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.
2025/01/13 02:46:23
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
catbarf wrote: Seems like a non sequitur to me. Did leafblower Guard in 5th suffer for lack of 3+ saves as they dominated tournaments? Have you never seen a reasonably balanced game where different factions have different capabilities and limitations?
Well rather the reverse, Leafblower dominated because they had more AP3 firepower than any other army and that let them counter the all-pervasive MEQs. Plus large blasts weren't bad against hordes in a pinch.
(Leafblower was a defense-skew as well but it was in form of AV12 vehicles, not an armor save.)
Anuvver fing - when they do sumfing, they try to make it look like somfink else to confuse everybody. When one of them wants to lord it over the uvvers, 'e says "I'm very speshul so'z you gotta worship me", or "I know summink wot you lot don't know, so yer better lissen good". Da funny fing is, arf of 'em believe it and da over arf don't, so 'e 'as to hit 'em all anyway or run fer it.
2025/01/13 10:29:53
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
Orkeosaurus wrote: Well rather the reverse, Leafblower dominated because they had more AP3 firepower than any other army and that let them counter the all-pervasive MEQs. Plus large blasts weren't bad against hordes in a pinch.
For reference - as a lot of people won't know what leafblower was.
5e guard, immediately after Cruddace had chopped down prices, upped weapon availability, allowed multiple heavy units in single slots, moved elite and fast units into troops and turned chimeras into discount pillboxes.
2500pts in an era where a 'normal' game would be around 1750 due to many factions not being able to spend efficiently into the higher points (limited slots, no ability to double and triple up like guard).
~d3+9 blasts when most factions were capped at 3. No armour saves, all toughness 3, reliant on the newly harder to kill vehicles and 5es high (4++) cover saves, and simply being too cheap to trade out.
Typical composition:
-two command squads with 6 plasma guns, feel no pain, chimeras, and the ability to screw with reserves and drop orbital barrages (the only AP3 weapon in the list)
-allied inquisitor to intercept reserves, psyker squad in chimera (another pieplate blast)
-cheap infantry and special weapon infantry with flamers and heavy weapons (autocannon or mortars), more chimeras - the list could build something of a wall against the hordes and just flame out of it
-veteran (but still dirt cheap) infantry with meltaguns, more chimeras, and demolition charges more powerful than heavy artillery because Cruddace
-medusas - more large blasts this time strength 10 AP2
-manticore - more strength 10 large blasts
-valkyries - more more large blasts, this time chaff clearers
-and finally multiple hydras as they ignored cover saves for fast moving vehicles (bikes) allowing them to beat up on some key xenos troop units
Most common save: 5+, most common weapon AP: 4 (but also more than twice as many strength 8-10 AP2 large blasts than anyone else could field)
~70-ish infantry, ~17 vehicles.
The high points level alone hampered a lot of older lists, the force org chart was built to limit how many stones and scissors you could bring to your paper/scissors/stones game and guard turned every slot into a squadron while handing out mobility and hand-held artillery for peanuts. The dozen or so forward vehicle hulls were just tough enough to shrug off the hand to hand and small arms of factions without grenades creating a wall and subsequently difficult terrain between the armies and the whole thing just shelled you off the board by being cheaper and better armed than every other book.
In an era of codex creep that saw such munchkins as the 5e Grey Knights the guard book managed to stay top tier as just the second release of the edition.
This message was edited 1 time. Last update was at 2025/01/13 12:23:03
2025/01/13 12:18:54
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
catbarf wrote: Seems like a non sequitur to me. Did leafblower Guard in 5th suffer for lack of 3+ saves as they dominated tournaments? Have you never seen a reasonably balanced game where different factions have different capabilities and limitations?
Automatically Appended Next Post:
tauist wrote: I am just trying to wrap my head around the probabilities of staggered diceroll mechanics. It goes against common sense that if you have a process of elimination, where each stage drops unqualifying attempts, that it would not matter if the first trials were more difficult, and the later ones easier, than the other way round..
