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.

 Evileyes wrote:
If I were to consider this logically, I would say it's the armour save.

Why?

I can't imagine the daemon, feeling suddenly cocky, rip's off his own armour and goes "Booyah! I'm taking this one head on!"

Well, it is a khornate demon. Common sense isn't really it's strong point

But yes, I agree. You use the armor save.


Automatically Appended Next Post:

 Awfeel wrote:

 CthuluIsSpy wrote:
They do not account for rerolls when they say best save available. You use the 3+.

says what? what page is that on exactly?

Pg19.

Under "Models with more than one save." The bold part.

 AndrewC wrote:
Thats bad maths. Statistically, you still rolled 6 dice, for a 50% chance of acheiving the aim.

No, your math is bad.
Statistically it has a 75% chance of succeeding.

If you've rolled the first and failed, NOW you have a 50% chance of succeeding again.

But if you have 100 rolls to roll ONE 4+, you cannot in all seriousness think it good math to say there is a 50% chance of succeeding that? Because if you do, I'd like to see you bet me 1:1 that I'll succeed. Hell, I'll give you 10:1 odds. It's 50%, right? Money in the bank for you.

It's 75%. 1/2 chance of passing, half chance of a re-roll. The re-roll is again 1/2 chance of passing. So 1/2 + 1/2 of 1/2 is 3/4.

Otherwise, the odds and probabilities don't really work when talking about related events. Yes it's 50% per dice roll but that doesn't equate to it is always 50% probability of getting the result you want.

 liturgies of blood wrote:
It's 75%. 1/2 chance of passing, half chance of a re-roll. The re-roll is again 1/2 chance of passing. So 1/2 + 1/2 of 1/2 is 3/4.

This is right, although I would notate it as
.5+(.5*.5)=.75 or 1/2+(1/2*1/2)=3/4
It gets significantly harder when you're trying to calculate the odds of one 4+ in a batch of dice, or figure out (as my Psyfleman likes to do) the chance of a pen on a Flyer. Four TL Str8 shots, no Skyfire, AV12. I need 6's to hit, then I need 5+ to pen, he might Evade. Easy enough with one die...confusing with four.

I find it easier to demonstrate this in terms of coinfliips:

A 4+ save is (in theory) equivalent to a tossing a coin and getting a heads; both have a 50% chance of success.

Now, saying "you take a wound if the first save fails, you re-roll, and the second save fails" is equivalent to "you take a wound if you roll twice, and both fail" - if you're unclear on that, just imagine that having rolled them, the one nearer to you is automatically the "first" roll.

So a 4+ rerollable save is equivalent to tossing two coins, and only dying if both are tails. So look at the possibilities:

HH - Survives!
HT - Survives!
TH - Survives on the reroll!
TT - Dies...

Three times out of four you survive, so a 4+ rerollable has a 75% chance of success.

If you're still uncertain of this, please let me know which part of the reasoning you don't get and I'll be happy to clarify.

 Elric Greywolf wrote:

 liturgies of blood wrote:
It's 75%. 1/2 chance of passing, half chance of a re-roll. The re-roll is again 1/2 chance of passing. So 1/2 + 1/2 of 1/2 is 3/4.

This is right, although I would notate it as
.5+(.5*.5)=.75 or 1/2+(1/2*1/2)=3/4
It gets significantly harder when you're trying to calculate the odds of one 4+ in a batch of dice, or figure out (as my Psyfleman likes to do) the chance of a pen on a Flyer. Four TL Str8 shots, no Skyfire, AV12. I need 6's to hit, then I need 5+ to pen, he might Evade. Easy enough with one die...confusing with four.

In the future calculate your 1 die result and realize you have the chance 4 times. Obviously they can take into unrealistic territory like say 110% chance to succeed. but this will give you relative odds. The thing people tend to forget when calculating odds in this game is each die roll is an independent trial and previous die rolls do not impact future die rolls. Many people will say that rolling a 6 followed by a 6 is a 1 in 36 chance. But if you have already rolled the first six you are no more or less likely to get a 6 on the second roll meaning you have a 1 in 6 chance to get that second 6 once the first is rolled.

I'm running an Autarch with Harlequin's Mantle (or was it the wings that gave cover re-roll?) across open ground and he's then shot at by a group of tactical marines.

