player. If LOV is crushing everyone else that means there are 24 factions aka 24/1 people being negatively affected by this. If LOV sucked then while its still undesirable you at least only have 1 faction being negatively affected so 1 compared to 24 others.

So you then go on to admit that they will probably land between a 45-55% win rate so what exactly is the issue? It honestly sounds like you are mad you didnt get a few weeks of auto wins.


Once again because you apparently either have short term memory issues or simply new new material:

Dudeface wrote:And you're here leveling accusations I want free wins for a faction I don't own, don't want to own and openly stated I feel needed a nerf in this very thread and comment chain.


But since I'm an idiot can you tell me how the LOV player is affecting 24 players across 5 games? It doesn't matter if they're 90% or 10% they affect 5 other peoples games/day. The only difference is at 90% the top spot is likely pre-determined, at 10% there's 5 players who have a better shot at winning than the other 19 depending on pairing techniques. Both are bad and both would need banning.

Does that make sense or are you still concerned about the people buying OP units because they're OP and feeling hurt? That seemed to be a big concern to you, maybe you're just happy your meta chasing got sorted before you wasted money.
   
Made in dk
Loyal Necron Lychguard






It wouldn't be one LoV player, it'd be 60% of the field, so it'd be every player, just like when any faction has reigned as the only tier 0 army.
   
Made in gb
Gore-Drenched Khorne Chaos Lord




 vict0988 wrote:
It wouldn't be one LoV player, it'd be 60% of the field, so it'd be every player, just like when any faction has reigned as the only tier 0 army.


That's a community issue spawned from a combination of competitive obsession and GW leaving enough imbalance to let them leverage it.

The point is and I agree that those examples are worthless beyond showing how people think. They'll never happen realistically. I don't think even nids have accounted for 60% pickrate have they though?
   
Made in us
Stealthy Kroot Stalker





FezzikDaBullgryn wrote:
So now that we have a new "imperial faction" does GW make an equal but opposite new "Dark Votann" faction?


It's not a new Imperium faction, so no.
   
Made in us
Ancient Venerable Dreadnought




San Jose, CA

Xenos...cuz THEYRE NOT SQUATS!!!
   
Made in us
Hurr! Ogryn Bone 'Ead!





By the way, if I remember correctly, the amount of time needed to exhaustively test a codex was based on testing a codex against every other codex (don't remember if supplements counted separately) for every Nephlim mission. It might have also included enough games so that each secondary could be picked once in each match up (I feel like the calculation was tailored towards LoV)? The end number of days/months/years it would take may also have taken into account 8-hour work days. I don't think it was counting "man-hours" (which would double the number of required play testing hours to account for two human players), but I could be wrong. I think it was assuming only one game occurring at a time.

I wasn't the one who made the calculation, just sharing what I remember.

This message was edited 1 time. Last update was at 2022/11/01 17:56:01


 
   
Made in gb
Gore-Drenched Khorne Chaos Lord




DeadliestIdiot wrote:
By the way, if I remember correctly, the amount of time needed to exhaustively test a codex was based on testing a codex against every other codex (don't remember if supplements counted separately) for every Nephlim mission. It might have also included enough games so that each secondary could be picked once in each match up (I feel like the calculation was tailored towards LoV)? The end number of days/months/years it would take may also have taken into account 8-hour work days. I don't think it was counting "man-hours" (which would double the number of required play testing hours to account for two human players), but I could be wrong. I think it was assuming only one game occurring at a time.

I wasn't the one who made the calculation, just sharing what I remember.


2 people
3 hour game
25 factions
8 missions
3 * 8 * 2 * 25 hours = 1200 hours
+ 15% for admin/feedback/etc = 1215 hours

Average working day in the UK is 7.5 hours, 162 man hours, assuming that every second of working time fits nicely, which it won't. That you don't need to make changes and reset the cycle, which they will.

Real terms 2 games a day of the 600 needed = 300 days
   
Made in us
Fixture of Dakka




NE Ohio, USA

 Unusual Suspect wrote:
FezzikDaBullgryn wrote:
So now that we have a new "imperial faction" does GW make an equal but opposite new "Dark Votann" faction?


It's not a new Imperium faction, so no.


