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.

My eyes played a trick on me - so we agree - cents aren't good. So - Cents say hi. My point is that Oblitz were OP before and they are fixing that by making them a higher % of your army. You still get about the same damage efficiency with them at the new points with stratagems. Which is an OP efficiency BTW.

 Xenomancers wrote:
My eyes played a trick on me - so we agree - cents aren't good. So - Cents say hi. My point is that Oblitz were OP before and they are fixing that by making them a higher % of your army. You still get about the same damage efficiency with them at the new points with stratagems. Which is an OP efficiency BTW.

50% increase in Firepower at max squad size, but points went up by more than 75%. So no-they're not as efficient as before. More buffs have become available, but they all cost points.

They did get more resilient against some things (Lascannons by a good amount, small arms a good amount, OC Plasma a little) but are just as vulnerable to other things (Autocannons, for instance). And their melee is better, but that's negligible.

 JNAProductions wrote:

 Xenomancers wrote:
My eyes played a trick on me - so we agree - cents aren't good. So - Cents say hi. My point is that Oblitz were OP before and they are fixing that by making them a higher % of your army. You still get about the same damage efficiency with them at the new points with stratagems. Which is an OP efficiency BTW.

50% increase in Firepower at max squad size, but points went up by more than 75%. So no-they're not as efficient as before. More buffs have become available, but they all cost points.

They did get more resilient against some things (Lascannons by a good amount, small arms a good amount, OC Plasma a little) but are just as vulnerable to other things (Autocannons, for instance). And their melee is better, but that's negligible.

You are still paying the same CP to increase that firepower though and autocannons are useless against units with 2+ saves. It is pretty dang easy to set up in cover with 24" range.

Short Answer: NO
Long Answer: AAAAW HELLL NAW

But Seriously, what do Oblits do that isn't done better by other units for less points? Oblits are great stat lines, but worthless at the cost.

Also, 6 completely random shots? CRAAAAAAAAAAAPPPPP


Automatically Appended Next Post:
Wait, just checked. Is this right? HAVOCS SHOOT MORE REAPER DAKKA THEN A CHAOS TERMINATOR???? Bacon Cat Bug......explain!

Seriously, Havocs get 6 shots, Termies get 4. Facepalm.

Wayniac wrote:
The problem with pointing out RAI, in this case, is that Shadowspear is not something everyone may have. The Codex is. If someone has Obliterators already, and bought the new Codex (let's say they came back to the game) and turns up for a game, what? You tell them they're cheating because a different boxed set that they didn't hear of (that is, as of this writing, temporarily out of stock) changed the points but the codex was reprinted with the old points?

I'd be fine with it as long as they don't want to bring single model units. (Actually, I'd be fine with that too in regular play, as I'm pretty relaxed in my games, and I'd like to see how they do). It's not something I would let slide in a tourney setting. Official ruling on that. I wouldn't do it myself in relaxed games as that is not the price will be and I want to get a good sense of my list for the more competitive stuff.

 JNAProductions wrote:

 Xenomancers wrote:
My eyes played a trick on me - so we agree - cents aren't good. So - Cents say hi. My point is that Oblitz were OP before and they are fixing that by making them a higher % of your army. You still get about the same damage efficiency with them at the new points with stratagems. Which is an OP efficiency BTW.

50% increase in Firepower at max squad size, but points went up by more than 75%. So no-they're not as efficient as before. More buffs have become available, but they all cost points.

They did get more resilient against some things (Lascannons by a good amount, small arms a good amount, OC Plasma a little) but are just as vulnerable to other things (Autocannons, for instance). And their melee is better, but that's negligible.

You forget the non-negligible toughness and wounds increase. It's a complicated measure though, because there's fewer total wounds but higher toughness.

Also, one unit is outputting the same firepower as two old units now, so you're paying less for equal firepower. A new unit costs 345 points, compared to 2 old units costing 390. Both output 36 shots, and your dice-fixing buffs will go further on the new unit so your obliterators perform better more consistently too.

Also, with the ability to field them singly, they're better throwaway suicide drop units, since a 115 point guy is cheap enough to easily sacrifice, and he's tough enough to draw antitank guns, or at least a very large amount of power.


If you're fielding 3 or 1, you're doing better than before, but if you're fielding 2 you're doing worse.

