I'm popping in to eat my crow. I thought that Sisters had the tools to overcome their limitations and it's very clear that they don't. I'm still a little curious to see if they work in any kind of Imperial allies lists as the cheap objective holders but I suspect that if such a list was good we'd be seeing it make waves by now.

The Repressor is so wildly superior to anything else in the index it's kind of infuriating to be honest.

Any big tournaments doing legends anyone know about?

Here's something to ponder. According to 40K Stats:

Adepta Sororitas 10th Ed Win Rate: 48.04% over 1403 Games

Adepta Sororitas Leviathan GT Win Rate: 43.56% over 225 Games

Adeptus Mechanicus 10th Ed Win Rate: 33.37% over 833 Games

Adeptus Mechanicus Leviathan GT Win Rate: 47.95% over 146 Games

What the heck is going on here? AS win rate drops in GT format while AM increased by nearly 50%

 Mr Morden wrote:

 Totalwar1402 wrote:
Pretty easy to solve melta. Just say it gets anti vehicle 4 plus within melta range.

But this is the issue. You can’t make a change like that without doing it to everybody because the melta is a common weapon.

Technically they can do exactly that as weapons are datasheet based.

They can, but they shouldn't. It doesn't really fix the issue - which is that so many units were "built" with melta as a/the tank killer option for the unit and/or the entire army - and fixing it just for sisters will be both a tacit admission they screwed up, and a giant middle finger to everyone they don't fix it for. We only have to back track to the posts from the origins/precursors of the "Melta" rule and Firedragons not getting that boost to see a preview of those results. The players are as unlikely to care about the difference between a Sisters of Battle Holy Melta vs a Space Marine Multi Melta as they were about the difference between a Meltagun and a Firepike or fusion gun.


Automatically Appended Next Post:

 alextroy wrote:
Here's something to ponder. According to 40K Stats:

Adepta Sororitas 10th Ed Win Rate: 48.04% over 1403 Games

Adepta Sororitas Leviathan GT Win Rate: 43.56% over 225 Games

Adeptus Mechanicus 10th Ed Win Rate: 33.37% over 833 Games

Adeptus Mechanicus Leviathan GT Win Rate: 47.95% over 146 Games

What the heck is going on here? AS win rate drops in GT format while AM increased by nearly 50%

Experience in build/play styles and vs. Target Priority?

 alextroy wrote:
Here's something to ponder. According to 40K Stats:

Adepta Sororitas 10th Ed Win Rate: 48.04% over 1403 Games

Adepta Sororitas Leviathan GT Win Rate: 43.56% over 225 Games

Adeptus Mechanicus 10th Ed Win Rate: 33.37% over 833 Games

Adeptus Mechanicus Leviathan GT Win Rate: 47.95% over 146 Games

What the heck is going on here? AS win rate drops in GT format while AM increased by nearly 50%

Player skill? It might be a difference of who was playing where & who/what they played against. And how.
Just because you're using x faction does NOT guarantee you x% chance of winning.....

 alextroy wrote:
Here's something to ponder. According to 40K Stats:

Adepta Sororitas 10th Ed Win Rate: 48.04% over 1403 Games

Adepta Sororitas Leviathan GT Win Rate: 43.56% over 225 Games

Adeptus Mechanicus 10th Ed Win Rate: 33.37% over 833 Games

Adeptus Mechanicus Leviathan GT Win Rate: 47.95% over 146 Games

What the heck is going on here? AS win rate drops in GT format while AM increased by nearly 50%

Ugh. See I am worried they didn’t mention Sisters in the metawatch video.

I’d question where that win rates coming from when you can get 10 Black Templars for near enough 10 Sisters of Battle. Objectively a balanced or fluffy army should lose against a fluffy marine army. Not sure how you’re an MSU army if you aren’t even half the points of a marine squad. People are probably taking the most tailored list they can and exploiting the way objectives are generated and the terrain.

So I can see them staying the course and thinking the “MSU”, you might win the game by hiding in the terrain stops the opponent attacking you and you get tabled. I don’t really count that as a win TBH. I am old fashioned, you start balance with two armies in an open field. You shouldn’t make assumptions about terrain and the objectives.

Saying, “oh well in a game of 40K it should be so packed with LOS blocking terrain that all those long range guns can’t shoot most of the time” or “we’ll count your game as a win even if you get tabled, you get cards that give you points for doing nothing and your opponent doesn’t lose any models”. Like I don’t really care too much what the cards say, if you have a bad game then it’s a bad game even if I am moving stuff around in my back line and deep striking or redeploying stuff arbitrarily your opponent can’t always counter. I can see them using that as an excuse not to look lethality or durability or relative points costs.

Plus they might all just be spamming Exorcists and arco flaggelants.

Depends how they classify armies.

Right now I’d think the best way to play SB is to go all in on the allies. I don’t know GT rules but I’d guess there are a lot of SB “armies” that don’t actually have much SB in them.

Can you even do allies, outside Agents of the Imperium?

You can take a specific number of Agents of the Imperium characters & units along with 1 Imperial Knight unit.

 JNAProductions wrote:
Can you even do allies, outside Agents of the Imperium?

They can do one big knight or a couple small ones I think. A discount Aeldari thing. Sort of.

 alextroy wrote:
Here's something to ponder. According to 40K Stats:

Adepta Sororitas 10th Ed Win Rate: 48.04% over 1403 Games

Adepta Sororitas Leviathan GT Win Rate: 43.56% over 225 Games

Adeptus Mechanicus 10th Ed Win Rate: 33.37% over 833 Games

Adeptus Mechanicus Leviathan GT Win Rate: 47.95% over 146 Games

What the heck is going on here? AS win rate drops in GT format while AM increased by nearly 50%

Admech have some okay units in the dex, the Kataphrons mainly.

Ad Mecha to work need an expert pilot playing them and a very specific build to be used. In case of SoB the level of player has less impact on win rates.