I bake 40 cookies. My mom takes 1/2 of the cookies. I now have 20 cookies. You take 3/4 of what remains. I have 5 cookies.
I bake 40 cookies. My mom takes 3/4 of the cookies. I now have 10 cookies. You take 1/2 of what remains. I have 5 cookies.
It doesn't matter what order you apply the downselections in.
yeah, that's cookies. but does this take into account that the chance of 6+ with 6 die is only 42%, not 100%? ie. the amount of die affect non-linearly the probabilities of each step..
This message was edited 3 times. Last update was at 2025/01/13 12:30:47
"The larger point though, is that as players, we have more control over what the game looks and feels like than most of us are willing to use in order to solve our own problems"
2025/01/13 12:33:46
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
That's a different issue. It is a real balance issue that makes low reliability units like Orks even worse in practice than in paper, but doesn't has much to do with the order of operations.
2025/01/13 12:34:30
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
catbarf wrote: Seems like a non sequitur to me. Did leafblower Guard in 5th suffer for lack of 3+ saves as they dominated tournaments? Have you never seen a reasonably balanced game where different factions have different capabilities and limitations?
Automatically Appended Next Post:
tauist wrote: I am just trying to wrap my head around the probabilities of staggered diceroll mechanics. It goes against common sense that if you have a process of elimination, where each stage drops unqualifying attempts, that it would not matter if the first trials were more difficult, and the later ones easier, than the other way round..
I bake 40 cookies. My mom takes 1/2 of the cookies. I now have 20 cookies. You take 3/4 of what remains. I have 5 cookies.
I bake 40 cookies. My mom takes 3/4 of the cookies. I now have 10 cookies. You take 1/2 of what remains. I have 5 cookies.
It doesn't matter what order you apply the downselections in.
yeah, that's cookies. but does this take into account that the chance of 6+ with 6 die is only 42%, not 100%? ie. the amount of die directly affect the probabilities of each step..
Just trying to make sure I've internalized this correctly
The expected value does yes, it’s a sum of all the possible outcomes multiplied by their likelihood.
There are other measures of how ‘swingy’ something is such as the standard deviation.
This message was edited 1 time. Last update was at 2025/01/13 12:35:13
2025/01/13 12:38:56
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
If anyone has any links pointing to indepth tutorial on how to properly calculate these sort of things, warts and all, I'd appreciate it. I should have paid more attention to long math at school instead of daydreaming about playing Epic Space Marine hehehe!
"The larger point though, is that as players, we have more control over what the game looks and feels like than most of us are willing to use in order to solve our own problems"
2025/01/13 13:18:53
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
Tyran wrote: That's a different issue. It is a real balance issue that makes low reliability units like Orks even worse in practice than in paper, but doesn't has much to do with the order of operations.
Not necessarily worse but yes, more swingy.
But average mathhammer (I.E. I hit on 3s, wound on 3s, and they get a 5+ save = 2/3*2/3*2/3=8/27) has always been just about good enough, since you are mostly interested in comparisons.
Because its not really about units - or indeed about armies. Its about points. If my 200 point unit is considerably better than your 200 point unit, I have an advantage. If my 2000 point army is considerably better than your 2000 point army, I have an advantage. I'm "effectively" playing with more points than you are - so I have an advantage.
And this in turn why points don't have to be perfect - and you don't need to agonize over the value of bolt pistols vs las pistols on characters. Take the best army in the game - and say it has to leave a two units behind. Take the worst army and say they can get 2 extra units. There's been some moments when the game is so uninteractive it perhaps wouldn't matter - but odds are you'd have a much closer game. In part because a 4 unit swing should make it much easier for the weaker army to play objectives if nothing else.
3rd-7th had more janky rules that perhaps made estimating the value of a unit for its points compared with the rest of 40k for its points harder. But I'm not really convinced it felt that way if you were steeped in the game.