2 old units did 36 with Endless Cacophony.

What about those of us that DON'T run Slaanesh?

Considered with the full range of buffs that Chaos is getting? Absolutely worth their points.

That's the question I get almost everytime I put Obliterators on the table "why aren't they Slaanesh?"

Ok, first, cents are good. Perhaps not the best for the role they fill, but with a captain/chapter master, leiutenant and relic banner, (all essentials for a gunline anyway) they are good, maybe even throw in an apothecary. Not great, but good. Also oblits seem fairly worth the cost, especially with the buffs available from the new psychic discipline. Maybe a little overcosted, but not to the point of not taking.

 JNAProductions wrote:
2 old units did 36 with Endless Cacophony.

What about those of us that DON'T run Slaanesh?

That's not my problem. A 3-man unit is pretty much the ideal target for Endless Cacophony. If you're not using Endless Cacophony on them because you have a better/more essential target for it that's understandable, and you can consider the newly-opened alternate use case of a single obliterator on a suicide mission. He's still solo strong enough to inconvenience a tank [or even outright kill it if you're lucky], needs to die right away, and takes heavy fire to dislodge.

But if you're making them Tzeentch or something, you probably also don't care about the 5% change in cost-efficiency, right?

I think without taking buffs into account they are overcosted by 15-25 points. One Obliterator is only a little cheaper than, say, a Venomcrawler with a similar gun and 2/3 the shots but massively better survivability and melee.

Add buffs to the equation and it is still the best AT shooting unit Chaos has, to my knowledge. So I think they are still worth taking even if they are overcosted in an abstract sense.

 JNAProductions wrote:
2 old units did 36 with Endless Cacophony.

What about those of us that DON'T run Slaanesh?

Which is the problem with costing units like they always have a Strategem active. Why is the Strategem not priced like it'll be used on Obliterators instead?

 CREEEEEEEEED wrote:
Ok, first, cents are good. Perhaps not the best for the role they fill, but with a captain/chapter master, leiutenant and relic banner, (all essentials for a gunline anyway) they are good, maybe even throw in an apothecary. Not great, but good. Also oblits seem fairly worth the cost, especially with the buffs available from the new psychic discipline. Maybe a little overcosted, but not to the point of not taking.

Normally I would not agree but seen as those things are pretty much a must take anyway I see your point, I'd like to try them out but damn those models are ugly as all hell, thinking of just getting the primaris ones and just slapping them on a bigger base as counts as, I don't use any other primaris so it should not be confusing.

 Inquisitor Lord Katherine wrote:
I did some comparative math regarding their cost efficiency back when the cost was first previewed.

The short answer is that you're getting a better unit for units of 1 or 3 Obliterators, but it's less efficient to run 2, which happens to be the number in the box.

Spoiler:

The formula weights the importance of the various factors [down to resistance to various categories of weapons], and then computes an average change in the values. 1 and 3 oblit are largely always over the line, 2 is always below the line, and the tail for fielding a ton of them is generally down unless you want the shooting more than the price.

This is an... interesting assertion to make. Can you explain what you're doing here - what the actual formula is, any assumptions you've made etc - as I'm really not sure how you're justifying your statement.

On sunday my opponent was running two units of three, old models of course, but new datasheet. He brought as sorcerer and an apostle to buffen them.

A unit of 3 now has 18 shots, where it had 12 before, so when shooting additional times, buffing to hit (prescience, warp sight plea, many stratagems) and/or buffing to wound (VotLW), you get more bang for your buck.
My opponent could provide two units of obliterators with +1 to hit at all times, with one shooting twice. So for 690 he had 54 shots hitting on 2+, dealing 26.667 damage to your average death guard model (3+ T5-7, FNP), so 0.03865 damage per point
Three units from the old datasheet would yield 36 shots hitting on 2+ and 12 hitting on 3+ dealing 22.519 for 585, which is 0.03849 damage per point.

So with those two buffs, my opponent was actually doing 0.3% more damage per point spent than he would have done before the change. So basically, they stayed the same, as you are not going to notice a 0.3% change when rolling d6.

 Xenomancers wrote:


50% increase in Firepower at max squad size, but points went up by more than 75%. So no-they're not as efficient as before. More buffs have become available, but they all cost points.