It shows in some of the good armies too. GSC win rates for tournaments, and over all are not the same. With a much bigger spread then lets say with eldar.

 JNAProductions wrote:
Can you even do allies, outside Agents of the Imperium?

Isn't it slightly different to 9th edition where you can take any Imperium/Chaos unit but you lose all detachment abilities?

Jarms48 wrote:

 JNAProductions wrote:
Can you even do allies, outside Agents of the Imperium?

Isn't it slightly different to 9th edition where you can take any Imperium/Chaos unit but you lose all detachment abilities?

Usually what happens is you keep your Detachment abilities, but most detachment abilites only work on your main faction through faction/keyword limitations - so allies can't be targeted by Detachment Strats, just the BRB ones. There are also generally some pretty tight points/unit count restrictions. I can't find/remember which free PDF had the rules for a knight allied contingent - but I seem to remember that one being points based (up to 500 Points so 1 big, or two small) vs Unit Count like for Imperial Agents (2 and 2) Oop - there it is Freeblades in the Knight PDF 1 Titanic, or 3 Armigers

Big knights without rotating ion shields and even more their Knight Code. Armigers could be good, if they were -2AP.

Three Warglaives with Thermal Spears makes up for a lot of Sisters Anti-Tank issues.

Well, hides the issues. Armigers aren't sisters so it's not really a fix.

In my expiriance 3 of the smaller knights in to anything which is basic army now, just dies. GSC have numbers and fire power to kill big knights, and the only saving grace of the faction is that it is both expensive and few people owned it, at least in places where people don't look for leaked rules.

Eldar just blow up everything. Custodes can actualy melee kill the small knights. On top of that, because people don't like towering, terrain becomes very hard on non LoS ignoring or super fast stuff. SoB are just too expensive, too slow, with not efficient enough melee and with core mechanics designed for a horde army. If SoB had the GSC numbers and super powerful melee units of some sort, then they would work.

The real problems are going to start, in 1-2 months, when even the sunday gamers are going to catch up to people playing 10+ games a week. Then no army that is mid or bad is going to work. But well at least we have the first army, at least to my knowladge, that burst the 90% win rate.

Karol wrote:
In my expiriance 3 of the smaller knights in to anything which is basic army now, just dies. GSC have numbers and fire power to kill big knights, and the only saving grace of the faction is that it is both expensive and few people owned it, at least in places where people don't look for leaked rules.

Eldar just blow up everything. Custodes can actualy melee kill the small knights. On top of that, because people don't like towering, terrain becomes very hard on non LoS ignoring or super fast stuff. SoB are just too expensive, too slow, with not efficient enough melee and with core mechanics designed for a horde army. If SoB had the GSC numbers and super powerful melee units of some sort, then they would work.

The real problems are going to start, in 1-2 months, when even the sunday gamers are going to catch up to people playing 10+ games a week. Then no army that is mid or bad is going to work. But well at least we have the first army, at least to my knowladge, that burst the 90% win rate.

Last top 5 GT list for Sisters was 3 armigers.

Interestingly, people are still trying to make garbage units like Sacresants and Dominions work. Once people accept that it's Exorcist+Seraphim.Index, I think the winrate will level out to about 40% and sit there until the next rebalance.

Not so sure about that in non-tourney games. We've got repressors datacards now, wo anyone who kept their Repressor spam is going to be able to get doms punching above their weight again.

PenitentJake wrote:
Not so sure about that in non-tourney games. We've got repressors datacards now, wo anyone who kept their Repressor spam is going to be able to get doms punching above their weight again.

Nothing that happens in non-tourney games matters in relation to unit balance.

Try to go tell that to 4 dudes that just started and 2 have 40-50% win rate lists, one has a sub 40% win rates list and the 4th plays a curb stomper like knights or custodes.
Balance internal and external is essential especialy at the non tournament level. At the tournament level the problem of 27% win rate Votan isn't a problem, because if you remove fresh players from the rankings there is exacly 0 people playing them in tournaments. The same with too good armies. 90% win rate of veteran players with GSC or +70% of eldar, what problem is that for people that are trying to win events? It is a non problem.
Problem at event level are local FAQs, what ever you can get models painted on time and the ones you want etc.

The main problem for most custodes or IK players right now, isn't what ever they are good or bad, but what ever they can get their hands on 10 arbitors.

ERJAK wrote:

PenitentJake wrote:
Not so sure about that in non-tourney games. We've got repressors datacards now, wo anyone who kept their Repressor spam is going to be able to get doms punching above their weight again.

Nothing that happens in non-tourney games matters in relation to unit balance.

A bold statement considering that less than 10% of the playerbase compete in tournaments. Many countries where GW products are sold don't even have official tournaments.

Maybe: "Nothing that happens in non-tourney games matters in relation to balance for (Dakkanauts, tourney players, the internet, you personally, etc)" would have been an easier argument to support.

Of course, it's fair to point out that being an OOP FW unit probably DOES mean that only the tourney-minded will have access to it, and you are certainly correct that this won't make a difference to that subset of the player base, so I suppose there is some validity to your point.

ERJAK wrote:
Nothing that happens in non-tourney games matters in relation to unit balance.

Strongly disagree, but GW does not have the means in place to get any other data.

They don't need much data. If new GSC players are draggint the veteran players win rates of +90% to a measly +70%, then one expect that the unprepared FLGS games enjoyer will not have much fun playing against the army, piloted by a good or new player. On the flip side 27% win rates of Votan, and 0% non new GT players tell GW, that absolutly no one in the tournament circle thought that Votan are worth bringing to an event. And I would like to point out that GWs own last big tournament had a dude that brough minimum GK and a thunder hawk to the event. Votan don't even the "for fun" players, because right now they are not fun.

If ad mecha win rates and participation are in the dumps, and the only wins come from veteran players playing a list that looks like a clone of other winning lists, then how much fun will a non tournament, non optimised list, non veteran player have?