Why's that make the answer "No"? Eldar aren't Imperial and never have been & that didn't preclude Dark Eldar from becoming a thing.
   
Made in us
Arch Magos w/ 4 Meg of RAM






Dudeface wrote:
 Flinty wrote:
It would be an unpleasant task, and individually demoralising. But the backdrop is that out of 25 possible factions, each with different levels of popularity, it is better to have to tell one of the factions this, rather than having to tell 24 factions this.


Sorry, to drop that same metrics back out, in that hypothetical perfect spread 25 person 5 round event, you're telling 24 people there's a 4% chance they'll have a game that's almost a dead certain loss.


which is better than telling 24 people theres a 96% chance they'll get rammed by an OP codex
   
Made in gb
Gore-Drenched Khorne Chaos Lord




 VladimirHerzog wrote:
Dudeface wrote:
 Flinty wrote:
It would be an unpleasant task, and individually demoralising. But the backdrop is that out of 25 possible factions, each with different levels of popularity, it is better to have to tell one of the factions this, rather than having to tell 24 factions this.


Sorry, to drop that same metrics back out, in that hypothetical perfect spread 25 person 5 round event, you're telling 24 people there's a 4% chance they'll have a game that's almost a dead certain loss.


which is better than telling 24 people theres a 96% chance they'll get rammed by an OP codex


Well, that's wrong, my example was actually that 5 people out of 24 will play the votann.

For a dice and probability game based in maths, people aren't great at it.

This message was edited 1 time. Last update was at 2022/11/01 18:31:32


 
   
Made in us
Longtime Dakkanaut




Dudeface wrote:
Asmodios wrote:

uhhhhh the only "basic math" that is flawed is your own. If there are 25 factions (just assuming all have the same popularity to keep it simple). That means that there are 24 players for every 1 votan player. If LOV is crushing everyone else that means there are 24 factions aka 24/1 people being negatively affected by this. If LOV sucked then while its still undesirable you at least only have 1 faction being negatively affected so 1 compared to 24 others.

So you then go on to admit that they will probably land between a 45-55% win rate so what exactly is the issue? It honestly sounds like you are mad you didnt get a few weeks of auto wins.


Once again because you apparently either have short term memory issues or simply new new material:

Dudeface wrote:And you're here leveling accusations I want free wins for a faction I don't own, don't want to own and openly stated I feel needed a nerf in this very thread and comment chain.


But since I'm an idiot can you tell me how the LOV player is affecting 24 players across 5 games? It doesn't matter if they're 90% or 10% they affect 5 other peoples games/day. The only difference is at 90% the top spot is likely pre-determined, at 10% there's 5 players who have a better shot at winning than the other 19 depending on pairing techniques. Both are bad and both would need banning.

Does that make sense or are you still concerned about the people buying OP units because they're OP and feeling hurt? That seemed to be a big concern to you, maybe you're just happy your meta chasing got sorted before you wasted money.

because with a 90% win rate only LOV would win events. Every other faction at the event will have to pull off a miracle to win thus all other factions are disenfranchised to go because you will have to play LOV to realistically have a chance. A 10% win faction on the other hand will never advance past the 1st or 2nd round for contention thus meaning the other 24 factions are not disenfranchised. So its in unequivocally better for the game that 1 faction be horrendously underpowered than horrendously overpowered.

I play imperial guard. My faction has been bottom of the barrel for almost all edition..... Id rather that be the case than we have a 90% win rate because once again that would ruin the game for all factions that don't play IG. Also the "meta chaser" knows what he's getting into and will continue to switch units and factions in order to win. The timing of the nerf has little effect on that person because they are ready to drop and move on to the next army. The people that would have been hurt the most by a late nerf are those taking time to plan out and build and army and either don't have the budget or time to switch once they are locked in. A late nerf would have hurt casual LOV players the most. Now they can better plan their purchases without having to worry about massive swings in power that might make certain builds a waste of time.
   
Made in us
Banelord Titan Princeps of Khorne




Noctis Labyrinthus

Love the seething over LoV getting exactly what they deserve. Feels good man.
   