They did get more resilient against some things (Lascannons by a good amount, small arms a good amount, OC Plasma a little) but are just as vulnerable to other things (Autocannons, for instance). And their melee is better, but that's negligible.

You are still paying the same CP to increase that firepower though and autocannons are useless against units with 2+ saves. It is pretty dang easy to set up in cover with 24" range.

Slayer-Fan123 wrote:

 JNAProductions wrote:
2 old units did 36 with Endless Cacophony.

What about those of us that DON'T run Slaanesh?

Which is the problem with costing units like they always have a Strategem active. Why is the Strategem not priced like it'll be used on Obliterators instead?

Because then it's miscosted for use on a CSM squad or whatever else you want to use it for.

There exists no static price points (A, B, C) such that
-A is the cost for a 3man Oblit squad
-B is the cost of 3 CSM
-C is the cost of Cacophony

Such that [A, B, A+B, A+C] are all fair.

Sort of as if Strategems are a bad, gimmicky mechanic, eh?

 Excommunicatus wrote:
Sort of as if Strategems are a bad, gimmicky mechanic, eh?

Considering that what Bharring said is applicable for 90% of the mechanics of this game, of AoS, of Warma/hordes, of Kill Team, of X-Wing... actually of almost all the games i know with somewhat complex rules, i don't undersand your comment.

Actually what Bharring said is "This game has non linear interacions", which is a definition at least as generic as "This game uses dices".

What it means is that, without much more complicated rules, it can't be fair.

The saddest thing I've ever read was a proof that no non-trivial voting system can be fair. Incredibly sad, because it wasn't just an argument or some evidence - it was a formal proof with extremely defensible assumptions. There's better and worse voting systems. But there *is no* perfect voting system. Not because nobody's thought it up. Not because we haven't figured it out yet. It is demonstrable that it *can't* be figured out.

If that's the sad truth about balance in the voting system, what does that say about seeking balance in something so much more trivial? So much less important? With so many fewer eyes?

The game will never be balanced. It's a question of how much more or less balanced it will be, not a question of how to balance it perfectly.

So the next time you have the "perfect" answer, realize that you don't. Perhaps it is better. In most ways. But it won't be perfect.

What I said can be applied to most games. You can add complexity to "fix" it (make Cacophony cost an amount based on the unit size and value). But that's a hell of a lot of bloat. GW - and many players - would rather the rules as-is. Because the bloat-vs-balance tradeoff is too high.

Bharring wrote:
What it means is that, without much more complicated rules, it can't be fair.

The saddest thing I've ever read was a proof that no non-trivial voting system can be fair. Incredibly sad, because it wasn't just an argument or some evidence - it was a formal proof with extremely defensible assumptions. There's better and worse voting systems. But there *is no* perfect voting system. Not because nobody's thought it up. Not because we haven't figured it out yet. It is demonstrable that it *can't* be figured out.

If that's the sad truth about balance in the voting system, what does that say about seeking balance in something so much more trivial? So much less important? With so many fewer eyes?

The game will never be balanced. It's a question of how much more or less balanced it will be, not a question of how to balance it perfectly.

So the next time you have the "perfect" answer, realize that you don't. Perhaps it is better. In most ways. But it won't be perfect.

What I said can be applied to most games. You can add complexity to "fix" it (make Cacophony cost an amount based on the unit size and value). But that's a hell of a lot of bloat. GW - and many players - would rather the rules as-is. Because the bloat-vs-balance tradeoff is too high.

We agree that it needs more complicated rules to be fair, but that is not something stricly related to stratagems. Psy powers work like that. Also chapter traits and almost any kind of buff.
If we want to go all the way, any shooting unit is also not defineable by a point value, because it has a range, and range has a non linear impact on the game.

We all know that points aren't perfect, but is a reasonable approach.


Defining stratagems as a "bad, gimmicky mechanic" because they can't be properly costed, means showing a deep ignorance of the game.
Note: This is also true for those whose mantra is "Any unit can be balanced at the right point cost", there is nothing wronger than that assumption.

Aelyn wrote:

 Inquisitor Lord Katherine wrote:
I did some comparative math regarding their cost efficiency back when the cost was first previewed.

The short answer is that you're getting a better unit for units of 1 or 3 Obliterators, but it's less efficient to run 2, which happens to be the number in the box.