GW doesn't need to have all the casual data in the world, or even any casual data, to have an idea what the impact of outliers is. Now they may struggle to find out how it is for the around 50% win rate, because here the difference between and old marine army with tacticals and a triple desolotor squad can be huge.
But GW knows the sales, they know how the tournament lists look like and they know much many units of stuff they produced and sold. If the tournament lists can only be run, by lets say 1000 people world wide and everyone else would have to play the left overs. Then the casual player may not have the access to some of those obligatory things, needed to run certain armies. Custodes or knights without an assains and arbitors are an army that can lose turn one, if they push fails.

Playing marines and being unable to buy desolators, may not be very fun in to good armies and people that could get their stuff.

As I said some place else, Jimmy with his christmas box of custodes is going to wreck face in his casual circle of "just stared w40k and 10th" guys.

a_typical_hero wrote:

ERJAK wrote:
Nothing that happens in non-tourney games matters in relation to unit balance.

Strongly disagree, but GW does not have the means in place to get any other data.

It seems like you do agree then - GW can't/don't collect non-tourney data (besides whatever passes for their in-house playtesting (HA!)), so those games don't matter when it comes to whatever GW does concerning unit balance.

If you want to balance your game properly, you need more data than just faction attendance for events and victorious army lists.

Just as an example, there is currently no way for GW to track the "MMR" (for the lack of a better word) of any given player. The publicly available information of tournament factions and placements is tiny compared to all the unreported games being played everyday.

GW can make adjustments on what gets reported or complained about, but that is just a matter of "the squeaky wheel gets the grease". If nobody ever plays LoV and we have no data on any of their units in a balls to the walls competitive environment, where both players know what they are doing, how well designed can a possible buff be? Especially given GW's track record showing their lack of understanding how their own game is being played by competitive players.

I think this is one of the reasons why the balance pendulum swings constantly and why it swings hard at times. Lack of data to base a decision on.

 Lord Damocles wrote:

a_typical_hero wrote:

ERJAK wrote:
Nothing that happens in non-tourney games matters in relation to unit balance.

Strongly disagree, but GW does not have the means in place to get any other data.

It seems like you do agree then - GW can't/don't collect non-tourney data (besides whatever passes for their in-house playtesting (HA!)), so those games don't matter when it comes to whatever GW does concerning unit balance.

No, casual data is important and should be collected. I merely acknowledge that GW currently does not possess the ability to do so.

a_typical_hero wrote:
No, casual data is important and should be collected. I merely acknowledge that GW currently does not possess the ability to do so.

Is there an intelligent way to do this?

Not asking you to design the whole system, just have a conceptual model of how this could be done that prevents:
1) malicious data manipulation
2) unintelligible results that require large amounts of work to parse (e.g. text input feedback fields).

 Unit1126PLL wrote:

a_typical_hero wrote:
No, casual data is important and should be collected. I merely acknowledge that GW currently does not possess the ability to do so.

Is there an intelligent way to do this?

Not asking you to design the whole system, just have a conceptual model of how this could be done that prevents:
1) malicious data manipulation
2) unintelligible results that require large amounts of work to parse (e.g. text input feedback fields).

It's not infallible, but you could integrate score-keeping and game tracking functionality similar to the ITC Battles app into the official 40K app.

Useful for getting a larger picture of army balance as well as secondaries and rates of CP expenditure. If it's linked into the army builder, you can dig a little deeper to see stuff like Eldar win rates with and without a Wraithknight.

 Unit1126PLL wrote:
Is there an intelligent way to do this?

Not asking you to design the whole system, just have a conceptual model of how this could be done that prevents:
1) malicious data manipulation
2) unintelligible results that require large amounts of work to parse (e.g. text input feedback fields).

Let's start with number 2) first, as I think this is easier or more straight forward to do:
- Make a proper list building app and website, so you have exact unit configuration saved while providing an incentive for the user to do it.
- The app could assist during the game so you can keep track of the overall score and all secondaries.
- Submit the result at the end of the game to a server for parsing.
- The app could assign a "MMR" to you which gets higher if you win against other people of the same or higher MMR than you and gets lower if you lose, or stays the same if you continuously play against the same people.

1) is harder to get right, as there is simply no way of preventing any malicious use while keeping it accessable enough. You could, however, do something like this:
- After you submit your battle data, generate a (QR) code for the other player to enter, so the result gets confirmed and the armies get paired.
- Introduce different kind of "trust levels" by which you can filter the results. Something like "Garagehammer (only confirmed by participating players)", "Bunker alliance" for games taking place in affiliated clubs and stores and finally "Tournament and Warhammer shops confirmed". So for Bunker and Tournament you need something like a third party confirmation that the game did indeed happen like this.

Those are good ideas actually, appreciate the insights.

Not flawless ofc, but conceptually sound as far as I am willing to analyze them.

a_typical_hero wrote:
If you want to balance your game properly, you need more data than just faction attendance for events and victorious army lists.

Just as an example, there is currently no way for GW to track the "MMR" (for the lack of a better word) of any given player. The publicly available information of tournament factions and placements is tiny compared to all the unreported games being played everyday.

GW can make adjustments on what gets reported or complained about, but that is just a matter of "the squeaky wheel gets the grease". If nobody ever plays LoV and we have no data on any of their units in a balls to the walls competitive environment, where both players know what they are doing, how well designed can a possible buff be? Especially given GW's track record showing their lack of understanding how their own game is being played by competitive players.

IMO, the fact that no non new players play Votan in tournaments is a HUGE sign on the quality of rules. Same with win rates. If GW sees that veteran players using GSC are having, at lower attendance then eldar, comperable or higher win rates and that new GT players are super swingy as far as wins/loses goes. Then there is a ton of data to get from this. First of all this means the army is, in proper hands, too powerful. Second if it is being dragged down to "only" 76% win rates by the "noobs", and then you find out that the "noobs" lose mirrors and to other top armies and still stomp the mid and bottom tier armies, then the army is really too powerful. Because the casual player is going to be that mid to low skilled player, with potentialy a not fully optimised list.