What prevents me from rolling 3+ armour save on the first save and 4+ cover save on the re-roll in case the first save failed?

The problem with that particular naivety, of course, is when a Farseer with a Warlock's armour strengthening suffers wounds from a vehicle explosion while under the effect of Null Zone and then rolls a 6 on the first save roll - an armour save would have succeeded but the invulnerable must be re-rolled.

* I might get some names wrong, I've got the rulebooks tucked away in my gaming bags and/or iPad. You get the gist of it.

The wording of the special rule themselves.

The vast majority that I just flicked through all state you can re-roll failed X saves. In order to have failed a roll, you need to make it in the first place. If you roll a different type of save then the one being listed under X, and fail that save, you still have not met the requirements to be allowed to re-roll. Any special rule that is a re-roll on all type of saving throws are worded in such a way that you would still be re-rolling the save you failed, as that would still be the best available save.

To enter the situation you are describing the rule would have to be worded along the lines of, 'if you fail your saving throw you may take an additional saving throw using the best available cover save.'

tgf wrote:


In the future calculate your 1 die result and realize you have the chance 4 times. Obviously they can take into unrealistic territory like say 110% chance to succeed. but this will give you relative odds. The thing people tend to forget when calculating odds in this game is each die roll is an independent trial and previous die rolls do not impact future die rolls. Many people will say that rolling a 6 followed by a 6 is a 1 in 36 chance. But if you have already rolled the first six you are no more or less likely to get a 6 on the second roll meaning you have a 1 in 6 chance to get that second 6 once the first is rolled.

This is terrible math. I realise that rolling one 6 on a single die is 1/6. But the chance of one 6 showing up on four dice rolls is not 1/6, it's much higher.
Someone else above brought up a similar problem. What's the chance of one 4+ showing up when I roll 100 dice? It's greater than 50%, that's for sure!

tgf wrote:

 Elric Greywolf wrote:

 liturgies of blood wrote:
It's 75%. 1/2 chance of passing, half chance of a re-roll. The re-roll is again 1/2 chance of passing. So 1/2 + 1/2 of 1/2 is 3/4.

This is right, although I would notate it as
.5+(.5*.5)=.75 or 1/2+(1/2*1/2)=3/4
It gets significantly harder when you're trying to calculate the odds of one 4+ in a batch of dice, or figure out (as my Psyfleman likes to do) the chance of a pen on a Flyer. Four TL Str8 shots, no Skyfire, AV12. I need 6's to hit, then I need 5+ to pen, he might Evade. Easy enough with one die...confusing with four.

In the future calculate your 1 die result and realize you have the chance 4 times. Obviously they can take into unrealistic territory like say 110% chance to succeed. but this will give you relative odds. The thing people tend to forget when calculating odds in this game is each die roll is an independent trial and previous die rolls do not impact future die rolls. Many people will say that rolling a 6 followed by a 6 is a 1 in 36 chance. But if you have already rolled the first six you are no more or less likely to get a 6 on the second roll meaning you have a 1 in 6 chance to get that second 6 once the first is rolled.

You really don't understand statistics, and I would submit you should stop advising others on how they work. While every die roll is indeed independent, you can make statistical analysis of multiple die rolls. You can never get over 100% chance. Ever. The fact that you said you can just shows you don't have a firm grasp of this area of study.

I don't mean this as an attack, but rather a plea. There are many resources that teach how statistics work and can be calculated.

For instance - lets say you roll two dice and want at least one result of 2 or better. So we know that you have a 5/6 chance of getting either die to roll a 2+. However, if you want to know your chance of getting at least one success between the two dice, you figure out the odds of failing both, and subtract from one.

So in this example, you on two dice you get non-snake eyes 1 - (1/6 * 1/6) = 97%. Your average number of successes is different, giving you 10/6 or 1.66667 successes on average with 2 successes being you mean result.

Statistics can, will and should be used for these purposes.

The best available save is not clearly defined in the BRB.

So if 4++ gives him a 75% chance of survival and the 3+ only a 66.67% chance, than it's clear what the best save is.

Kangodo wrote:
The best available save is not clearly defined in the BRB.

So if 4++ gives him a 75% chance of survival and the 3+ only a 66.67% chance, than it's clear what the best save is.

Let me ask you this then - which is the better save (not best, just better), 4++ with re-roll or 3+?