Spoiler:

The formula weights the importance of the various factors [down to resistance to various categories of weapons], and then computes an average change in the values. 1 and 3 oblit are largely always over the line, 2 is always below the line, and the tail for fielding a ton of them is generally down unless you want the shooting more than the price.

This is an... interesting assertion to make. Can you explain what you're doing here - what the actual formula is, any assumptions you've made etc - as I'm really not sure how you're justifying your statement.

It's a series of weighted averages of percent changes in the various metrics based on the most-similar configuration enemy unit.
The overweights are noted at the top of the plot, but the weights inside the amalgamated "resilience, firepower, cqc, move, and cost" aren't shown. In this specific plot, Firepower is rated 2, resilience 1, and cost 3. Move and cqc are both rated 0 because there's no change to move, and the CQC improvement is irrelevant to the unit.

Each metric is calculated as a weighted average in the percent change of its subcomponents.

For firepower, only shot count is weighted; because the other values didn't change. I could introduce weights to allow me to implement the effects of CP re-rolls and new psychic buffs, but that would be complex once introducing the existence of buffed and un-buffed units, and while it's a safe bet that they will recieved Cacophony, it's not a safe bet on them receiving the entire potential buff suite.

Cost is simple, just backwards since cost decreases are good and increases are bad.

Resiliency has 3 components, save, toughness, and wound count. Save is weighted 0, since it's the same, toughness is valued more than wound count, but barely.
Toughness is considered like strength, and that's a weighted average of the odds of being wounded by hits of various strengths, weighted by their relevance. S4 and S3 are considered highly relevant, S5, S7, S9, and S10 are low relevance, S6 and S8 are considered moderate relevance.This is justified on the principle that antitank guns cost a lot more than anti-infantry guns, and the idea that the enemy will generally have more anti-infantry weapons than things to shoot them at but fewer antitank weapons than things they want to kill with them, so Obliterators dying to lascannons is less of a cause for concern since those aren't targeting your other high value models, or using their AP and range against entrenched havocs, or whatever.
Wound count is the total wound count in the unit. Wounds-to-degredation isn't accounted for, because it's more complicated, but could arguably be applied to Obliterators, since they have a fairly discontinuous degredation.


As for picking the most-similar unit to compare against at each level, I did this generally based on closest to having the same fire output, on the grounds that fire output is the primary driver in choosing to select a unit of obliterators to bring. It's notable that 1 does not assume the use of Endless Cacophony, on the grounds that the use case for a single solo suicide obliterator is different from the use case of a large squad. 2 newblitz are compared to 3 oldblitz, 3 newblitz are compared to 6 oldblitz, 4 newblitz are still compared to 6 oldblitz, and 5+ newblitz are compared to 9 oldblitz.

I can make a new, graph to account for the introduction of CP-re-rolls and re-rolls from the new psychic powers on gun stats, since that's pretty important to the primary use case of obliterators, and investing heavily in them also logically leads to being willing to invest in supporting them. In addition, I could try to account for the importance of their resistance to weapons of different damage class, since the change in wounds/guy is actually more important than just an overall decrease of wounds in the unit, because the one additional wound can absorb multiple points of damage from common multi-damage weapons used to eliminate obliterators. It's also notable that the relative importance of the various aspects of the unit vary by use case, so it's also worth taking it with a grain of salt there, since a single suicide newblit might care more about toughness and low cost than a big squad of obliterators which is much more interested in achieving destructive fire [and cost].

It's notable that varying the weights, or even the sub-weights can make the graph change a lot. However, for most "realistic" weightings [high importance of cost, medium-high importance of firepower, low importance of toughness], it looks like a buff at 1 and 3, but a nerf at most other values.


Interpreting the graph is also important, because it doesn't just say that 3 newblitz are going to be awesome. It says that 3 newblitz are more cost efficient than 6 oldblitz, their most-similar unit counterpart. Use case is important, and you might only really be interested in a weapons unit at the price point of 200 points, at which point you don't really care that they're more efficient at 3 than they used to be, because the 155 points you have to drop matter more to you than the overall improvement of that efficiency. And lists don't come down to buying units of maximum efficiency ["good unit soup" only wins games against "bad unit soup"], they come down to interactions between units, so variation in efficiency doesn't tell the whole story. In addition, it's percent change, and is in most cases is pretty low, like 5%, which you could easily write off as mostly irrelevant at the level of "use them if you liked them before"; and if you thought they weren't worth it before, a 5% improvement over bad is still bad.