From other armies win rates GW can learn other stuff. Veteran playing ad mecha win games, in 9th they won games post nerfs even. But the noobs didn't. Lesson from that, army is powerful, but not too powerful, as the avarge player is not going to be able to play it.

No one playing faction X, and I mean like really non one, when it has a codex, is a sign that the ball was dropped somewhere writing the rules. The rest can only be checked with outside house testing, because the DT is clearly biased in how they write rules. And it is even technicaly done for GW. All players test the game post release for GW. They even pay for the opportunity to participate in the tests. The problem comes from how GW fixes problems.
They try to fix specific units or rule problems with points hikes, which at best can achive a dead unit/army/faction, which isn't much of a fix for the FLGS player.

A lot of the stuff GW knows, but as long as it doesn't make them that "minimum 100k" they will just not do the change, because it isn't worth it for them. Plus the design team lives in this limbo, when for the problems we have right now, they have a fix (like a codex or seson detachment/secondaries change) they play in house now, but we will play in maybe 6-12 months. This gives them that odd what is the problem perspective, because to them faction X, Y or Z being too good or too bad, is already fixed. It is just isn't fixed to the players/buyers of their game.

 Lord Damocles wrote:

a_typical_hero wrote:

ERJAK wrote:
Nothing that happens in non-tourney games matters in relation to unit balance.

Strongly disagree, but GW does not have the means in place to get any other data.

It seems like you do agree then - GW can't/don't collect non-tourney data (besides whatever passes for their in-house playtesting (HA!)), so those games don't matter when it comes to whatever GW does concerning unit balance.

GW can barely use the results from tournament games effectively. Collecting data from casual games would be more disastrous.

Some of the other games I play have player-run databases that assign you an Elo Score (or some bespoke version of an Elo). All they require is a unique number or name for each player, and they keep track of the performance of both the player and the army -- in some cases, they even keep track of the individual army rosters. I know nothing about computers but I imagine such a database should be fairly easy to create, if vulnerable to dishonest reporting.

asoiaf-stats.com and artdelaguerre.fr are two examples of such sites.

 Saber wrote:
if vulnerable to dishonest reporting.

And that's the fatal problem with it. A database like that only works for major tournaments where honesty is enforced, anything else has worse than zero value in balance or player strength analysis.


Automatically Appended Next Post:

PenitentJake wrote:
A bold statement considering that less than 10% of the playerbase compete in tournaments.

Bold maybe, but 100% accurate. Tournament games are the only context where you have both reliable data and games with symmetrical missions and both players prioritizing winning the game over any other concerns. A story-focused game about, say, a last stand by the SoB against a horde of Tyranids may be fun to play but it doesn't tell you anything about balance.

Of course, it's fair to point out that being an OOP FW unit probably DOES mean that only the tourney-minded will have access to it

Exactly the opposite. Tournament players don't care about OOP FW stuff that doesn't have real rules, if they happen to own any of it they're likely to cash it in on ebay. The people who have those models are painters and narrative-focused players like you, who prioritize having the cool model even if it doesn't have rules and can't be used outside of your heavily house ruled narrative games.

Even if you got perfect data from the app one of the biggest issues is the rock-paper-scissors nature of 40k match-ups. 40k, outside of the TAC tournament meta, can skew horribly and I don't think it would be a desirable trait to allow an anti-infantry skew list to have even a 40% win rate into a competent list that skews into tough units/vehicles. The amount of data processing required would be immense and the number of games reported would be too low to give great data.

I've suggested a full-on 40k simulator that GW could gather data from but as people pointed GW is very much not a tech company.

There is a super easy solution for your concerns and I already gave the answer in my last post.

Introduce "ranked" 40k games with adjusting MMR values for each player. Both players have to agree on it beforehand, of course, but this way you
a) create the incentive to win
b) can filter the results by skill level
c) enforce a standardised way of play (so results are comparable)*

*provide a "map pack" for matched play with different pre defined layouts.

You can still gather the data for all other games and let players use the app to help during the game and with list building, but those "open" games could be filtered in or out as needed.

a_typical_hero wrote:
There is a super easy solution

Both players have to agree on it beforehand

You and I define super easy very differently.

Sorry to interrupt a thread about Sisters with a xenos diversion again, but this won't stand unanswered:

 Void__Dragon wrote:

 Insectum7 wrote:
What a silly argument. Tyranid Warriors say hi. And there are billions of them.

I wonder if you could provide a citation proving that there are billions of Tyranid Warriors?

Sure:



In short, Tyranid Warriors form "the heart" of swarms of billions of Tyranid creatures. And while the initial quote of "untold* millions" is vague, by the numbers if Tyranid Warriors ("the heart of swarms") represented only one-in-a-million Tyranid organisms, that's in the billions ("a million ships each with billions of organisms") or even tens or much more ("a billion times a billion"), because despite numerous hive fleets in the galaxy, there are still more coming from the intergalactic void. But hey, even if you don't accept that, "untold millions" is still means plenty more Tyranid Warriors than there are Marines.

*untold - meaning "too much or too many to be counted or measured." by the online dictionary. So not like . . . 10. Many many.

 Void__Dragon wrote:
Or even that they are more formidable than Marines man per man.