 Happyjew wrote:
Let me ask you this then - which is the better save (not best, just better), 4++ with re-roll or 3+?

Best is a term when you compare two (or more) things.
The one that is better than any other is considered the best.

Seeing as 4++ has a reroll, it is the best.

Kangodo wrote:

 Happyjew wrote:
Let me ask you this then - which is the better save (not best, just better), 4++ with re-roll or 3+?

Best is a term when you compare two (or more) things.
The one that is better than any other is considered the best.

Seeing as 4++ has a reroll, it is the best.

Except the rules define lower as better - therefore a 3+ is better than a 4++ as the 3+ is lower. In this case, since we are comparing two things, the better (or lower) is the best.

 Happyjew wrote:
Except the rules define lower as better - therefore a 3+ is better than a 4++ as the 3+ is lower. In this case, since we are comparing two things, the better (or lower) is the best.

No, it doesn't. I have the rule right in front of me.
It says that a lower ARMOUR save is better.
So 3+ is better than 4+.
But 3+ is not better than 4++ with a reroll.

I've read that argument and I'm actually amazed that nobody pointed out it spoke about ARMOUR saves and not all saves in general.

Kangodo wrote:

 Happyjew wrote:
Except the rules define lower as better - therefore a 3+ is better than a 4++ as the 3+ is lower. In this case, since we are comparing two things, the better (or lower) is the best.

No, it doesn't. I have the rule right in front of me.
It says that a lower ARMOUR save is better.
So 3+ is better than 4+.
But 3+ is not better than 4++ with a reroll.

I've read that argument and I'm actually amazed that nobody pointed out it spoke about ARMOUR saves and not all saves in general.

Invul saves only list one area they're different - the fact that you can always take them (ie you can't AP them away).
Which means that rules that apply to armor saves also apply to invul saves.

Also, this was covered. In this thread. I guarantee it.

Actually if you read page 19, Models with More than One Save again points to lower save value being better.

 Happyjew wrote:
Actually if you read page 19, Models with More than One Save again points to lower save value being better.

It says that when you have multiple saves you use the best one.
It does not hint to the 'lower value being better'.

rigeld2 wrote:
Invul saves only list one area they're different - the fact that you can always take them (ie you can't AP them away).
Which means that rules that apply to armor saves also apply to invul saves.

Also, this was covered. In this thread. I guarantee it.

Can you quote me the line that says invulnerable-saves should be treated as armour-saves?
My book doesn't have that line.

The 'lower is better' only applies to armour-saves, which is only a small part of all saving throws.

 Elric Greywolf wrote:

 liturgies of blood wrote:
It's 75%. 1/2 chance of passing, half chance of a re-roll. The re-roll is again 1/2 chance of passing. So 1/2 + 1/2 of 1/2 is 3/4.

This is right, although I would notate it as
.5+(.5*.5)=.75 or 1/2+(1/2*1/2)=3/4
It gets significantly harder when you're trying to calculate the odds of one 4+ in a batch of dice, or figure out (as my Psyfleman likes to do) the chance of a pen on a Flyer. Four TL Str8 shots, no Skyfire, AV12. I need 6's to hit, then I need 5+ to pen, he might Evade. Easy enough with one die...confusing with four.

The odds of one 4+ in a batch of dice is straight into your formula. It doesn't matter if you roll them one by one or all in a pile, it's the exact same thing. It's not like your odds of saving 6 wounds would change if you decided to roll them one by one after all
Although I have a friend that would disagree with that on superstitious terms. He always rolls important saves one by one and keeps saving against all odds...

Calculating pens is mostly confusing because it goes through so many different rolls. Not like one set save.
1-(5/6)^2 will give you the odds of one twin linked shot landing, if it hits on 6es.
1-(1-(5/6)^2)^4 should then give you the probability of that happening in 4 such shots (which is to say it's a statistical probability, over 99%, that you'll get atleast one hit in.)
But it's more interesting at this point to instead say "How many hits will I statistically get.
so we say
(1-(5/6)^2)*4 which tells us you'll statistically get 1.22 hits. So that sounds reasonable.
So since there is a 1/3 chance those hits will pen, we can just go and divide it all by 3 to get the chance of getting that penetrating blow.
(1-(5/6)^2)*4/3
That leaves us at a 40.7% chance for a penetrating blow. (but a 61% chance of a pen OR a glancing.)
Of course, this guy was paranoid and decided to evade and gained a 5+ jink when you started showing interest.
So we have to remove 1/3 of the chances, since that's his odds of avoiding it.
(I am getting really tired, so I made a really ugly hack here. The math is fine, but it makes for really ugly equations, implying the rest hasn't been hacked together so far. and I've started adding parenthasis everywhere to keep it apart in my head. There's even one in here that I didn't use just above this.)
((((1-(5/6)^2)*4)/3)/3)*2)