Are the points costs in shadowspear not geared specifically for all the models that come in the box? what I mean is isn't shadowspear it's own mini-game and the models in the box only interacting with each other?

I haven't read any of the literature that came with the box so I'm just throwing this out there.

 Jidmah wrote:
On sunday my opponent was running two units of three, old models of course, but new datasheet. He brought as sorcerer and an apostle to buffen them.

A unit of 3 now has 18 shots, where it had 12 before, so when shooting additional times, buffing to hit (prescience, warp sight plea, many stratagems) and/or buffing to wound (VotLW), you get more bang for your buck.
My opponent could provide two units of obliterators with +1 to hit at all times, with one shooting twice. So for 690 he had 54 shots hitting on 2+, dealing 26.667 damage to your average death guard model (3+ T5-7, FNP), so 0.03865 damage per point
Three units from the old datasheet would yield 36 shots hitting on 2+ and 12 hitting on 3+ dealing 22.519 for 585, which is 0.03849 damage per point.

So with those two buffs, my opponent was actually doing 0.3% more damage per point spent than he would have done before the change. So basically, they stayed the same, as you are not going to notice a 0.3% change when rolling d6.

It's not just that - being a random weapon that when it shoot twice retains the roll for the whole phase. A command reroll (one time use) has significantly more potential on 1 unit.

 Inquisitor Lord Katherine wrote:

Aelyn wrote:

 Inquisitor Lord Katherine wrote:
I did some comparative math regarding their cost efficiency back when the cost was first previewed.

The short answer is that you're getting a better unit for units of 1 or 3 Obliterators, but it's less efficient to run 2, which happens to be the number in the box.

Spoiler:

The formula weights the importance of the various factors [down to resistance to various categories of weapons], and then computes an average change in the values. 1 and 3 oblit are largely always over the line, 2 is always below the line, and the tail for fielding a ton of them is generally down unless you want the shooting more than the price.

This is an... interesting assertion to make. Can you explain what you're doing here - what the actual formula is, any assumptions you've made etc - as I'm really not sure how you're justifying your statement.

Spoiler:

It's a series of weighted averages of percent changes in the various metrics based on the most-similar configuration enemy unit.
The overweights are noted at the top of the plot, but the weights inside the amalgamated "resilience, firepower, cqc, move, and cost" aren't shown. In this specific plot, Firepower is rated 2, resilience 1, and cost 3. Move and cqc are both rated 0 because there's no change to move, and the CQC improvement is irrelevant to the unit.

Each metric is calculated as a weighted average in the percent change of its subcomponents.

For firepower, only shot count is weighted; because the other values didn't change. I could introduce weights to allow me to implement the effects of CP re-rolls and new psychic buffs, but that would be complex once introducing the existence of buffed and un-buffed units, and while it's a safe bet that they will recieved Cacophony, it's not a safe bet on them receiving the entire potential buff suite.

Cost is simple, just backwards since cost decreases are good and increases are bad.

Resiliency has 3 components, save, toughness, and wound count. Save is weighted 0, since it's the same, toughness is valued more than wound count, but barely.
Toughness is considered like strength, and that's a weighted average of the odds of being wounded by hits of various strengths, weighted by their relevance. S4 and S3 are considered highly relevant, S5, S7, S9, and S10 are low relevance, S6 and S8 are considered moderate relevance.This is justified on the principle that antitank guns cost a lot more than anti-infantry guns, and the idea that the enemy will generally have more anti-infantry weapons than things to shoot them at but fewer antitank weapons than things they want to kill with them, so Obliterators dying to lascannons is less of a cause for concern since those aren't targeting your other high value models, or using their AP and range against entrenched havocs, or whatever.
Wound count is the total wound count in the unit. Wounds-to-degredation isn't accounted for, because it's more complicated, but could arguably be applied to Obliterators, since they have a fairly discontinuous degredation.