Sure:

In every edition they've held better stats in the Attacks, Wounds and Ld. department, Marines having a better BS and Save. In all the editions post-1st but pre 8th Warriors tended to have better WS at 5. When Initiative was a thing, the Warriors had higher. In 1st ed, interestingly, Marines had higher Strength. But at the same time the Tyranid forces had to be made up of at least half Zoats, which were far superior to a Marine in stats. Amusingly, Warriors were also smarter than Marines In 2nd ed, Warriors were much faster with a movement of 6 vs the Marine 4. At worst post RT, a Tyranid Warrior still held higher Attacks and more Wounds, but could then be upgraded to (and beyond!) their previous levels of power. In 2nd ed, when psychology was a thing, Marines had to test for Fear against Warriors. The latest editions have just confirmed their stat superiority, updating them to T5 S5, and even the Ranged-Warrior boasts 5 Attacks with its generic claws. In short, Tyranid Warriors have consistently been portrayed as some combination of bigger, stronger, tougher, faster, and nastier in CC than typical Marines. Marines are better shots, but the basic weapons like Tyranid Deathspitters were/are better than Bolters. The primary advantage Marines have had is their armor. Your typical Marine will lose to a typical Tyranid Warrior in either a cage fight or a firefight.

 Void__Dragon wrote:
They certainly didn't look particularly formidable when Titus ripped one's head off of its shoulders with nothing but his own brute strength in the Space Marine 2 trailer.

"I saw a thing in a trailer once." does not make for good data. Is Titus representative of a typical Marine? Is the trailer unbiased in its depiction? Did he just roll sixes?

Poor argument.

 Void__Dragon wrote:

Also, Necrons could be better individually than Marines, with access to teleportation, and still not auto-win conflicts. Because in case you forget, the Imperium has numerous fighting forces other than the Marines, such as the Guard, the Navy, the Mechanicum, the Sororitas etc. Plus, the Necrons have their own alternate conflicts that require attention.

What does that have to do with anything?

It shows your extremely poor understanding of how conflicts can play out. Technical and numerical superiority don't mean auto-win when there are other variables at play.

 Void__Dragon wrote:

No idea.

Because people like you can't help but respond to nine day old posts apparently.

The magnitude of your wrongitude is so great that it beckons across time and space. In short, the idea that no faction can have a superior-to-marine type of unit and also vastly outnumber marines in the lore. . . is total bunk!


Thank you for your patience everyone! You may now return to your regularly scheduled Sisters thread.

a_typical_hero wrote:
There is a super easy solution for your concerns and I already gave the answer in my last post.

An answer that does nothing to address the issues with bad data from fraudulent games.

Breton wrote:
You and I define super easy very differently.

"Ranked" is basically the lowest common denominator for pick up games already. And as I already wrote: Different kind of trust/quality levels for results.
At the low end you have "not confirmed by other player, only one army list and wether it was a victory or defeat" and at the other end a "confirmed by authorised and trusted third party, both army lists, final score, secondaries by round".

 ThePaintingOwl wrote:
An answer that does nothing to address the issues with bad data from fraudulent games.

Yes I did. Trust levels.

a_typical_hero wrote:

Breton wrote:
You and I define super easy very differently.

"Ranked" is basically the lowest common denominator for pick up games already. And as I already wrote: Different kind of trust/quality levels for results.
At the low end you have "not confirmed by other player, only one army list and wether it was a victory or defeat" and at the other end a "confirmed by authorised and trusted third party, both army lists, final score, secondaries by round".

If the low end is "garbage data that is easily faked" why collect that data at all?

Time-delayed confirmation of the result.

One person does not have their mobile with them, or is only using the website, or there is no internet.

Edit: And it is probably more work to build in restrictions. But if not, then don't collect it. What does it matter? The idea does not stand and fall with the detail on how to collect the low end data, as you can always filter the results.

a_typical_hero wrote:
What does it matter?

Ever hear the phrase "garbage in, garbage out"?

a_typical_hero wrote:
There is a super easy solution for your concerns and I already gave the answer in my last post.

Introduce "ranked" 40k games with adjusting MMR values for each player. Both players have to agree on it beforehand, of course, but this way you
a) create the incentive to win
b) can filter the results by skill level
c) enforce a standardised way of play (so results are comparable)*

*provide a "map pack" for matched play with different pre defined layouts.

You can still gather the data for all other games and let players use the app to help during the game and with list building, but those "open" games could be filtered in or out as needed.

Chess has an ELO system, people still screw around and don't take games seriously. Just because a game is ranked doesn't mean that players will play to the best of their ability, that it isn't an outlier in terms of hot/cold dice, that there aren't rules misplayed, that false data wasn't submitted etc.

The only way to gather clean enough data would be via a fully digital boardgame that enforces rules perfectly.

EviscerationPlague wrote:

 Lord Damocles wrote:

a_typical_hero wrote:

ERJAK wrote:
Nothing that happens in non-tourney games matters in relation to unit balance.

Strongly disagree, but GW does not have the means in place to get any other data.

It seems like you do agree then - GW can't/don't collect non-tourney data (besides whatever passes for their in-house playtesting (HA!)), so those games don't matter when it comes to whatever GW does concerning unit balance.

GW can barely use the results from tournament games effectively. Collecting data from casual games would be more disastrous.

Also, this isn't league of legends where EVERY game has to follow the same rules. Trying to collect data from a beer and pretzels game where your eldar buddy spots you 200pts, or where you use legends units or illegal wargear options from 5th edition, would be completely pointless.

Non-tournament games DO NOT matter when it comes to unit balance because it's not even the same game.


Automatically Appended Next Post:

a_typical_hero wrote:
There is a super easy solution for your concerns and I already gave the answer in my last post.

Introduce "ranked" 40k games with adjusting MMR values for each player. Both players have to agree on it beforehand, of course, but this way you
a) create the incentive to win
b) can filter the results by skill level
c) enforce a standardised way of play (so results are comparable)*

*provide a "map pack" for matched play with different pre defined layouts.

You can still gather the data for all other games and let players use the app to help during the game and with list building, but those "open" games could be filtered in or out as needed.

I like how you're trying to pass the wargaming equivalent of discovering the holy grail and then using it to do cold fusion and also capture the higgs boson, as 'super easy'.

 ThePaintingOwl wrote:
Ever hear the phrase "garbage in, garbage out"?