Anyway, that gives us about a 27.1% chance for a penetrating blow. Of course, what we want is the tremendous fireball. That's on a 6, since we're psyflemen and have no frickin bonuses at all to our vehicular damage.
So let's bring in another frickin parenthasis for the hell of it. (Yeah, it's not needed.)
(((((1-(5/6)^2)*4)/3)/3)*2)/6

Well, so that brings us to a neat 4.527% chance of blowing his sh*t out of the skies.

... I don't think the psyfleman was made for aerial combat, to be honest.

Kangodo wrote:

rigeld2 wrote:
Invul saves only list one area they're different - the fact that you can always take them (ie you can't AP them away).
Which means that rules that apply to armor saves also apply to invul saves.

Also, this was covered. In this thread. I guarantee it.

Can you quote me the line that says invulnerable-saves should be treated as armour-saves?
My book doesn't have that line.

The 'lower is better' only applies to armour-saves, which is only a small part of all saving throws.

Invulnerable saves are different to armour saves because they may always be taken whenever the model suffers a Wound- the Armour Piercing value of attacking weapons has no effect.

Now this comprises pretty much all of the rules for invul saves. Meaning without referring to armor saves you have no idea how to use them.
Do you roll a d4, d6, d8, 2d4? Do you need to roll above or below? Is it a hex, decimal, or octal number?
Armor save rules answer all of these.

Happyjew wrote:

Kangodo wrote:

 Happyjew wrote:
Let me ask you this then - which is the better save (not best, just better), 4++ with re-roll or 3+?

Best is a term when you compare two (or more) things.
The one that is better than any other is considered the best.

Seeing as 4++ has a reroll, it is the best.

Except the rules define lower as better - therefore a 3+ is better than a 4++ as the 3+ is lower. In this case, since we are comparing two things, the better (or lower) is the best.

Stating a single aspect of something is better does not make the one with the most of that aspect the best. If more is better than the most is the best is a leap of logic (though quite a natural one) when dealing with a subject that has several aspects to it. While the number is noted as being better if it is lower that does not account for any special rules as they are not mentioned. Either including or disregarding them is in this case not expressly permitted and it is left undefined as to what the 'best' is.

They say the lower number is better. That does not equal the lowest number is best. That would be akin to saying the fastest car is the best car when there is more to it than just top speed. Sure the higher the top speed the better but that does not mean the highest top speed will be best.

You ask if a lower number is better? I say yes.
I ask, is a re-roll option better than a single roll?... you say? No? Does the book really need to spell that out for you?

Obviously a save with a re-roll is better than a save without one.

rigeld2 wrote:

Kangodo wrote:

 Happyjew wrote:
Except the rules define lower as better - therefore a 3+ is better than a 4++ as the 3+ is lower. In this case, since we are comparing two things, the better (or lower) is the best.

No, it doesn't. I have the rule right in front of me.
It says that a lower ARMOUR save is better.
So 3+ is better than 4+.
But 3+ is not better than 4++ with a reroll.

I've read that argument and I'm actually amazed that nobody pointed out it spoke about ARMOUR saves and not all saves in general.

Invul saves only list one area they're different - the fact that you can always take them (ie you can't AP them away).
Which means that rules that apply to armor saves also apply to invul saves.

Also, this was covered. In this thread. I guarantee it.

Technically invulnerable saves are in fact different and they only compare armor saves. Strictly speaking the rule says nothing about them or cover saves for that matter but I will run on the same assumptions you are and agree on this point as it's the only way anything will make sense here.

Sadly the distinction between a 4++ save and a 4++ save and a re-roll vs a 3+ isn't made in the BRB. All the math-hammering behind a 4++ re-roll is metagaming, and while it will increase your odds of saving wounds, the book doesn't take that information into account for determining "best" save.

