Have we been playing saves with modifiers and rerolls wrong this whole time?

[Thomas Lyons]

So, hear me out.  Here are the save rules:

The last sentence says that re-rolls happen before modifiers are applied (positive or negative).  So what this means is that every time a save is rolled, the following steps occur:

1. Roll dice.  
2. Determine if a reroll happens.
3. Add modifiers.
4. Determine final results.

So, when you don't have re-rolls, the calculation is very straightforward.  Rerolls seem to be a different beast.  So, for example, lets take a Disk Lord that re-rolls 1's and has a +2 to his save.  He would:

1. Roll a save dice.
2. If the result is a 1 on a dice, he rerolls it.  If not, move on to step 3.
3. +2 is added to the dice result.
4. Determine if the dice result is 4 or higher to see if he succeeded.  

Fairly straightforward.  Now, lets give that same Disk Lord a re-roll to all saves. 

1. Rolls a save dice.
2. If the result is a 1-3 on a dice, he rerolls it (wording will determine if units must re-roll failed saves, or "may" re-roll failed saves).  If not, move on to step 3.
3. +2 is added to the dice result.
4. Determine if the dice result is 4 or higher to see if he succeeded.

Still fairly straightforward.  Now, lets take the Tzeentch item that grants the Disk Lord an additional +2 on saves but says he must re-roll successes.  Now he: 

1. Rolls a save dice.
2. If the result is a 1-3 on a dice, move on to step 3.  If a 4+, which would be a success, he must re-roll it.
3. +4 is added to the dice result (+2 from his normal ability, +2 from the Tzeentch Paradoxical shield item).
4. Determine if the dice result is 4 or higher to see if he succeeded.

What this seems to imply is that, with no rerolls, the Disk Lord fails on a 1, succeeds on a 2 and 3, and must re-roll on a 4+.  On the reroll, he will succeed on 2+.  If this Lord has re-roll 1s, then he re-rolls 1s and 4-6s, and is just successful on 2s and 3s.

This is obviously important for any rerolls, including hits and wounds, because this means that more of these dice may need to be rerolled (which would be good if you are fishing for higher numbers to trigger abilities).

Is this correct?  

Edit:  So, it seems any time you have bonuses to a roll and re-roll all failed rolls coupled with an ability to activate on a 6+ that things really get interesting.  Here is an example where this might be significant and used for an advantage.  

Take for example a Plague Monks that wound one 4+ with their Foetid Blades.  Imagine that the Plague Monk is getting +2 to Wound from nearby Plague Priests, has reroll all wounds from a Plague Furnace , and is in the Pestilent Clawpack battalion that says on a Wound of 6+ they do double damage. Normally, we would consider reroll failed wounds would only apply on natural 1s since you have +2 to wound (thus successfully wounding on a 2+ effectively) and you would trigger the double damage on 4+ of the die roll. Under this insight, I could opt to still reroll 2s and 3s, even though they would be successes with modifiers, to fish for more 4+s which will do double damage since rerolls are determined to occur before the modifiers are applied. All of my rerolls would still land on 2+ but rerolling 2s and 3s gives me another bite at that double damage I get from die results of 4+.  This is just one implication of how this would work.  It would work similarly for Stormcast effected by a nearby Lord Castellant's Lantern who are wanting to fish for healing on save rolls and anyone else with a reroll all ability, bonuses, and a triggering ability. Hope this is a bit clearer than my original example.

[Denneysman]

Am I wrong in thinking this same debate pops up every month?

[Mc1gamer]

This is usually easy to figure out, as the rerolls are often based on failures, but this is wacky and forces it on successes. I'm tagging for knowledge base

[Arkiham]

So, the tl:Dr version is.

"Can I reroll rolls before modifiers to proc bonuses on certain numbers then add the modifiers in order to proc them on the rerolled dice."

In short. Yes, it was done in 8th edition to fish for killing blows or poision attacks on various units.

the reason is, it says you can reroll blah blah on the abilities. 

The faq states that you reroll before modifiers.

 

I've been playing it like this all the time, I assumed everyone else did anyway.

[Dave Fraser]

it's clearly just loose GW rules writing.  They have 2 types of reroll:

1. Reroll [specific value] where the are usually 1s which now always fail.  If applied post modifier then often the end result wouldn't be 1 so you'd not get to reroll despite being a failure so they applied the wording above

2. Reroll successes/failures - The success/fail isn't determined until after modifiers so applying the sequencing above leads to some perverse behaviour where you end up rerolling successes.  I'd postulate this isn't the intent but GW are hopeless at writing tight rules as they simply don't work that way so you can apply it as you say but to my mind it's an application of a mechanic at the detriment of the quicker and less intrusive rules of AoS.

[Squirrelmaster]

Daemonettes also have a “re-roll to-hit rolls of 6 or more” which only makes sense if it's done after modifiers.