It is literally as complicated as ticking a checkbox which data to include. I like how you comfortably ignore the delayed confirmation part where a bad dataset can have a higher quality later on, which is a reasonable enough use case to keep the functionality to report one sided results.

 Canadian 5th wrote:
Chess has an ELO system, people still screw around and don't take games seriously. Just because a game is ranked doesn't mean that players will play to the best of their ability, that it isn't an outlier in terms of hot/cold dice, that there aren't rules misplayed, that false data wasn't submitted etc.

The only way to gather clean enough data would be via a fully digital boardgame that enforces rules perfectly.

And all kind of competitive online games got trolls, cheater and people going afk. I still trust the LoL analytics websites with what champ and item got a good winrate. And again, if you don't want to base your arguments on a given dataset, pick a high enough MMR and/or trust level. "No data is viable because of uncontrollable luck/misplays/x outside of a simulation" is demonstrably false, as the data we currently do have (tournament results) is consistent with actual game outcomes.

ERJAK wrote:
I like how you're trying to pass the wargaming equivalent of discovering the holy grail and then using it to do cold fusion and also capture the higgs boson, as 'super easy'.

Are you living in bizarro world where "2k matched play, latest MFM" isn't the norm for pick up games at a store? Great, so the biggest part is already mutually agreed upon. "Can I get your code to confirm our game? / Can you please use this code and confirm later when you are back home?" really sounds like " trying to pass the wargaming equivalent of discovering the holy grail and then using it to do cold fusion and also capture the higgs boson" /s

Most of your guys arguments boil down to "add another checkbox so the result can be filtered in or out as needed".


Edit: We could continue coming up with theoretical problems and theoretical solutions for some time now, but GW is likely never going to do something similar, so it is a mood topic. The core of the argument is that, in my opinion, casual data matters and should be collected and considered.

Even if I play super random at home with brothers/some friends, I agree balance in 40k shouldnt be around us filthy casuals.

BUT, changes MUST be made. Its crazy nuts a devastator unit is similar price to retributors. Or how our best vehicle, the Castigator, is same price but absolute trash other than +1T when compared to an Eldar Fire Prism with 1 full reroll, much better mobility, 2 types of ammo, and vast superior antitank while having similar anti-infantry capabilities.

Yes, all factions have their strength and weaknesses, but those differences are just too high that actively hurt even the most casual of games with friends.

My last game against my eldest brother as tyranids (his first game) we just saw a Carnifex is same price as an Haruspex, the latter being vastly superior in damage and much tougher. Or 5 Bargaunts are 15 points cheaper than 3 guarsmen mortar while hitting more accurately.

The balance is just wild.

a_typical_hero wrote:
It is literally as complicated as ticking a checkbox which data to include. I like how you comfortably ignore the delayed confirmation part where a bad dataset can have a higher quality later on, which is a reasonable enough use case to keep the functionality to report one sided results.

And the point is that under no circumstances would GW ever want to tick the "include the garbage data" checkbox so there is no point in including that data. Anything other than organized events is garbage data and organized event data will all be submitted and processed as a single data set once the event concludes.

"Can I get your code to confirm our game? / Can you please use this code and confirm later when you are back home?"

What does that accomplish? Having confirmation codes agreed on between two players doesn't touch the problem of players colluding to submit false data for personal benefit, or even a single player with two accounts submitting a bunch of games showing that {their favorite faction} always loses to {their most hated faction} and skewing GW's data in the direction that benefits them.

a_typical_hero wrote:
And all kind of competitive online games got trolls, cheater and people going afk. I still trust the LoL analytics websites with what champ and item got a good winrate. And again, if you don't want to base your arguments on a given dataset, pick a high enough MMR and/or trust level. "No data is viable because of uncontrollable luck/misplays/x outside of a simulation" is demonstrably false, as the data we currently do have (tournament results) is consistent with actual game outcomes.

Yes, but they also don't get accidental list errors, intentional bad data, games played using rules that aren't part of the game's code, and all the other errors that will render all the data outside of organized play (which is already tracked) completely useless. You simply can't get clean enough data by asking players to submit their own gameplay results unless you can control every variable outside of how players choose to play within the systems provided.

All it takes to break your "players exchange game codes" idea is for a relatively small number of users choosing to create accounts on multiple devices and then spamming data from games that were never played. Knowing GW it wouldn't even be hard to make a script to do this for you to get units you dislike buffed or nerfed. Even if it can't be automated given how few games of 40k are played each week even relatively small efforts will corrupt the data significantly.

They get plenty of casual info from their staff games over lunch and that seems to be the only "playtesting" that is done

 ThePaintingOwl wrote:

What does that accomplish? Having confirmation codes agreed on between two players doesn't touch the problem of players colluding to submit false data for personal benefit, or even a single player with two accounts submitting a bunch of games showing that {their favorite faction} always loses to {their most hated faction} and skewing GW's data in the direction that benefits them.

I remember when it was 'common knowledge' that everyone and their grandma was faking the results of games during the Eye of Terror campaign and that's why [insert faction here depending on conspiracy theory] won.

Hell, just a couple of weeks ago on this forum people were advocating faking the results of games to get GW to reveal Tyranid releases before Marines...

People would definitely falsify results to stick it to whichever faction's players it's cool to hate on this week.

This discussion is nice and all, but it doesn't have a lot to do with Adepta Sororitas. Maybe spin off to it's own thread?

And yes, any dataset gathered outside of highly controlled conditions is subject to corruption by bad actors. That doesn't mean you should throw up your hands and just do nothing. You do your best to insure some fidelity in your collection method and take your results with a grain of salt. It's not like somebody is going to open a million accounts so that they can pollute the data to enhance or tank a specific faction. Too much work for too little gain.