While mathematically the 4++ is the better save, book asks for you to use the model's "best" save, which it defines as the lowest save. You are not allowed to ignore your armor save if it is better. It is akin to a model suddenly losing its armor in the middle of a fight (failing all saves of a 3 instead of passing) and then have it reappear. While that may work for Chaos, it wouldn't work for any other army. If the Chaos Codex had a special rule allowing you to choose your save, I would have a different opinion.

As it stands, I would say using the 4++ re-roll when you can still use your 3+ save, would be either powergaming or fudging the rules to get an advantage. It is great vs high AP weapons, but useless against bolter fire.

Neither of these saves is affected by the AP of the weapon, so the Captain uses the cover save to give him the best chance of surviving.

This is from page 19 of the BRB, clearly best means best chance of surviving as it is used here.

As well as..

"Invulnerable saves are different to armour saves because they MAY ALWAYS BE TAKEN whenever the model suffers a wound - .."

This is on page 17 of the BRB

[delete]

If I must use a 3+ instead of a 4++, I do not have the advantage of using the best available save, I am using the save at a disadvantage

This is getting heated
I would simply use math to decide which is BEST!
In every sense of the word.
BEST simply means the one with better probability of making the saving throw..

I WISH THERE WAS A POLL FOR THIS FORUM HAHA

Abandon - they define lower as better. By definition the lowest of two is the best of those two.

Your car analogy is flawed - as you havent defined "better". If your determinant is "top speed" and you say higher is better, then two cars, one with a top peed of 100mph and one with a top speed of 150mph, based on the criteria given the 150mph car is the best car of the two.

They absolutely and precisely state where an invulnerable save is different to armour; it is unaffected by AP. As such you MUST use the lower of armour, cover, invulnerable, and are not allowed to use other criteria when detrmining your save

Permissive ruleset. You have no permission, NONE, to pick a reroll 4+ over a 3+.

The rules only tell you what is 'better' or 'best' when comparing Armour Saves to Armour Saves (pg 2.), or Cover Saves to Cover Saves (pg 19.).

When comparing different types of Saves, we are told that a model 'has the advantage of always using the best available save.' And the example says: '. . so the Captain uses the cover save to give him the BEST CHANCE OF SURVIVAL.'

A save that has a lower statistical chance of success is not the 'Best save available' nor does it give a model 'the best chance of survival.'

nosferatu1001 wrote:
Abandon - they define lower as better.

No, they don't.
That has be proven wrong as it only counts for armour saves.

They absolutely and precisely state where an invulnerable save is different to armour

"It's different because", does NOT imply that every other aspect is the same.

As such you MUST use the lower of armour, cover, invulnerable, and are not allowed to use other criteria when detrmining your save

You just made that up and you know it.

Permissive ruleset. You have no permission, NONE, to pick a reroll 4+ over a 3+.

Yes, you do. They give you permission the moment they say that you "pick the best save available".

rigeld2 wrote:
Now this comprises pretty much all of the rules for invul saves. Meaning without referring to armor saves you have no idea how to use them.
Do you roll a d4, d6, d8, 2d4? Do you need to roll above or below? Is it a hex, decimal, or octal number?
Armor save rules answer all of these.

I don't give a gak.
You are saying it follows the rules for armour-saves, so you better quote the rule that says that or take your words back.

I firmly believe the wording of the characteristic descriptions on page 2, which are there to give beginners a feel for what characteristics they will come across and what they mean are being miss used to apply an absolute meaning to a rule which includes, but is not limited to that particular characteristic.

Page 2 goes over all the characteristics found in a model's profile, and gives a very high level view of how the characteristics work. 'SV' characteristic, 'Armour Save' is inclusive and it does state

Unlike other characteristics, the lower the armour save is, the better

which is of course correct, but does not confer an absolute meaning of 'Best' when working out different types of saves, and deciding which save to make, this does not go into modifiers, does not go into re re-rolls, this deals with 1 stat which as a base is better when lower.

Frankly, the interpretation that 'best' has a absolute meaning of 'lower value' contradicts the sentencing of the rule in question.

but has the advantage of always using the best available save

If I must use the lowest value save, I do not have the advantage of using the best available save, I am at the disadvantage of using the best[lowest value] save, for sure 3+ VS 4++ I am disadvantaged by using 3+.