As for picking the most-similar unit to compare against at each level, I did this generally based on closest to having the same fire output, on the grounds that fire output is the primary driver in choosing to select a unit of obliterators to bring. It's notable that 1 does not assume the use of Endless Cacophony, on the grounds that the use case for a single solo suicide obliterator is different from the use case of a large squad. 2 newblitz are compared to 3 oldblitz, 3 newblitz are compared to 6 oldblitz, 4 newblitz are still compared to 6 oldblitz, and 5+ newblitz are compared to 9 oldblitz.

I can make a new, graph to account for the introduction of CP-re-rolls and re-rolls from the new psychic powers on gun stats, since that's pretty important to the primary use case of obliterators, and investing heavily in them also logically leads to being willing to invest in supporting them. In addition, I could try to account for the importance of their resistance to weapons of different damage class, since the change in wounds/guy is actually more important than just an overall decrease of wounds in the unit, because the one additional wound can absorb multiple points of damage from common multi-damage weapons used to eliminate obliterators. It's also notable that the relative importance of the various aspects of the unit vary by use case, so it's also worth taking it with a grain of salt there, since a single suicide newblit might care more about toughness and low cost than a big squad of obliterators which is much more interested in achieving destructive fire [and cost].

It's notable that varying the weights, or even the sub-weights can make the graph change a lot. However, for most "realistic" weightings [high importance of cost, medium-high importance of firepower, low importance of toughness], it looks like a buff at 1 and 3, but a nerf at most other values.

Interpreting the graph is also important, because it doesn't just say that 3 newblitz are going to be awesome. It says that 3 newblitz are more cost efficient than 6 oldblitz, their most-similar unit counterpart. Use case is important, and you might only really be interested in a weapons unit at the price point of 200 points, at which point you don't really care that they're more efficient at 3 than they used to be, because the 155 points you have to drop matter more to you than the overall improvement of that efficiency. And lists don't come down to buying units of maximum efficiency ["good unit soup" only wins games against "bad unit soup"], they come down to interactions between units, so variation in efficiency doesn't tell the whole story. In addition, it's percent change, and is in most cases is pretty low, like 5%, which you could easily write off as mostly irrelevant at the level of "use them if you liked them before"; and if you thought they weren't worth it before, a 5% improvement over bad is still bad.

That's... an awful lot of words that don't really answer my questions. You say that interpreting the graph is important, but you've not really given enough information for us to meaningfully interpret it. No offence intended, but this is something of a blackbox - you haven't explained how you actually derived the numbers, so how can I decide whether the output provided is meaningful?

It just felt odd to me that you claim 2 obliterators to be "less efficient" than 1 or 3, and the only thing I can see that could justify this is how you've chosen to compare 2 of the new to 3 of the old, but 3 of the new to 6 of the old. I'm pretty sure that's where the "2 is less efficient" idea is coming from - you're comparing to a pseudo-arbitrary number of the old ones, with an inconsistent conversion ratio.

EDIT: To clarify, I think this paragraph is the crux of the whole point:

As for picking the most-similar unit to compare against at each level, I did this generally based on closest to having the same fire output, on the grounds that fire output is the primary driver in choosing to select a unit of obliterators to bring. It's notable that 1 does not assume the use of Endless Cacophony, on the grounds that the use case for a single solo suicide obliterator is different from the use case of a large squad. 2 newblitz are compared to 3 oldblitz, 3 newblitz are compared to 6 oldblitz, 4 newblitz are still compared to 6 oldblitz, and 5+ newblitz are compared to 9 oldblitz. 

I don't see how you're justifying the underlined figures.

Since they need to be FAQ'd anyway, GW should take the opportunity to make them an even 100 points each. Because at 115 nobody will take them.

Aelyn wrote:

 Inquisitor Lord Katherine wrote:

Aelyn wrote:

 Inquisitor Lord Katherine wrote:
I did some comparative math regarding their cost efficiency back when the cost was first previewed.

The short answer is that you're getting a better unit for units of 1 or 3 Obliterators, but it's less efficient to run 2, which happens to be the number in the box.

Spoiler:

The formula weights the importance of the various factors [down to resistance to various categories of weapons], and then computes an average change in the values. 1 and 3 oblit are largely always over the line, 2 is always below the line, and the tail for fielding a ton of them is generally down unless you want the shooting more than the price.