Made in us
Longtime Dakkanaut




Dudeface wrote:
DeadliestIdiot wrote:
By the way, if I remember correctly, the amount of time needed to exhaustively test a codex was based on testing a codex against every other codex (don't remember if supplements counted separately) for every Nephlim mission. It might have also included enough games so that each secondary could be picked once in each match up (I feel like the calculation was tailored towards LoV)? The end number of days/months/years it would take may also have taken into account 8-hour work days. I don't think it was counting "man-hours" (which would double the number of required play testing hours to account for two human players), but I could be wrong. I think it was assuming only one game occurring at a time.

I wasn't the one who made the calculation, just sharing what I remember.


2 people
3 hour game
25 factions
8 missions
3 * 8 * 2 * 25 hours = 1200 hours
+ 15% for admin/feedback/etc = 1215 hours

Average working day in the UK is 7.5 hours, 162 man hours, assuming that every second of working time fits nicely, which it won't. That you don't need to make changes and reset the cycle, which they will.

Real terms 2 games a day of the 600 needed = 300 days

LOL "25 factions" there are not 25 factions. You have:
1. Loyal Marines
2. Evil Marines split into three codices unnecessarily
3. Knights split into two codices unnecessarily
4. Custodes
5. Sisters
6. Necrons
7. Eldar
8. Evil Eldar
9. Orks
10. Tau
11. Loyal Marines but they're all Psykers
12. Daemons
13. Genestealer Cults
14. IG
15. AdMech
16. Tyranids
   
Made in gb
Gore-Drenched Khorne Chaos Lord




EviscerationPlague wrote:
Dudeface wrote:
DeadliestIdiot wrote:
By the way, if I remember correctly, the amount of time needed to exhaustively test a codex was based on testing a codex against every other codex (don't remember if supplements counted separately) for every Nephlim mission. It might have also included enough games so that each secondary could be picked once in each match up (I feel like the calculation was tailored towards LoV)? The end number of days/months/years it would take may also have taken into account 8-hour work days. I don't think it was counting "man-hours" (which would double the number of required play testing hours to account for two human players), but I could be wrong. I think it was assuming only one game occurring at a time.

I wasn't the one who made the calculation, just sharing what I remember.


2 people
3 hour game
25 factions
8 missions
3 * 8 * 2 * 25 hours = 1200 hours
+ 15% for admin/feedback/etc = 1215 hours

Average working day in the UK is 7.5 hours, 162 man hours, assuming that every second of working time fits nicely, which it won't. That you don't need to make changes and reset the cycle, which they will.

Real terms 2 games a day of the 600 needed = 300 days

LOL "25 factions" there are not 25 factions. You have:
1. Loyal Marines
2. Evil Marines split into three codices unnecessarily
3. Knights split into two codices unnecessarily
4. Custodes
5. Sisters
6. Necrons
7. Eldar
8. Evil Eldar
9. Orks
10. Tau
11. Loyal Marines but they're all Psykers
12. Daemons
13. Genestealer Cults
14. IG
15. AdMech
16. Tyranids


Good, I look forward to your footnotes on the thousand sons discipline from your deathguard games and nuanced feedback on voidweavers in your ulthwe game.
   
Made in gb
Longtime Dakkanaut




I think the argument is more "have more testers".

Whether its worthwhile for GW to have 10-20 people who just play 2 games of 40k every working day is an open question - but really, its probably not *that* expensive. That would give you say 200-400 games a month. Which would probably give you decent indications.

Admittedly whether these people would go mad playing this much and trying to keep the countless different rules versions in their heads is an open question.

I guess for "cheap" you could have say 5-10 archetype armies that you play into and then see how it feels. If its a bit much, it should be obvious. I feel though there are quite a few different sort of lists out there, and so going "right, one game into Marines, tick" isn't really going to give you much.
   
Made in pl
Fixture of Dakka




 Flinty wrote:
It would be an unpleasant task, and individually demoralising. But the backdrop is that out of 25 possible factions, each with different levels of popularity, it is better to have to tell one of the factions this, rather than having to tell 24 factions this.


Oh I am sure that people who were getting farmed by the top factions are delighted to hear, that after the nerfs the army is not only better then theirs, but is losing to the armies they are losing too. Also absolutly doesn't explain litteral months of other factions siting at 60%+ with no reaction from GW.
It creates another bad precedents, aspecialy when parts of the community say, I don't play them, so it is okey.