[wayniac]

TBH I am chalking this up to "GW cannot write clear rules, and writes unclear rules in a conversational tone not an instructional one"

[Lucio]

I think the different wordings trigger subtly different behaviour. Abilities which reroll on a specific number, reroll before modifiers. Those that reroll on the result (i.e. success / fail) reroll after modifiers.

 

This prevents most of the possible abuses, particularly as you can't reroll a reroll

[Squirrelmaster]

So what about the daemonettes' Locus of Beguilement?

[Lucio]

4 hours ago, Squirrelmaster said:

So what about the daemonettes' Locus of Beguilement?

Do seem to fit into the latter catagory, re-roll after modifiers, but the justification isn't great either way.

[BaldoBeardo]

Yeah, semantics aren't great here. Would be better if they distinguished between "roll" and "result".

Sent from my Nexus 5X using Tapatalk

[mmimzie]

This isnt a big deal as most of these abilities are optional. 

 

So you could choice to do the premodifier reroll or not. And maybe gain something with the higher result.

 

The only thing that comes to mind as I've been currently playing it. Is you can chose to tqke rerolls for a pheonix that has been buffed by sisters of the thorn.

 

So sisters of the thorn let you reroll failed saves and all saves of 6+ cause mortal woulfs back to melee opponents.

 

The pheonixs gets +1 armor on a 6+ for spells cast within range of them. Having a base save of 5+.

 

So using the rules. If i have a +2 to my armor and the sisters buff on a pheonix. I could reroll any die that is less than 5 if i wanted to. And try to proc more mortal wounds. Though in most positive cases this isnt a problem.

 

Now where it gets weird is forcing reroll on your opponent. Right now a RAW they only work on unmodified successes. I'd play it as RAW as i do all things until FAQ.

 

Please remember guys though AoS is an amazing game note that some of the rules are kind of 'old' and drawn up half jokingly and not meant for competive scrutiny. It is bad to assume that the point are balanced around an abigous interpretations as the points came after all the rules were written and so we could only assume the point take into account RAW

[Thomas Lyons]

On February 2, 2017 at 4:35 AM, Dave Fraser said:

it's clearly just loose GW rules writing.  They have 2 types of reroll:

1. Reroll [specific value] where the are usually 1s which now always fail.  If applied post modifier then often the end result wouldn't be 1 so you'd not get to reroll despite being a failure so they applied the wording above

2. Reroll successes/failures - The success/fail isn't determined until after modifiers so applying the sequencing above leads to some perverse behaviour where you end up rerolling successes.  I'd postulate this isn't the intent but GW are hopeless at writing tight rules as they simply don't work that way so you can apply it as you say but to my mind it's an application of a mechanic at the detriment of the quicker and less intrusive rules of AoS.

Is this found in the rules at all or are you concluding this just from looking across the system and noticing a pattern?  I ask only because it seems the rules only identify one type of reroll, and it always occurs before modifiers.  Two unqualified rules are established about rerolls: they never happen more than once and rerolls happen before the modifiers.  The distinction does not seem to exist in the rules.

[Dave Fraser]

No, that's not a rule thing, it's an observation thing, the rules make no distinction, there is only 1 type of reroll. 

As written you are entirely correct. But that doesn't make applying that strict adherence to the rules any less of a detriment to the flow of a game over being more streamlined and just rerolling post modified misses when it's not a Target Number reroll. 

 

 

[wayniac]

I think it just makes more sense to have there be two versions:

* Re-rolling a specific number happens before modifiers (e.g. 4+ save, re-rolling 1s with a +1 modifier would re-roll a 1 before it becomes a 2)

* Re-rolling with a generic condition (e.g. "re-roll failed armor saves") happens after modifiers (e.g. 4+ save, re-rolling fails, versus Rend -1 would re-roll on a 4 or less as the save would increase to 5+ and therefore a 4 fails, while without a roll of 4 would not trigger the re-roll but then fail the save when the Rend is applied)

This way it's not much more complicated and is more intuitive.

[Squirrelmaster]

It's certainly come up a few times before. Once-a-month sounds about right to me.

[ragnar]

So,if i am getting this correctly

If i have  a save value of 4+ and i have  reroll failed saves from sisters of the  thorn for example

and the enemy model has rend -1

When i roll my saves any 4s count for succesfull rolls and thats because i havent applied the modifier for rend ,so i  cant reroll it because it is considered succesfull  and i can reroll only 1s,2s,and 3s.

After that i aplly the modifier of -1 rend and all 4s are failures.

Am i saying  it correctly

[Thomas Lyons]

19 minutes ago, ragnar said:

So,if i am getting this correctly

If i have  a save value of 4+ and i have  reroll failed saves from sisters of the  thorn for example

and the enemy model has rend -1

When i roll my saves any 4s count for succesfull rolls and thats because i havent applied the modifier for rend ,so i  cant reroll it because it is considered succesfull  and i can reroll only 1s,2s,and 3s.

After that i aplly the modifier of -1 rend and all 4s are failures.