This is an... interesting assertion to make. Can you explain what you're doing here - what the actual formula is, any assumptions you've made etc - as I'm really not sure how you're justifying your statement.

Spoiler:

It's a series of weighted averages of percent changes in the various metrics based on the most-similar configuration enemy unit.
The overweights are noted at the top of the plot, but the weights inside the amalgamated "resilience, firepower, cqc, move, and cost" aren't shown. In this specific plot, Firepower is rated 2, resilience 1, and cost 3. Move and cqc are both rated 0 because there's no change to move, and the CQC improvement is irrelevant to the unit.

Each metric is calculated as a weighted average in the percent change of its subcomponents.

For firepower, only shot count is weighted; because the other values didn't change. I could introduce weights to allow me to implement the effects of CP re-rolls and new psychic buffs, but that would be complex once introducing the existence of buffed and un-buffed units, and while it's a safe bet that they will recieved Cacophony, it's not a safe bet on them receiving the entire potential buff suite.

Cost is simple, just backwards since cost decreases are good and increases are bad.

Resiliency has 3 components, save, toughness, and wound count. Save is weighted 0, since it's the same, toughness is valued more than wound count, but barely.
Toughness is considered like strength, and that's a weighted average of the odds of being wounded by hits of various strengths, weighted by their relevance. S4 and S3 are considered highly relevant, S5, S7, S9, and S10 are low relevance, S6 and S8 are considered moderate relevance.This is justified on the principle that antitank guns cost a lot more than anti-infantry guns, and the idea that the enemy will generally have more anti-infantry weapons than things to shoot them at but fewer antitank weapons than things they want to kill with them, so Obliterators dying to lascannons is less of a cause for concern since those aren't targeting your other high value models, or using their AP and range against entrenched havocs, or whatever.
Wound count is the total wound count in the unit. Wounds-to-degredation isn't accounted for, because it's more complicated, but could arguably be applied to Obliterators, since they have a fairly discontinuous degredation.


As for picking the most-similar unit to compare against at each level, I did this generally based on closest to having the same fire output, on the grounds that fire output is the primary driver in choosing to select a unit of obliterators to bring. It's notable that 1 does not assume the use of Endless Cacophony, on the grounds that the use case for a single solo suicide obliterator is different from the use case of a large squad. 2 newblitz are compared to 3 oldblitz, 3 newblitz are compared to 6 oldblitz, 4 newblitz are still compared to 6 oldblitz, and 5+ newblitz are compared to 9 oldblitz.

I can make a new, graph to account for the introduction of CP-re-rolls and re-rolls from the new psychic powers on gun stats, since that's pretty important to the primary use case of obliterators, and investing heavily in them also logically leads to being willing to invest in supporting them. In addition, I could try to account for the importance of their resistance to weapons of different damage class, since the change in wounds/guy is actually more important than just an overall decrease of wounds in the unit, because the one additional wound can absorb multiple points of damage from common multi-damage weapons used to eliminate obliterators. It's also notable that the relative importance of the various aspects of the unit vary by use case, so it's also worth taking it with a grain of salt there, since a single suicide newblit might care more about toughness and low cost than a big squad of obliterators which is much more interested in achieving destructive fire [and cost].

It's notable that varying the weights, or even the sub-weights can make the graph change a lot. However, for most "realistic" weightings [high importance of cost, medium-high importance of firepower, low importance of toughness], it looks like a buff at 1 and 3, but a nerf at most other values.

Interpreting the graph is also important, because it doesn't just say that 3 newblitz are going to be awesome. It says that 3 newblitz are more cost efficient than 6 oldblitz, their most-similar unit counterpart. Use case is important, and you might only really be interested in a weapons unit at the price point of 200 points, at which point you don't really care that they're more efficient at 3 than they used to be, because the 155 points you have to drop matter more to you than the overall improvement of that efficiency. And lists don't come down to buying units of maximum efficiency ["good unit soup" only wins games against "bad unit soup"], they come down to interactions between units, so variation in efficiency doesn't tell the whole story. In addition, it's percent change, and is in most cases is pretty low, like 5%, which you could easily write off as mostly irrelevant at the level of "use them if you liked them before"; and if you thought they weren't worth it before, a 5% improvement over bad is still bad.