So you then go on to admit that they will probably land between a 45-55% win rate so what exactly is the issue? It honestly sounds like you are mad you didnt get a few weeks of auto wins.

Because for factions with limited number of unit choices they have to be better then 45-55% win rate on day one, if they are to be okey in 12-18 months. Only harlequins are able to pulls stuff like that, and that is because for some reason GW writes their rules the way they do. Also for outside of tournament play, a LoV army ment that if someone new were to run in to a more optimised list run by someone playing for years, they would actualy have a chance to have a fun game, instead of just losing.

If you have to kill, then kill in the best manner. If you slaughter, then slaughter in the best manner. Let one of you sharpen his knife so his animal feels no pain. 
   
Made in us
Longtime Dakkanaut




Karol wrote:
 Flinty wrote:
It would be an unpleasant task, and individually demoralising. But the backdrop is that out of 25 possible factions, each with different levels of popularity, it is better to have to tell one of the factions this, rather than having to tell 24 factions this.


Oh I am sure that people who were getting farmed by the top factions are delighted to hear, that after the nerfs the army is not only better then theirs, but is losing to the armies they are losing too. Also absolutly doesn't explain litteral months of other factions siting at 60%+ with no reaction from GW.
It creates another bad precedents, aspecialy when parts of the community say, I don't play them, so it is okey.


So you then go on to admit that they will probably land between a 45-55% win rate so what exactly is the issue? It honestly sounds like you are mad you didnt get a few weeks of auto wins.

Because for factions with limited number of unit choices they have to be better then 45-55% win rate on day one, if they are to be okey in 12-18 months. Only harlequins are able to pulls stuff like that, and that is because for some reason GW writes their rules the way they do. Also for outside of tournament play, a LoV army ment that if someone new were to run in to a more optimised list run by someone playing for years, they would actualy have a chance to have a fun game, instead of just losing.

So you're saying the lower an army model count you should have a higher win%? I see no reason to artificially make the win rate higher because of amount of data slates
   
Made in gb
Gore-Drenched Khorne Chaos Lord




Tyel wrote:
I think the argument is more "have more testers".

Whether its worthwhile for GW to have 10-20 people who just play 2 games of 40k every working day is an open question - but really, its probably not *that* expensive. That would give you say 200-400 games a month. Which would probably give you decent indications.

Admittedly whether these people would go mad playing this much and trying to keep the countless different rules versions in their heads is an open question.

I guess for "cheap" you could have say 5-10 archetype armies that you play into and then see how it feels. If its a bit much, it should be obvious. I feel though there are quite a few different sort of lists out there, and so going "right, one game into Marines, tick" isn't really going to give you much.


I've no idea what an on the books playtester earns, I'd wager at least 25k though, so £250,000 for 40k testing at the 10 player mark for 220ish games a month. Agreed though it would be a migraine inducing job.
   
Made in pl
Fixture of Dakka




Asmodios 807483 11450408 wrote:
So you're saying the lower an army model count you should have a higher win%? I see no reason to artificially make the win rate higher because of amount of data slates


If it is suppose to function months in advance, and in case of LoV between a potential edition reset ? Of course. Which armies take nerfs over and over again, and are largely not impacted in their standing as good army? Those which have been designed, or even over designed , in the gear, point costs, rules etc sections. So even when GW nerfs something or intreduces a new army to the pot, the armies can adapt. How are GSC doing right now or GK ? One army was balanced on release and the other was balanced as part of the reigning in of multi detachment army. While an army like Harlequins is a top tier army, the second time in the same edition. And how do they do it? They have extremly powerful rules and undercosted unit, coupled with synergies that are not build for "fun", or not just for it, but also for game play efficiency.


Automatically Appended Next Post:
Dudeface 807483 11450409 wrote:

I've no idea what an on the books playtester earns, I'd wager at least 25k though, so £250,000 for 40k testing at the 10 player mark for 220ish games a month. Agreed though it would be a migraine inducing job.


If were to trust what ex studio members told about playtesting, then a ton of it is pushed on to people working at the studio, and then they are expected to either find a group of people who test it for them or they have to test it themselfs, with no extra budget to hire people.

This message was edited 1 time. Last update was at 2022/11/01 19:30:08


If you have to kill, then kill in the best manner. If you slaughter, then slaughter in the best manner. Let one of you sharpen his knife so his animal feels no pain. 
   