In other news, AS results over the last week are:

10th Edition Games: 47.96% over 269 games

Leviathan GT Games: 50.00% over 152 games

There appear to be ways to win game out there. On the other hand, our TiWP (4-0 Tournament appearances) is a pathetic 1.37% since June 29th. Sisters have no chance to win a tournament. However half the factions (14 of 28) are in the same or worst condition. GW has a lot of work to do.

 alextroy wrote:
This discussion is nice and all, but it doesn't have a lot to do with Adepta Sororitas. Maybe spin off to it's own thread?

And yes, any dataset gathered outside of highly controlled conditions is subject to corruption by bad actors. That doesn't mean you should throw up your hands and just do nothing. You do your best to insure some fidelity in your collection method and take your results with a grain of salt. It's not like somebody is going to open a million accounts so that they can pollute the data to enhance or tank a specific faction. Too much work for too little gain.

In other news, AS results over the last week are:

10th Edition Games: 47.96% over 269 games

Leviathan GT Games: 50.00% over 152 games

There appear to be ways to win game out there. On the other hand, our TiWP (4-0 Tournament appearances) is a pathetic 1.37% since June 29th. Sisters have no chance to win a tournament. However half the factions (14 of 28) are in the same or worst condition. GW has a lot of work to do.

I'd say that's signifanctly better than a lot of the comments in the run up to release made it sound like it would be as a bonus.

 alextroy wrote:

In other news, AS results over the last week are:

10th Edition Games: 47.96% over 269 games

Leviathan GT Games: 50.00% over 152 games

From where the hell is that data?

 alextroy wrote:
There appear to be ways to win game out there.

Yes, it's called playing pretty much Ministorum/Imperium Soup instead of Sisters.

 alextroy wrote:
On the other hand, our TiWP (4-0 Tournament appearances) is a pathetic 1.37% since June 29th. Sisters have no chance to win a tournament. However half the factions (14 of 28) are in the same or worst condition. GW has a lot of work to do.

Of course they can't win tourneys, especially if you play against IK you might as well just give up instead of wasting your time, and factions in even worse conditions are often those who can't just ally their way to higher win rates.

Dudeface wrote:
I'd say that's signifanctly better than a lot of the comments in the run up to release made it sound like it would be as a bonus.

And i'd say last weeks GT win rate of 35% according to Meta Monday and an overall GT win rate of 41% say it's not:

Sisters of Battle Index 10 19 55 0 35%(last week)0 41%(overall)

And it gets even more dire if we look at how one of the very few lists that managed a Top 5 placing(in a relatively small GT a week earlier) actually looked like:

CHARACTER

Daemonifuge (80 points) • 1x Ephrael Stern • 1x Bolt pistol 1x Sanctity • 1x Kyganil of the Bloody Tears • 1x The Outcast’s Weapons

Triumph of Saint Katherine (150 points) • Warlord • 1x Bolt pistols 1x Relic weapons

BATTLELINE

Battle Sisters Squad (110 points) • 1x Sister Superior • 1x Bolt pistol 1x Close combat weapon 1x Condemnor boltgun 1x Power weapon • 9x Battle Sister • 9x Bolt pistol 7x Boltgun 9x Close combat weapon 1x Meltagun 1x Multi-melta 1x Simulacrum Imperialis

OTHER DATASHEETS

Arco-flagellants (45 points) • 3x Arco-flails

Arco-flagellants (45 points) • 3x Arco-flails

Arco-flagellants (45 points) • 3x Arco-flails

Crusaders (20 points) • 2x Power weapon

Crusaders (20 points) • 2x Power weapon

Death Cult Assassins (35 points) • 2x Death Cult power blades

Exorcist (170 points) • 1x Armoured tracks 1x Exorcist missile launcher 1x Heavy bolter

Exorcist (170 points) • 1x Armoured tracks 1x Exorcist missile launcher 1x Heavy bolter

Exorcist (170 points) • 1x Armoured tracks 1x Exorcist missile launcher 1x Heavy bolter

Penitent Engines (60 points) • 1x Penitent flamers 1x Twin penitent buzz-blades

Penitent Engines (60 points) • 1x Penitent flamers 1x Twin penitent buzz-blades

Seraphim Squad (70 points) • 1x Seraphim Superior • 1x Close combat weapon 1x Plasma pistol 1x Power weapon • 4x Seraphim • 4x Bolt pistol 4x Close combat weapon 4x Ministorum hand flamer

Seraphim Squad (70 points) • 1x Seraphim Superior • 1x Close combat weapon 1x Plasma pistol 1x Power weapon • 4x Seraphim • 4x Bolt pistol 4x Close combat weapon 4x Ministorum hand flamer

Seraphim Squad (70 points) • 1x Seraphim Superior • 1x Close combat weapon 1x Plasma pistol 1x Power weapon • 4x Seraphim • 4x Bolt pistol 4x Close combat weapon 4x Ministorum hand flamer

ALLIED UNITS

Armiger Warglaive (140 points) • 1x Meltagun 1x Reaper chain-cleaver 1x Thermal spear

Armiger Warglaive (140 points) • 1x Meltagun 1x Reaper chain-cleaver 1x Thermal spear

Armiger Warglaive (140 points) • 1x Meltagun 1x Reaper chain-cleaver 1x Thermal spear

Callidus Assassin (115 points) • 1x Neural shredder 1x Phase sword and poison blades

Lord Inquisitor Kyria Draxus (75 points) • 1x Dirgesinger 1x Power fist 1x Psychic Tempest

That's as close to Imperium Soup as you can get in 10th, and the thread is called Adeptis(Adepta) Sororitas wtf and not Ministorum/Imperium Soup wtf after all.

It looks a lot better better with all kinds of games from the TTBattles App:

1656 45,62 49,21 47,34 % 784 793 79 47,44 %(mirror matches excluded)

, but even that is probably proped up by Soup, and the GT's are more relevant to determine the actual competetive strength of a faction imo.