For example "on paper" Dissie Ravagers should have been okay in 7th - as you'd think 9 S5 AP2 shots were quite lethal even without rerolling 1s to hit and wound. Unfortunately you had the defensive profile of a 35 point Rhino (worse actually, because of Open Topped) so you died if anything half sensible looked at you funny. Sure you could take Night Shields and hope your opponent shot you through cover - but its unlikely to be an issue for deep striking suicide melta.
Leafblower was busted because at the time it brought an unprecedented level of firepower for the points - and especially if it went first, just deleted armies. But this isn't really any more profound than Orks and Ad Mech doing so in early 9th edition (or frankly about half the armies in 9th without L-shaped ruins everywhere given some slightly above average dice.)
In terms of the probability chat, for first order mathhammer you are concerned with averages for comparison. If for example my unit "averages" out to a 40% points return on a shooting phase (and it will go higher or lower any time I roll the actual dice) versus a range of targets - while your unit only averages out to a 25% points return, then my unit is almost certainly better, with some caveats. And if you build a whole list on that basis, odds are your whole army will end up being better.
The second order stuff is more useful for trying to calculate the odds of something happening.
So for example the odds of rolling one 6 or more on 6 dice is 1-(5/6^6). Which is about 2/3rds. You can in turn use this sort of logic to calculate what your chances are for say unit X to kill unit Y by shooting. If you have say an 85% chance, you can feel fairly confident - and let down by dice if you fail. If however its only a 50% chance, then you may want to gamble (or have to given what else has happened on the table) - but its probably sensible to allocate another unit to potentially help out the first and ensure the kill.
This is where you can argue there's some value in units that have greater interactivity - for example with longer range on their guns. So to use the above - if the second unit sent to help the first only has 6" ranged weapons - and the first unit successfully destroys its target, they now may have nothing to shoot at and have wasted a shooting phase. Whereas if they have say 30" guns they can hit a lot of the table - and therefore keep their options open if other units roll well or roll badly. But arguably in turn the objectives of modern editions have somewhat undermined the benefits of range, since you need to gravitate towards the objectives - so odds are there will be something to shoot unless you are turtling in your deployment zone or everything nearby is already dead.
2025/01/13 15:08:43
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
3rd-7th had more janky rules that perhaps made estimating the value of a unit for its points compared with the rest of 40k for its points harder. But I'm not really convinced it felt that way if you were steeped in the game.
That's probably true for 3rd and 4th. I'm doubtful it remained true during 5th and it certainly wasn't true anymore durin 6th and 7th that were peak uninteractive rules.
This message was edited 2 times. Last update was at 2025/01/13 15:13:36
2025/01/13 16:38:03
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
Even during 3rd/3.5 when I started collecting properly it felt janky. Fleet of Foot being a 'patch' rule because they didn't include a movement stat so they had to have a special rule to represent moving faster, and another for moving slower.
hello
2025/01/13 16:59:15
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
Tyran wrote: ... Get a book in probability and statistics?
It is a whole academic subject on itself, which is probably half the reason GW sucks at balance as they don't want to hire actual mathematicians.
It's not a bad idea per se, but I am basically only interested in learning how to calculate odds for D6 rolls of specific number of die, against specific staggered tests, in the most accuracy possible. There are so many books in my "to study" list already, ranging from electronics engineering and acoustics to programming, that I'd prefer to keep things super focused on this subject instead of having to learn about the subject in lenght
"The larger point though, is that as players, we have more control over what the game looks and feels like than most of us are willing to use in order to solve our own problems"
2025/01/13 19:18:10
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
Tyran wrote: ... Get a book in probability and statistics?
It is a whole academic subject on itself, which is probably half the reason GW sucks at balance as they don't want to hire actual mathematicians.