Made in us
Longtime Dakkanaut




Karol wrote:
Asmodios 807483 11450408 wrote:
So you're saying the lower an army model count you should have a higher win%? I see no reason to artificially make the win rate higher because of amount of data slates


If it is suppose to function months in advance, and in case of LoV between a potential edition reset ? Of course. Which armies take nerfs over and over again, and are largely not impacted in their standing as good army? Those which have been designed, or even over designed , in the gear, point costs, rules etc sections. So even when GW nerfs something or intreduces a new army to the pot, the armies can adapt. How are GSC doing right now or GK ? One army was balanced on release and the other was balanced as part of the reigning in of multi detachment army. While an army like Harlequins is a top tier army, the second time in the same edition. And how do they do it? They have extremly powerful rules and undercosted unit, coupled with synergies that are not build for "fun", or not just for it, but also for game play efficiency.

the meta-watch article that explained the LOV nerfs (so it's just a couple weeks old) had GK at a 46% win rate and had grey knights at a 47% win rate. So both of them were fine, meanwhile, harlequins were at a 58% win rate so outside the 45-55% win rate. So wouldn't data support not breaking low model count books on purpose? Thousand sons were also at 49% and have one of the lowest model counts as well.
   
Made in it
Shas'la with Pulse Carbine





Dudeface wrote:
Tyel wrote:
I think the argument is more "have more testers".

Whether its worthwhile for GW to have 10-20 people who just play 2 games of 40k every working day is an open question - but really, its probably not *that* expensive. That would give you say 200-400 games a month. Which would probably give you decent indications.

Admittedly whether these people would go mad playing this much and trying to keep the countless different rules versions in their heads is an open question.

I guess for "cheap" you could have say 5-10 archetype armies that you play into and then see how it feels. If its a bit much, it should be obvious. I feel though there are quite a few different sort of lists out there, and so going "right, one game into Marines, tick" isn't really going to give you much.


I've no idea what an on the books playtester earns, I'd wager at least 25k though, so £250,000 for 40k testing at the 10 player mark for 220ish games a month. Agreed though it would be a migraine inducing job.

£250k for 220 games a month sounds like a bad return on investment, when a single weekend of tournaments gives you many more games to analyse.
Just release the rules, wait for results to come in and adjust accordingly in future updates. Oh, that's what they are doing already


 
   
Made in us
Regular Dakkanaut




No excuse for a company that pulls in this much profit and charges these rediculas prices.
If they need to hire more play testers wich they clearly do then they better get on it.

They want to charge premium prices there gak better be premium.
   
Made in us
Arch Magos w/ 4 Meg of RAM







Karol wrote:

Because for factions with limited number of unit choices they have to be better then 45-55% win rate on day one, if they are to be okey in 12-18 months. Only harlequins are able to pulls stuff like that, and that is because for some reason GW writes their rules the way they do. Also for outside of tournament play, a LoV army ment that if someone new were to run in to a more optimised list run by someone playing for years, they would actualy have a chance to have a fun game, instead of just losing.


actually braindead take lmao.

So Knights should have 100% winrate by that logic

This message was edited 2 times. Last update was at 2022/11/01 21:57:20


 
   
Made in us
Longtime Dakkanaut




Dudeface wrote:
EviscerationPlague wrote:
Dudeface wrote:
DeadliestIdiot wrote:
By the way, if I remember correctly, the amount of time needed to exhaustively test a codex was based on testing a codex against every other codex (don't remember if supplements counted separately) for every Nephlim mission. It might have also included enough games so that each secondary could be picked once in each match up (I feel like the calculation was tailored towards LoV)? The end number of days/months/years it would take may also have taken into account 8-hour work days. I don't think it was counting "man-hours" (which would double the number of required play testing hours to account for two human players), but I could be wrong. I think it was assuming only one game occurring at a time.

I wasn't the one who made the calculation, just sharing what I remember.


2 people
3 hour game
25 factions
8 missions
3 * 8 * 2 * 25 hours = 1200 hours
+ 15% for admin/feedback/etc = 1215 hours

Average working day in the UK is 7.5 hours, 162 man hours, assuming that every second of working time fits nicely, which it won't. That you don't need to make changes and reset the cycle, which they will.