Am i saying  it correctly

Correct.

 

 

[Celestant651]

Yeah same thing for fumlminators, as they have a 3 up save, but add 1 to all save rolls in the shooting phase, and they can re-roll rolls of 1.

[hobgoblinclub]

On 2017-2-13 at 5:08 PM, Thomas Lyons said:

Correct.

 

 

Then is surely a mistake in rules writing. It's a needless complication that'll force everyone to change the way they play. 

Instead of rolling dice and dividing them up into successes and failures, it'll have to be successes, and failures I can reroll and failures I can't. It'll get silly. 

[wayniac]

On 2/13/2017 at 0:08 PM, Thomas Lyons said:

Correct.

 

 

This makes it MORE confusing because in this case the re-roll is actually detrimental to you.  In the bonus example, it makes sense to re-roll before modifiers, but in the case of a negative, re-roll before modifiers actually hurts you MORE because it would mean a roll of a 4 does NOT trigger the re-roll even though it fails.  In fact, I've always played it the opposite, that in this example you WOULD get the re-roll because you failed the save.

I think this is partially why everyone plays it as modifying the stat, not the die roll, because it makes more sense when it comes to things like this.  You have a 4+ save and get hit with something that has Rend -1; you save on a 5+ so a 4+ triggers your re-roll because it fails.  Otherwise, negatives are worse than positives and there's no consistency.

[Squirrelmaster]

2 hours ago, wayniac said:

This makes it MORE confusing because in this case the re-roll is actually detrimental to you.  In the bonus example, it makes sense to re-roll before modifiers, but in the case of a negative, re-roll before modifiers actually hurts you MORE because it would mean a roll of a 4 does NOT trigger the re-roll even though it fails.  In fact, I've always played it the opposite, that in this example you WOULD get the re-roll because you failed the save.

I think this is partially why everyone plays it as modifying the stat, not the die roll, because it makes more sense when it comes to things like this.  You have a 4+ save and get hit with something that has Rend -1; you save on a 5+ so a 4+ triggers your re-roll because it fails.  Otherwise, negatives are worse than positives and there's no consistency.

The re-roll isn't detrimental; If you have a 4+ save and are hit with a -1 rend then a natural 4 falls anyway.

[wayniac]

5 minutes ago, Squirrelmaster said:

The re-roll isn't detrimental; If you have a 4+ save and are hit with a -1 rend then a natural 4 falls anyway.

But doesn't the re-roll apply before the rend?  So a 4 would not trigger the re-roll, then after that the -1 is applied to make it a 3 and you fail the save?

[Squirrelmaster]

1 hour ago, wayniac said:

But doesn't the re-roll apply before the rend?  So a 4 would not trigger the re-roll, then after that the -1 is applied to make it a 3 and you fail the save?

Yes, but if you didn't have the re-roll, you would have failed anyway.

It's not “using this interpretation makes having a re-roll detrimental (compared to not having a re-roll)”, it's “having a re-roll makes using this interpretation detrimental (compared to using the other interpretation)”. Do you see the difference?

[Thomas Lyons]

1 hour ago, wayniac said:

But doesn't the re-roll apply before the rend?  So a 4 would not trigger the re-roll, then after that the -1 is applied to make it a 3 and you fail the save?

Correct.  What it says is that Rend is actually better than many people have accorded it because it can create pockets in the dice range (when there aren't bonuses) where "reroll all failed saves" doesn't actually give a reroll.  This is one reason why the Paradoxical Shield is so interesting (btw, I brought this issue up in this very forum back in January when DOT was released).  

It is worth pointing out that this also has another benefit that most people haven't recognized.  Since rerolls come before modifiers, rerolling "failed" hits or wounds (before modifiers) could allow folks to reroll a "failure" (that would be a hit or wound with a modifier) in the instances where they are fishing for certain numbers (i.e. roll a 6+ abilities).  This might give folks more bites at activating these abilities than they currently do.  This is one potential side benefit of following the rules as written.

Now, is it counter intuitive?  Maybe, if you don't understand the order of operations.  Regardless, it is how the rules are currently written.  Otherwise, you get a situation where "reroll 1s to save" don't work when you get a modifier.  Every ability like this would need to be rewritten if you wanted to reorganize the order of operations (to gain bonuses and penalties before rerolls).  I will say, this does weaken Kurnoth Hunters and those units naturally sitting on rerolls.  If people had been playing this correctly, we might not have had the outcry that we did on these units.  

So, despite all of this, no one that I know is actually playing these rules accurately, even any of the folks I played at ACON.  I haven't raised a stink about this because when I brought it back up in February, most people seemed to either not understand the situation or they felt like it was counterintuitive and ignore the discussion.  I'm not sure what the fix is beyond GW offering a FAQ and/or a major rules rewrite. 

[Nico]

It does make rend (and especially rend -2 or -3) significantly better against some of the exceptionally good armour saves we see out there, which is arguably no bad thing.