That's... an awful lot of words that don't really answer my questions. You say that interpreting the graph is important, but you've not really given enough information for us to meaningfully interpret it. No offence intended, but this is something of a blackbox - you haven't explained how you actually derived the numbers, so how can I decide whether the output provided is meaningful?

It just felt odd to me that you claim 2 obliterators to be "less efficient" than 1 or 3, and the only thing I can see that could justify this is how you've chosen to compare 2 of the new to 3 of the old, but 3 of the new to 6 of the old. I'm pretty sure that's where the "2 is less efficient" idea is coming from - you're comparing to a pseudo-arbitrary number of the old ones, with an inconsistent conversion ratio.

EDIT: To clarify, I think this paragraph is the crux of the whole point:

As for picking the most-similar unit to compare against at each level, I did this generally based on closest to having the same fire output, on the grounds that fire output is the primary driver in choosing to select a unit of obliterators to bring. It's notable that 1 does not assume the use of Endless Cacophony, on the grounds that the use case for a single solo suicide obliterator is different from the use case of a large squad. 2 newblitz are compared to 3 oldblitz, 3 newblitz are compared to 6 oldblitz, 4 newblitz are still compared to 6 oldblitz, and 5+ newblitz are compared to 9 oldblitz. 

I don't see how you're justifying the underlined figures.

I'm comparing it to the most-equivalent unit configuration. There's only a 45 point cost difference between 3 newblitz and 6 oldblitz and no difference in fire output. There's 6 less wounds. This is a roughly equivalent unit composition.

Comparing 2 newblitz to 3 oldblitz are also roughly comparable units, with a similar price point [35 points difference], same number of shots, and similar number of wounds.

Obliterators aren't continous, they're discrete. You cannot field 5 oldblitz [or at the ridiculous, a unit of 4.3 models]. It's also important to compare units in equivalent use-cases, because an efficieny comparison between a Grot and a Warlord Titan are essentially meaningless.

Thats how the comparison targets are chosen, buy choosing what I judged to be the most similar possible oldbliterator configuration. I could make a 3d plot to compare all the possible configurations, but all in all, and a fair number of them aren't very useful comparisons [IE: if you are going to run 1 newblit as a suicide drop, you wouldn't run 9 oldblitz as a antitank battery, so that's not a useful comparison].

Anyway, long story short, because the new, stronger unit stretches the value you get out of buffs onto them, there's a notable buff at the point of overflow, where you only have 1 unit of newbliterators, but need 2 units of oldbliterators to achieve the same effect. In short, the real buff was increased unit total power.

Cynista wrote:
Since they need to be FAQ'd anyway, GW should take the opportunity to make them an even 100 points each. Because at 115 nobody will take them.

I will take three of them.

 Inquisitor Lord Katherine wrote:

I'm comparing it to the most-equivalent unit configuration. There's only a 45 point cost difference between 3 newblitz and 6 oldblitz and no difference in fire output. There's 6 less wounds. This is a roughly equivalent unit composition.

Comparing 2 newblitz to 3 oldblitz are also roughly comparable units, with a similar price point [35 points difference], same number of shots, and similar number of wounds.

Obliterators aren't continous, they're discrete. You cannot field 5 oldblitz [or at the ridiculous, a unit of 4.3 models]. It's also important to compare units in equivalent use-cases, because an efficieny comparison between a Grot and a Warlord Titan are essentially meaningless.

Thats how the comparison targets are chosen, buy choosing what I judged to be the most similar possible oldbliterator configuration.

Anyway, long story short, because the new, stronger unit stretches the value you get out of buffs onto them, there's a notable buff at the point of overflow, where you only have 1 unit of newbliterators, but need 2 units of oldbliterators to achieve the same effect. In short, the real buff was increased squad size.

Okay, but again, you didn't actually state the assumptions being used to get to your figures. I now realise that your graph was talking specifically about Slaaneshi Obliterators which get Endless Cacophony played on them - this was not mentioned in your first post, and only tangentially mentioned in the second post.

It also makes your graph useless to me, as I would run them in a Nurgle Epidemius force.

It's also worth mentioning that - since you're comparing one unit to two - you've tacitly ignored the additional flexibility gained by having two units, but have taken into account the reduced impacts of buffs. Again, this is not stated anywhere.

See how your message can be muddied if you don't make your assumptions explicit?