Bahahaha seeing Kyganil randomly pop up every once in awhile (exclusively in the SoB context) always makes me laugh and cry. It's, like, the perfect absurd reality for a Harlequins fan.

Also yeah lol that list ain't Sisters, I don't care what somebody wants to call it

gunchar wrote:

 alextroy wrote:

In other news, AS results over the last week are:

10th Edition Games: 47.96% over 269 games

Leviathan GT Games: 50.00% over 152 games

From where the hell is that data?

https://40kstats.goonhammer.com/#GbF filtered for the last week's games. There may have been data updates since I posted. As of now, 7/18-7/24 yields:

10th Edition Games: 54.52% over 365 games

Leviathan GT Games: 34.29% over 105 games

This shows a marked reduction in both GT Games played and win percentage. Non-GT is looking pretty good. I have no idea to what extent allies impact the numbers.

Automatically Appended Next Post:
GW has graced us with two data card errata's:

ADEPTA SORORITAS
Mortifiers – Abilities, Anguish of the Unredeemed
Change to ‘Each time this unit makes a Charge move, until the end of the turn, melee weapons equipped by models in this unit have the [SUSTAINED HITS 1] ability.’

Dominion Squad – Ranged Weapons, boltgun
Add ‘[ASSAULT, RAPID FIRE 1].’

Necessary, but no real help here.

So for some reason they nerfed dominions?

If you count correcting the obvious error of Dominion Boltguns being Rapid Fire 2 to the intended Rapid Fire 1 is nerfing, then yes they nerfed Dominons.

 alextroy wrote:
If you count correcting the obvious error of Dominion Boltguns being Rapid Fire 2 to the intended Rapid Fire 1 is nerfing, then yes they nerfed Dominons.

It was the only thing that made them even a little bit of an option. Now the only reason to take the unit at all is because you're a masochist who hates winning.

 alextroy wrote:
If you count correcting the obvious error of Dominion Boltguns being Rapid Fire 2 to the intended Rapid Fire 1 is nerfing, then yes they nerfed Dominons.

But they left pipe bombs respawning for GSC? I have a list of stuff which to me is obvious error (GM GK stats, GM GK having their ability once per game, when other marines and custodes have an indetical one once per turn, etc), yet somehow GW didn't fix those. IMO what ever something is a typo or an error GW sometimes decides after some time. Like the knight thing they said was working for all re-rolls, but when they found out that knights are a roadblock army, they suddenly back paddled to one dice etc.

Dominions need something else to be relevant. Perhaps move to 2A or WS 3+ in melee if GW wont make a new kit for regular Celestians now that they've gone?

But honestly, even if [Rapid Fire 2] was a kinda obvious errata, it should had stayed. They need to punch harder for that price.

 Guillérmidas wrote:
Dominions need something else to be relevant. Perhaps move to 2A or WS 3+ in melee if GW wont make a new kit for regular Celestians now that they've gone?

But honestly, even if [Rapid Fire 2] was a kinda obvious errata, it should had stayed. They need to punch harder for that price.

giving them crappy melee wont help them they either need a firepower increase or to be cheaper.

For 20 points more than a Battle Sisters squad you get Scout 6", Assault Weapons, Re-Roll Advance Rolls, and 4 Special Weapons instead of a Special+Special/Heavy. If Dominions are not relevant it is because Sisters are too expensive overall, not this specific unit.

Dominions should all have special weapons as base.
...because that's what a Dominion is.

 Guillérmidas wrote:
Dominions need something else to be relevant. Perhaps move to 2A or WS 3+ in melee if GW wont make a new kit for regular Celestians now that they've gone?

But honestly, even if [Rapid Fire 2] was a kinda obvious errata, it should had stayed. They need to punch harder for that price.

That would do literally nothing. That's like giving a rhino sustained hits. Sure, it's technically better, just not in a way that matters.


Automatically Appended Next Post:

 alextroy wrote:
For 20 points more than a Battle Sisters squad you get Scout 6", Assault Weapons, Re-Roll Advance Rolls, and 4 Special Weapons instead of a Special+Special/Heavy. If Dominions are not relevant it is because Sisters are too expensive overall, not this specific unit.

Battle Sisters are relevant because they sit on objectives and generate miracle dice. They're too expensive for what they do, but they have a niche.

Dominions:
1. Have scout in an edition where scout is pretty mid for a shooting unit. It does nothing for them on foot and isn't that significant when they're in a transport.
2. Have reroll advances when the army has miracle dice. If the distance you needed on the advance was important enough for a reroll to matter, you should have Miracle Diced it. Advance rolls are one of the few things in the game that can potentially be a more valuable use of the dice than damage rolls.
3. They are a damage dealing unit that doesn't do any damage. 130pts for 4 melta shots is terrible. 130pts for 4 flamers (even S5 flamers) is terrible, 130pts for 4 Stormbolters is beyond laughable.
4. Their only viable buff character doesn't buff 1/3rd of their loadouts.
5. They occupy a similar niche to Seraphim, who are easily the best unit in the entire index. The only situations where dominions are better than Seraphim, are situations where Retributors are better than dominions.

Dominions are a bad unit that serves no purpose. If you start pricing Battle Sister bodies more appropriately, every unit they compete with ends up going down at the same rate they do and they still end up in a place where; if you're thinking about bringing dominions, examine what you want them to do, and then bring retributors or Seraphim instead.

ERJAK wrote:

 Guillérmidas wrote:
Dominions need something else to be relevant. Perhaps move to 2A or WS 3+ in melee if GW wont make a new kit for regular Celestians now that they've gone?

But honestly, even if [Rapid Fire 2] was a kinda obvious errata, it should had stayed. They need to punch harder for that price.

That would do literally nothing. That's like giving a rhino sustained hits. Sure, it's technically better, just not in a way that matters.

You're not wrong, but when Sisters get their Detachment that's more focused on melee it would be more relevant. Sometimes amplifying melee a little bit goes a long way for a close quarters unit.