It's not a bad idea per se, but I am basically only interested in learning how to calculate odds for D6 rolls of specific number of die, against specific staggered tests, in the most accuracy possible. There are so many books in my "to study" list already, ranging from electronics engineering and acoustics to programming, that I'd prefer to keep things super focused on this subject instead of having to learn about the subject in lenght
Then I'd keep it simple and focus on the average outcome. You can't predict the future but it gives you an expected result. To keep it really basic:
For calculating odds:
6+ = 1/6 = 17%, or 0.17
5+ = 2/6 = 33%, or 0.33
4+ = 3/6 = 50%, or 0.5
3+ = 4/6 = 67%, or 0.67
2+ = 5/6 = 83%, or 0.83
1+ = 6/6 = 100%, or 1.0
Use this when mathing out what happens when unit x attacks unit y. You group all shots for the same weapons together.
Say I have 10 conscripts with lasguns in rapid fire range, and one has a plasma gun. They're shooting some termagants (T3, sv 6+, 1 wound).
9 conscripts with lasguns have 18 shots in rapid fire range, hit on 5+ and wound T3 targets on a 4+ (some of these stats may be off, doesn't matter).
Grab a calculator; we'll have to multiply number of shots with roll to hit, roll to wound and save to find out what happens. You'll need to use the 0.XX numbers provided above corresponding to the required dice roll as input. You'll also have to invert the target's armor save. A 6+ save means the opponent has a 6+ (1/6, 0.17) chance of passing the save, but we're interested in the 5/6 chance of them failing the save! So:
18 shots x 5+ (to hit) x 4+ (to wound) x 2+ (5/6 chance to bypass the save)
Looking at the table above that gives us the following input for the calculator:
18 x 0.33 x 0.5 x 0.83
If you hit enter that should give 2.5.
The plasma gunner shoots too (s7 and enough ap to ignore the 6+ save)
2 shots x 5+ to hit x 2+ to wound x 6/6 (100% success rate; there's no save left to invert!)
Translates to
2 x 0.33 x .83 x 1
The result is 0.55
Add the 2.5 from the lasguns and the conscripts kill a total of 3.05 gaunts. Round this up or down to the closest whole number (but only in the final step!) to get a ballpark expected result. So 3 in this case.
Note that this is the average result. You can roll ten dice and have them all come up as 1, it happens, but if you made the same roll 100 times the average result is what you'd see most of the time.
It's best to assign some "overkill" to protect against poor rolls, so if you want to kill 3 termagants you might assign enough extra lasgunners from another squad to kill a 4th gant. This reduces the odds of the dice gods screwing you over.
This can get more convoluted with gw's love for rerolls, partial rerolls, more rerolls, modifiers and whatnot, but this is the base of it all.
This message was edited 2 times. Last update was at 2025/01/13 19:21:01
2025/01/13 20:11:52
Subject: Re:The AP System. Fundamentally flawed, or just poorly implemented?
They showed you the most basic average mathhammer. If you want more detailed stats, I suggest looking at https://anydice.com/ that will allow you to really calculate all shorts of percentages of rolls and results.
2025/01/13 22:25:18
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
So is there a way to do a "hybrid" system? Something like binary but with the possibility of reducing the enemy save by 1 from certain weapons/circumstances? I certainly can see why -1AP on a heavy rifle makes sense against a guardsmen but is ludicrous against a leman russ, in terms of using the current AP system.
I've only ever played a game system once that didn't have the shot-at player rolling saves (Xenos Rampant I think it was) and I can certainly see why they are a thing. It's just much more boring when the enemy rolls two dice then you have to remove models. Getting to participate in a roll is better.
This message was edited 2 times. Last update was at 2025/01/13 22:28:10
2025/01/13 22:28:33
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
RustyNumber wrote: So is there a way to do a "hybrid" system? Something like binary but with the possibility of reducing the enemy save by 1 from certain weapons/circumstances?
I've only ever played a game system once that didn't have the shot-at player rolling saves (Xenos Rampant I think it was) and I can certainly see why they are a thing. It's just much more boring when the enemy rolls two dice then you have to remove models. Getting to participate in a roll is better.
There have been a few ideas discussed over the years.