Real terms 2 games a day of the 600 needed = 300 days

LOL "25 factions" there are not 25 factions. You have:
1. Loyal Marines
2. Evil Marines split into three codices unnecessarily
3. Knights split into two codices unnecessarily
4. Custodes
5. Sisters
6. Necrons
7. Eldar
8. Evil Eldar
9. Orks
10. Tau
11. Loyal Marines but they're all Psykers
12. Daemons
13. Genestealer Cults
14. IG
15. AdMech
16. Tyranids


Good, I look forward to your footnotes on the thousand sons discipline from your deathguard games and nuanced feedback on voidweavers in your ulthwe game.

Subfactions require less testing. It's easy to pinpoint if one is internally or externally broken, like if Iron Hands just had a 3+++ for example. I don't need to test that or how Votaan Rail weapons work but here we are with someone defending a trash release from GW because they don't get as many wins as they hoped for.


Automatically Appended Next Post:
Also are you seriously suggesting that Voidweavers are somehow performing different vs Ulthwe or Sam Hain? Man that's some non logic if I've ever seen it

This message was edited 1 time. Last update was at 2022/11/01 22:17:30


 
   
Made in us
Regular Dakkanaut




Wait are people actually saying things like 10 percent of the player base should be sacrificed so 90 percent of players can have a good time? Geez.


Automatically Appended Next Post:
Aecus Decimus wrote:
Dudeface wrote:
Excellent, better have a faction with 10% win rate than 90% right?


Yes. A faction with a 10% win rate makes things miserable for a small subset of players: the people who play that specific faction and do not have any other army. A faction with a 90% win rate makes things miserable for everyone except that faction. One of these groups is much larger than the other and if I have to choose which one to sacrifice the choice is obvious.


" Sometimes you must cull the weak to preserve the Strength of the Herd!" - Dr. Thraxx of the Global Liberation Army.

Seconded by Emperor Palpatine, Voldemort, Thanos, and other particularly pragmatic politicians.

On a more serious and mature note, this is ridiculous. A good, fair strategy game should be giving everyone a fair chance and a good time. Sacrificing ten percent of the player base or any percent for " The Greater Good" is ridiculous.

I mean by that reasoning up to 20 percent, or 30-40 percent can be sacrificed. This is absurd for a board game that's supposed to be fun and fairness is a large part of this. I can accept some imbalance for a little while for a new faction, especially if it's played by a minority and only possibly OP, but to say things like " Well sometimes you have to leave Grandma behind so the rest of us can make the trail" about players in a board game sounds like something from a cartoon.

This message was edited 2 times. Last update was at 2022/11/02 00:45:35


 
   
Made in us
Longtime Dakkanaut




Asenion wrote:
Wait are people actually saying things like 10 percent of the player base should be sacrificed so 90 percent of players can have a good time? Geez.


Automatically Appended Next Post:
Aecus Decimus wrote:
Dudeface wrote:
Excellent, better have a faction with 10% win rate than 90% right?


Yes. A faction with a 10% win rate makes things miserable for a small subset of players: the people who play that specific faction and do not have any other army. A faction with a 90% win rate makes things miserable for everyone except that faction. One of these groups is much larger than the other and if I have to choose which one to sacrifice the choice is obvious.


" Sometimes you must cull the weak to preserve the Strength of the Herd!" - Dr. Thraxx of the Global Liberation Army.

Seconded by Emperor Palpatine, Voldemort, Thanos, and other particularly pragmatic politicians.

On a more serious and mature note, this is ridiculous. A good, fair strategy game should be giving everyone a fair chance and a good time. Sacrificing ten percent of the player base or any percent for " The Greater Good" is ridiculous.

I mean by that reasoning up to 20 percent, or 30-40 percent can be sacrificed. This is absurd for a board game that's supposed to be fun and fairness is a large part of this. I can accept some imbalance for a little while for a new faction, especially if it's played by a minority and only possibly OP, but to say things like " Well sometimes you have to leave Grandma behind so the rest of us can make the trail" about players in a board game sounds like something from a cartoon.