One that comes up a lot is:
AP < Sv = Full save
AP = Sv = -1 to save
AP > Sv = No save
RustyNumber wrote: So is there a way to do a "hybrid" system? Something like binary but with the possibility of reducing the enemy save by 1 from certain weapons/circumstances? I certainly can see why -1AP on a heavy rifle makes sense against a guardsmen but is ludicrous against a leman russ, in terms of using the current AP system.
I've only ever played a game system once that didn't have the shot-at player rolling saves (Xenos Rampant I think it was) and I can certainly see why they are a thing. It's just much more boring when the enemy rolls two dice then you have to remove models. Getting to participate in a roll is better.
JNAProductions wrote: I prefer the binary system to the modifier system. Is that nostalgia talking? Probably.
That said! I do think modifiers have a place even in the binary system. Make a rule, High Impact [X] or whatever you want to call it. X is a penalty applied to armor save rolls made against this weapon.
So a Heavy 3 S5 AP4 Heavy Bolter hitting on a 3+ does 4/9ths a wound to MEQ. A Heavy 2 S7 AP4 High Impact [1] hitting on a 3+ Autocannon does 5/9ths.
A Power Weapon (which I am fine being a generic profile-Axes, Swords, and Mauls don't REALLY need differentiation) could be S:User or S:+1 AP3 High Impact [1].
See quote.
This message was edited 1 time. Last update was at 2025/01/13 22:32:08
Clocks for the clockmaker! Cogs for the cog throne!
2025/01/13 22:47:33
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
I think the natural solution is to move the whole system to D10 rather than D6, up armor values so there's more of a spread between a space marine, a terminator, a guardsman, and an eldar guardian. Then you can up the AP value of specific weapons and have it actually mean something. Provides more granularity but also more flexibility for designers. Will it happen? Nope, but I think its something to consider. We also said they'd never move from Hull Points to wounds and I think that has really refreshed the game immensely for vehicles.
RustyNumber wrote: So is there a way to do a "hybrid" system? Something like binary but with the possibility of reducing the enemy save by 1 from certain weapons/circumstances?
Rending kind of did. Of every four wounds inflicted against a marine by a weapon like an assault cannon (S6 AP4 rending) you would get three 3+ saves and one non-save (2 wounds), rather than four 4+ saves (2 wounds).
Lot of other circumstances where those numbers don't match the -1 save mod of course, and even that example is only accurate with mid edition rending and not 3e rending (on hit rather than on wound).
2025/01/14 01:46:17
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
There were also the Ork Choppas which just reduced any save better than 4 to 4, which was brutal. Marines got reduced to 4+, but so did Terminator 2+s. Those Boyz were proper scary.
Insectum7 wrote: There were also the Ork Choppas which just reduced any save better than 4 to 4, which was brutal. Marines got reduced to 4+, but so did Terminator 2+s. Those Boyz were proper scary.
That I find rather weird.
It's cool, but that'd feel more at home on some esoteric weapon rather than Choppas.
Clocks for the clockmaker! Cogs for the cog throne!
2025/01/14 01:51:39
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
Insectum7 wrote: There were also the Ork Choppas which just reduced any save better than 4 to 4, which was brutal. Marines got reduced to 4+, but so did Terminator 2+s. Those Boyz were proper scary.
That I find rather weird.
It's cool, but that'd feel more at home on some esoteric weapon rather than Choppas.
At the time, orks were S3 (and stayed S3 for quite a while). The choppa rule offset that somewhat. Still, even on the charge 20 slugga boyz would hit marines 40 times, wound ~16 times and kill 5-6. With choppas they killed 8.
To avoid the jankiness of terminators it probably could have been a -1 to save, rather than a flat value.
blockade23 wrote: I think the natural solution is to move the whole system to D10 rather than D6, up armor values so there's more of a spread between a space marine, a terminator, a guardsman, and an eldar guardian. Then you can up the AP value of specific weapons and have it actually mean something. Provides more granularity but also more flexibility for designers. Will it happen? Nope, but I think its something to consider. We also said they'd never move from Hull Points to wounds and I think that has really refreshed the game immensely for vehicles.
Go back a few pages
2025/01/14 02:22:10
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
Insectum7 wrote: There were also the Ork Choppas which just reduced any save better than 4 to 4, which was brutal. Marines got reduced to 4+, but so did Terminator 2+s. Those Boyz were proper scary.
That I find rather weird.
It's cool, but that'd feel more at home on some esoteric weapon rather than Choppas.
At the time, orks were S3 (and stayed S3 for quite a while). The choppa rule offset that somewhat. Still, even on the charge 20 slugga boyz would hit marines 40 times, wound ~16 times and kill 5-6. With choppas they killed 8.
To avoid the jankiness of terminators it probably could have been a -1 to save, rather than a flat value.
I don't question whether or not it was EFFECTIVE-because I get the mechanical idea.
It just feels narratively weird, that's all. Was there an explanation for why it was so?
Clocks for the clockmaker! Cogs for the cog throne!
2025/01/14 02:33:25
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
I think just because they were big and heavy. As likely to smash as to chop.
The philosophy of 3rd ed mechanics was all about effective design rather than narrative. Space Marines recieving no benefit from cover against small arms is the prime example. Shuriken Catapults becoming short ranged, but having Assault so the wielders were free to maneuvre and Assault with no loss to firepower is another. No split-fire for Heavy weapons, forcing hard choices, etc.
Tyran wrote: ... Get a book in probability and statistics?
It is a whole academic subject on itself, which is probably half the reason GW sucks at balance as they don't want to hire actual mathematicians.
It's not a bad idea per se, but I am basically only interested in learning how to calculate odds for D6 rolls of specific number of die, against specific staggered tests, in the most accuracy possible. There are so many books in my "to study" list already, ranging from electronics engineering and acoustics to programming, that I'd prefer to keep things super focused on this subject instead of having to learn about the subject in lenght
Then I'd keep it simple and focus on the average outcome. You can't predict the future but it gives you an expected result. To keep it really basic:
For calculating odds:
6+ = 1/6 = 17%, or 0.17
5+ = 2/6 = 33%, or 0.33
4+ = 3/6 = 50%, or 0.5
3+ = 4/6 = 67%, or 0.67
2+ = 5/6 = 83%, or 0.83
1+ = 6/6 = 100%, or 1.0
Use this when mathing out what happens when unit x attacks unit y. You group all shots for the same weapons together.
Say I have 10 conscripts with lasguns in rapid fire range, and one has a plasma gun. They're shooting some termagants (T3, sv 6+, 1 wound).
9 conscripts with lasguns have 18 shots in rapid fire range, hit on 5+ and wound T3 targets on a 4+ (some of these stats may be off, doesn't matter).
Grab a calculator; we'll have to multiply number of shots with roll to hit, roll to wound and save to find out what happens. You'll need to use the 0.XX numbers provided above corresponding to the required dice roll as input. You'll also have to invert the target's armor save. A 6+ save means the opponent has a 6+ (1/6, 0.17) chance of passing the save, but we're interested in the 5/6 chance of them failing the save! So:
18 shots x 5+ (to hit) x 4+ (to wound) x 2+ (5/6 chance to bypass the save)
Looking at the table above that gives us the following input for the calculator:
18 x 0.33 x 0.5 x 0.83
If you hit enter that should give 2.5.
The plasma gunner shoots too (s7 and enough ap to ignore the 6+ save)
2 shots x 5+ to hit x 2+ to wound x 6/6 (100% success rate; there's no save left to invert!)
Translates to
2 x 0.33 x .83 x 1
The result is 0.55
Add the 2.5 from the lasguns and the conscripts kill a total of 3.05 gaunts. Round this up or down to the closest whole number (but only in the final step!) to get a ballpark expected result. So 3 in this case.
Note that this is the average result. You can roll ten dice and have them all come up as 1, it happens, but if you made the same roll 100 times the average result is what you'd see most of the time.
It's best to assign some "overkill" to protect against poor rolls, so if you want to kill 3 termagants you might assign enough extra lasgunners from another squad to kill a 4th gant. This reduces the odds of the dice gods screwing you over.
This can get more convoluted with gw's love for rerolls, partial rerolls, more rerolls, modifiers and whatnot, but this is the base of it all.
I appreciate you writing all that out for me, but the "simplified" method of calculating odds is not enough to satisfy my curiousity. I want to see how the odds change when the amount of die change, in order to gain insight into how much "dice power" weights the outcomes towards succcess/failure, and the simplified system just doesnt account for that.
Automatically Appended Next Post:
alextroy wrote: They showed you the most basic average mathhammer. If you want more detailed stats, I suggest looking at https://anydice.com/ that will allow you to really calculate all shorts of percentages of rolls and results.
Thank you! I'll take a look.
This message was edited 2 times. Last update was at 2025/01/14 08:28:55
"The larger point though, is that as players, we have more control over what the game looks and feels like than most of us are willing to use in order to solve our own problems"
2025/01/14 08:43:28
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
Tyran wrote: ... Get a book in probability and statistics?
It is a whole academic subject on itself, which is probably half the reason GW sucks at balance as they don't want to hire actual mathematicians.
It's not a bad idea per se, but I am basically only interested in learning how to calculate odds for D6 rolls of specific number of die, against specific staggered tests, in the most accuracy possible. There are so many books in my "to study" list already, ranging from electronics engineering and acoustics to programming, that I'd prefer to keep things super focused on this subject instead of having to learn about the subject in lenght
Then I'd keep it simple and focus on the average outcome. You can't predict the future but it gives you an expected result. To keep it really basic:
For calculating odds:
6+ = 1/6 = 17%, or 0.17
5+ = 2/6 = 33%, or 0.33
4+ = 3/6 = 50%, or 0.5
3+ = 4/6 = 67%, or 0.67
2+ = 5/6 = 83%, or 0.83
1+ = 6/6 = 100%, or 1.0
Use this when mathing out what happens when unit x attacks unit y. You group all shots for the same weapons together.
Say I have 10 conscripts with lasguns in rapid fire range, and one has a plasma gun. They're shooting some termagants (T3, sv 6+, 1 wound).
9 conscripts with lasguns have 18 shots in rapid fire range, hit on 5+ and wound T3 targets on a 4+ (some of these stats may be off, doesn't matter).
Grab a calculator; we'll have to multiply number of shots with roll to hit, roll to wound and save to find out what happens. You'll need to use the 0.XX numbers provided above corresponding to the required dice roll as input. You'll also have to invert the target's armor save. A 6+ save means the opponent has a 6+ (1/6, 0.17) chance of passing the save, but we're interested in the 5/6 chance of them failing the save! So:
18 shots x 5+ (to hit) x 4+ (to wound) x 2+ (5/6 chance to bypass the save)
Looking at the table above that gives us the following input for the calculator:
18 x 0.33 x 0.5 x 0.83
If you hit enter that should give 2.5.
The plasma gunner shoots too (s7 and enough ap to ignore the 6+ save)
2 shots x 5+ to hit x 2+ to wound x 6/6 (100% success rate; there's no save left to invert!)
Translates to
2 x 0.33 x .83 x 1
The result is 0.55
Add the 2.5 from the lasguns and the conscripts kill a total of 3.05 gaunts. Round this up or down to the closest whole number (but only in the final step!) to get a ballpark expected result. So 3 in this case.
Note that this is the average result. You can roll ten dice and have them all come up as 1, it happens, but if you made the same roll 100 times the average result is what you'd see most of the time.
It's best to assign some "overkill" to protect against poor rolls, so if you want to kill 3 termagants you might assign enough extra lasgunners from another squad to kill a 4th gant. This reduces the odds of the dice gods screwing you over.
This can get more convoluted with gw's love for rerolls, partial rerolls, more rerolls, modifiers and whatnot, but this is the base of it all.
I appreciate you writing all that out for me, but the "simplified" method of calculating odds is not enough to satisfy my curiousity. I want to see how the odds change when the amount of die change, in order to gain insight into how much "dice power" weights the outcomes towards succcess/failure, and the simplified system just doesnt account for that.
Automatically Appended Next Post:
alextroy wrote: They showed you the most basic average mathhammer. If you want more detailed stats, I suggest looking at https://anydice.com/ that will allow you to really calculate all shorts of percentages of rolls and results.
Thank you! I'll take a look.
Ah, yeah then a dice roller like the one alextroy linked is the better fit. Have fun mathhammering
This message was edited 1 time. Last update was at 2025/01/14 08:44:14
2025/01/14 10:02:18
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
JNAProductions wrote: It's cool, but that'd feel more at home on some esoteric weapon rather than Choppas.
Chaos 3.5 had it as well as a 1pt upgrade to chainswords on khorne units.
Replaced in both cases with furious charge in later editions.
A number of rules got more or less inclusive over the editions - rending lost its 'ignore wound roll' ability making it less effective against monsters, poison gained an added effect against low toughness creatures rather than being only good against monsters, sniper changed to use the firers BS score, etc.
2025/01/14 20:45:49
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
I appreciate you writing all that out for me, but the "simplified" method of calculating odds is not enough to satisfy my curiousity. I want to see how the odds change when the amount of die change, in order to gain insight into how much "dice power" weights the outcomes towards succcess/failure, and the simplified system just doesnt account for that.
As in you want to know how the bell curve changes based on how many dice are being rolled? As previously mentioned, I think you want to look into "standards of deviation."
Automatically Appended Next Post:
blockade23 wrote: I think the natural solution is to move the whole system to D10 rather than D6, up armor values so there's more of a spread between a space marine, a terminator, a guardsman, and an eldar guardian. Then you can up the AP value of specific weapons and have it actually mean something. Provides more granularity but also more flexibility for designers. Will it happen? Nope, but I think its something to consider. We also said they'd never move from Hull Points to wounds and I think that has really refreshed the game immensely for vehicles.
This is a common suggestion, but it ultimately doesn't really do much except give you more granularity. So let's say you double the die size of 40k to d12. Now you can have a +1 to-hit be a ~8.3% boost to the chances of a given to-hit roll succeeding instead of a ~16.7% chance. Any even-numbered modifier is the same % change as a modifier on a d6 (that is, +4 on a d12 is the equivalent of +2 on a d6). Any odd-numbered modifier on a d12 is only ~8% off of a value you could already land on using a d6. So unless you think the magic ingredient missing to fix the AP system is that extra 8% of granularity, this change doesn't accomplish much. That is, a shuriken catapult at AP-1 could go from making a marine fail a save 50% of the time to merely making them fail their save about %42 of the time.
And in exchange for that small boost in granularity, you have to make everyone throw out their d6s and go source an equal number of d12s. Which would be a pain for a number of small reasons.
That said, where the added granularity could help is just in the "feel" of certain rules. For instance, we could probably go back to having stacking to-hit modifiers without the math getting too crazy too fast. Functionally, the overall value of an odd number on a d12 would only be about an 8% difference from an even number, but you could have barrel rolls and cover and the enemy moving before shooting a heavy weapon all be factored into that roll without making marines hit on 6+ and guardsmen on 7+.
This message was edited 1 time. Last update was at 2025/01/14 20:57:20
ATTENTION. Psychic tests are unfluffy. Your longing for AV is understandable but misguided. Your chapter doesn't need a separate codex. Doctrines should go away. Being a "troop" means nothing. This has been a cranky service announcement. You may now resume your regularly scheduled arguing.
2025/01/14 21:14:03
Subject: The AP System. Fundamentally flawed, or just poorly implemented?
I think stacking negative modifiers could still work as long as there are potential stacking positive modifiers, and a reintroduction of the 7+ rol potential. (Roll a D6 and on a 6 you get a chance to roll a 7 on a 4+). Or even adjust everyon'es BS downward by a point .