Nobody was advocating for 10% win rates. But if faced with a 10% or 90% a 10% would be healthier for the game... obviously the goal is between 45-55 as stated multiple times by GW

This message was edited 2 times. Last update was at 2022/11/02 01:09:16


 
   
Made in us
Regular Dakkanaut




Asmodios wrote:
Asenion wrote:
Wait are people actually saying things like 10 percent of the player base should be sacrificed so 90 percent of players can have a good time? Geez.


Automatically Appended Next Post:
Aecus Decimus wrote:
Dudeface wrote:
Excellent, better have a faction with 10% win rate than 90% right?


Yes. A faction with a 10% win rate makes things miserable for a small subset of players: the people who play that specific faction and do not have any other army. A faction with a 90% win rate makes things miserable for everyone except that faction. One of these groups is much larger than the other and if I have to choose which one to sacrifice the choice is obvious.


" Sometimes you must cull the weak to preserve the Strength of the Herd!" - Dr. Thraxx of the Global Liberation Army.

Seconded by Emperor Palpatine, Voldemort, Thanos, and other particularly pragmatic politicians.

On a more serious and mature note, this is ridiculous. A good, fair strategy game should be giving everyone a fair chance and a good time. Sacrificing ten percent of the player base or any percent for " The Greater Good" is ridiculous.

I mean by that reasoning up to 20 percent, or 30-40 percent can be sacrificed. This is absurd for a board game that's supposed to be fun and fairness is a large part of this. I can accept some imbalance for a little while for a new faction, especially if it's played by a minority and only possibly OP, but to say things like " Well sometimes you have to leave Grandma behind so the rest of us can make the trail" about players in a board game sounds like something from a cartoon.

Nobody was advocating for 10% win rates. But if faced with a 10% or 90% a 10% would be healthier for the game... obviously the goat is between 45-55 as stated multiple times by GW


Healthier is questionable here. That's like saying Cancer is Healthier then AIDS.

Technically the K-T Meteor Strike was better then the Methane Explosion of the Great Dying, but that's far from ideal. Certainly the developers and community for a board game, meant for fun with nigh endless resources can do better then that!

This message was edited 2 times. Last update was at 2022/11/02 01:15:01


 
   
Made in us
Longtime Dakkanaut




Asenion wrote:
Asmodios wrote:
Asenion wrote:
Wait are people actually saying things like 10 percent of the player base should be sacrificed so 90 percent of players can have a good time? Geez.


Automatically Appended Next Post:
Aecus Decimus wrote:
Dudeface wrote:
Excellent, better have a faction with 10% win rate than 90% right?


Yes. A faction with a 10% win rate makes things miserable for a small subset of players: the people who play that specific faction and do not have any other army. A faction with a 90% win rate makes things miserable for everyone except that faction. One of these groups is much larger than the other and if I have to choose which one to sacrifice the choice is obvious.


" Sometimes you must cull the weak to preserve the Strength of the Herd!" - Dr. Thraxx of the Global Liberation Army.

Seconded by Emperor Palpatine, Voldemort, Thanos, and other particularly pragmatic politicians.

On a more serious and mature note, this is ridiculous. A good, fair strategy game should be giving everyone a fair chance and a good time. Sacrificing ten percent of the player base or any percent for " The Greater Good" is ridiculous.

I mean by that reasoning up to 20 percent, or 30-40 percent can be sacrificed. This is absurd for a board game that's supposed to be fun and fairness is a large part of this. I can accept some imbalance for a little while for a new faction, especially if it's played by a minority and only possibly OP, but to say things like " Well sometimes you have to leave Grandma behind so the rest of us can make the trail" about players in a board game sounds like something from a cartoon.

Nobody was advocating for 10% win rates. But if faced with a 10% or 90% a 10% would be healthier for the game... obviously the goat is between 45-55 as stated multiple times by GW


Healthier is questionable here. That's like saying Cancer is Healthier then AIDS.

A 90% win rate faction impacts every other faction in the game negatively. a 10% would only hurt its individual player base. Luckily according to the most recent data the lowest win rate faction had a 39% which is bad but a far cry from 10%
   
Made in us
Regular Dakkanaut




Please don't spam the forum.

This message was edited 1 time. Last update was at 2022/11/02 17:21:35


 
   
 
Forum Index » 40K General Discussion
Go to: