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.

 Galas wrote:
So powerfist and powerklas no longer "auto" destroy 100-200 point vehicles and for that they are useless...

Yeah... Look, I also loved when my tactical sargeant one-shot that mechanical monster that charged his unit with a bunch of lucky 6's, but that wasn't good.

The present incarnation of power fist/claws as anti-elite infantry/characters meele-weapon is good enough.

The obvious difference being that the 215pt Ork Boyz Mob had to run across the entire table and then get in charge distance before bringing that 25pt Powerklaw to bear against said vehicle. So yes, it should auto-kill that 100-200pt vehicle after getting raked with fire for at least 2 full turns.

But even if you don't agree that PKs and PFs should kill a vehicle in CC then lets at least do a comparison. 25pts = dead vehicle in 7th, 13pts = minor inconvenience to vehicle (2ish damage), then depending on the type of vehicle and faction it can simply waltz out of combat and shoot the ever loving crap out of those orkz who are now stranded and in rapid fire range. Since we went from a PK doing a complete kill or at the least a vehicle being unable to move or shoot the next turn to being only mildly inconvenienced I think we should look at points costs and adjustments. A Big Choppa should be 3pts and a PK should be 5pts. Since they don't actually accomplish there mission anymore.


Automatically Appended Next Post:

 Imateria wrote:

I really hope you weren't blowing up vehicles in 7th on a 5+ with an AP2 weapon as an explodes result was a 7 on the table, so you needed 6's. Also, if a combat unit charged you (say my Incubi), I'd challenge your Nob and probably kill him bepfre he gets to attack since I5 beats I1/2 any day.

1/6th chance + 1/6th chance = 1/3rd chance. And against Open Topped vehicles it was closer to a 2/3rd chance.

As for your challenge nonsense, yep, that ruined a lot of my games and tactics because the ork faction REQUIRED PK nobz in order to kill vehicles and a single well equipped Sergeant or equivalent could ruin that function. Of course, if you rolled poorly my PK nob would squish (instant Kill) any character in the game who didn't have Eternal warrior or T5. Something else that doesn't happen anymore.

Actually, you don't just add the 1/6th chances together. You square the odds of it NOT doing anything (5/6)^2=25/36, then subtract that from 1.

So you have just under a 1/3 chance, or 11/36 chance.

 JNAProductions wrote:
Actually, you don't just add the 1/6th chances together. You square the odds of it NOT doing anything (5/6)^2=25/36, then subtract that from 1.

So you have just under a 1/3 chance, or 11/36 chance.

Which we then round up to.....1/3rd chance. You know....cause its a D6 game not a D36.

SemperMortis wrote:

 JNAProductions wrote:
Actually, you don't just add the 1/6th chances together. You square the odds of it NOT doing anything (5/6)^2=25/36, then subtract that from 1.

So you have just under a 1/3 chance, or 11/36 chance.

Which we then round up to.....1/3rd chance. You know....cause its a D6 game not a D36.

Why would you do that? Just because the game is only as granular as to go by increments of 1/6 doesn't mean our math should be less precise.

 JNAProductions wrote:

SemperMortis wrote:

 JNAProductions wrote:
Actually, you don't just add the 1/6th chances together. You square the odds of it NOT doing anything (5/6)^2=25/36, then subtract that from 1.

So you have just under a 1/3 chance, or 11/36 chance.

Which we then round up to.....1/3rd chance. You know....cause its a D6 game not a D36.

Why would you do that? Just because the game is only as granular as to go by increments of 1/6 doesn't mean our math should be less precise.

The problem with being so precise is that it generally becomes irrelevent, you can't get 0.1432455 of a dice roll which makes going to such a granular level a waste if time since in practice it makes no difference.

 Imateria wrote:

 JNAProductions wrote:

SemperMortis wrote:

 JNAProductions wrote:
Actually, you don't just add the 1/6th chances together. You square the odds of it NOT doing anything (5/6)^2=25/36, then subtract that from 1.

So you have just under a 1/3 chance, or 11/36 chance.

Which we then round up to.....1/3rd chance. You know....cause its a D6 game not a D36.

Why would you do that? Just because the game is only as granular as to go by increments of 1/6 doesn't mean our math should be less precise.

The problem with being so precise is that it generally becomes irrelevent, you can't get 0.1432455 of a dice roll which makes going to such a granular level a waste if time since in practice it makes no difference.

I mean, when you get right down to it, this is entirely impractical anyway. The better calculation is finding out how likely you are to get an "Explodes" or "Immobilization".

Ah yes we only ever roll that dice once so no averaging out due to attacking say more than once in your lifetime. Jep jep makes sense.

SemperMortis wrote:

 JNAProductions wrote:
Actually, you don't just add the 1/6th chances together. You square the odds of it NOT doing anything (5/6)^2=25/36, then subtract that from 1.

So you have just under a 1/3 chance, or 11/36 chance.

Which we then round up to.....1/3rd chance. You know....cause its a D6 game not a D36.

I'm not entirely sure why you're facepalming or belittling precision, especially because though you are correct that 40k isn't a d36 game it is a d216 game (d6 to hit * d6 to wound * d6 to save) most of the time. Rerolls, two-dice-pick-highest, and FNPs extend it to a d279936 (6^7, though it could be 6^8 if there's a mechanism to reroll FNP, I don't know of any off the top of my head) under extreme circumstances, though this has been limited by removing FNP-stacking. Trying to round all probabilities to multiples of 1/6 because "this is a d6 game" is silly; consider the chance of an Ork shoota getting a wound through on a Land Raider, for instance; 2/6 to hit * 1/6 to wound * 1/6 to get through the save gets you 1/108, and trying to round that to either 0 or 1/6 are both misleading.

And that's just the odds of one result happening. Trying to look for the odds of (A) results on (B) dice it gets even more fun.

The abstraction mechanism you're using ("get the results of one attack, then multiply by the number of attacks") gets you something called the expected value of an attack, which is the average number of successes if you roll that attack an infinite number of times, and it's a pretty useful shortcut when considering 8e. If you want to go back to 7e where asking about the odds of getting at least one success on N attacks (to explode vehicles, for instance) was a thing, then the math works exactly how JNA just told you it does; by "rounding" you are attempting to claim that getting 2 explodes results on the same vehicle is somehow more valuable than getting one result.

 AnomanderRake wrote:

SemperMortis wrote:

 JNAProductions wrote:
Actually, you don't just add the 1/6th chances together. You square the odds of it NOT doing anything (5/6)^2=25/36, then subtract that from 1.

So you have just under a 1/3 chance, or 11/36 chance.

Which we then round up to.....1/3rd chance. You know....cause its a D6 game not a D36.

I'm not entirely sure why you're facepalming or belittling precision, especially because though you are correct that 40k isn't a d36 game it is a d216 game (d6 to hit * d6 to wound * d6 to save) most of the time. Rerolls, two-dice-pick-highest, and FNPs extend it to a d279936 (6^7, though it could be 6^8 if there's a mechanism to reroll FNP, I don't know of any off the top of my head) under extreme circumstances, though this has been limited by removing FNP-stacking. Trying to round all probabilities to multiples of 1/6 because "this is a d6 game" is silly; consider the chance of an Ork shoota getting a wound through on a Land Raider, for instance; 2/6 to hit * 1/6 to wound * 1/6 to get through the save gets you 1/108, and trying to round that to either 0 or 1/6 are both misleading.

And that's just the odds of one result happening. Trying to look for the odds of (A) results on (B) dice it gets even more fun.

The abstraction mechanism you're using ("get the results of one attack, then multiply by the number of attacks") gets you something called the expected value of an attack, which is the average number of successes if you roll that attack an infinite number of times, and it's a pretty useful shortcut when considering 8e. If you want to go back to 7e where asking about the odds of getting at least one success on N attacks (to explode vehicles, for instance) was a thing, then the math works exactly how JNA just told you it does; by "rounding" you are attempting to claim that getting 2 explodes results on the same vehicle is somehow more valuable than getting one result.

Right, so from now on instead of saying I have a 1/3 chance to hit something I'll go ahead and do this "33.33333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333" because we are trying to be as precise as possible right?

Except 1/3 is more precise. (Not to mention, it'd be .33 repeating or 33.33 repeating PERCENT, not 33.33 repeating flat.)

And everything in moderation. And clarity. 1/3 is easy to understand, and clear. 11/36 is a little less clear, but it's easy to compare to 12/36 (the same as 1/3).

.33333333... Is clear, I guess, but annoying to read.

 Imateria wrote:

 JNAProductions wrote:

SemperMortis wrote:

 JNAProductions wrote:
Actually, you don't just add the 1/6th chances together. You square the odds of it NOT doing anything (5/6)^2=25/36, then subtract that from 1.

So you have just under a 1/3 chance, or 11/36 chance.

Which we then round up to.....1/3rd chance. You know....cause its a D6 game not a D36.

Why would you do that? Just because the game is only as granular as to go by increments of 1/6 doesn't mean our math should be less precise.

The problem with being so precise is that it generally becomes irrelevent, you can't get 0.1432455 of a dice roll which makes going to such a granular level a waste if time since in practice it makes no difference.

No, but none of this math is for telling you "You got 0.1432455 of a die roll!", it's for telling you "If you do (x) you get an average of 0.1432455 results, whereas if you do (y) you get an average of 0.1532455 results, which means (y) is slightly better".

0.1432455 is not there because 0.1432455 is a number that might come up in the game, it's a mechanism for determining which thing is better. That said it is true that computing something out to seven decimal places doesn't really help, but that isn't because "we're using d6s", it's because we don't roll enough dice over the course of a game for that level of difference to establish itself, and because the way weapon stats work there are easy approximate shortcuts. Imagine I'm considering a lascannon versus a missile launcher fired at a Rhino; they're identical except that the lascannon has better AP and slightly higher average damage, so I can conclude that the lascannon is a better choice without pulling out my calculator.

Doing the math in detail is useful and can be helpful in building an army list, but it isn't important enough to pull out a calculator during a game because your target priority decisions tend to be either a relatively easy decision to make on the fly by quick statline comparisons or something you worked out earlier when building your army list or when deciding which units to buy/how to build them. On top of that the calculator may give you bogus data simply because of how difficult it is to try and value range/movement against opponents with a variety of desired range/movement values themselves.

Using a calculator out to many decimal places is a useful tool; it doesn't give you all the answers but belittling it because it doesn't give you all the answers isn't helpful either because it does give you bits of the answers.

Also, if you notice in other threads, I almost always stay in fractions till the end, then convert out to two decimal places. (Occasionally three, but usually two.) That way it's highly precise, but easy to understand.

SemperMortis wrote:
...Right, so from now on instead of saying I have a 1/3 chance to hit something I'll go ahead and do this "33.33333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333" because we are trying to be as precise as possible right?

Depends. What's the margin you're working with? Is there another value you're trying to compare it to that's 33.3333...4% or 33.3333...2%?

As a general rule of thumb when dealing with 40k I would say that rounding things to the nearest 1%-2% is almost always enough to make an accurate decision; rounding things to the nearest 17% like you did when trying to mock JNA earlier is going to produce inaccurate or misleading results if you want to try and make any actual decisions based on your math.

 AnomanderRake wrote:

SemperMortis wrote:
...Right, so from now on instead of saying I have a 1/3 chance to hit something I'll go ahead and do this "33.33333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333" because we are trying to be as precise as possible right?

Depends. What's the margin you're working with? Is there another value you're trying to compare it to that's 33.3333...4% or 33.3333...2%?

As a general rule of thumb when dealing with 40k I would say that rounding things to the nearest 1%-2% is almost always enough to make an accurate decision; rounding things to the nearest 17% like you did when trying to mock JNA earlier is going to produce inaccurate or misleading results if you want to try and make any actual decisions based on your math.

So rounding 1-2% is absolutely fine. SO when I said 1/3rd and made my comment you made your rebutall.....so what is 11/36? 30.5% or roughly 2.8% difference. So you are now saying 2% is ok but 2.8 is just too imprecise to round to and therefore we have to have a discussion on why rounding up that extra .8% is just way to far and ruins the point someone is trying to make. Or conversely you just like arguing over nonsense.

SemperMortis wrote:

 AnomanderRake wrote:

SemperMortis wrote:
...Right, so from now on instead of saying I have a 1/3 chance to hit something I'll go ahead and do this "33.33333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333" because we are trying to be as precise as possible right?

Depends. What's the margin you're working with? Is there another value you're trying to compare it to that's 33.3333...4% or 33.3333...2%?

As a general rule of thumb when dealing with 40k I would say that rounding things to the nearest 1%-2% is almost always enough to make an accurate decision; rounding things to the nearest 17% like you did when trying to mock JNA earlier is going to produce inaccurate or misleading results if you want to try and make any actual decisions based on your math.

So rounding 1-2% is absolutely fine. SO when I said 1/3rd and made my comment you made your rebutall.....so what is 11/36? 30.5% or roughly 2.8% difference. So you are now saying 2% is ok but 2.8 is just too imprecise to round to and therefore we have to have a discussion on why rounding up that extra .8% is just way to far and ruins the point someone is trying to make. Or conversely you just like arguing over nonsense.

The issue I had was that your math was completely and utterly wrong. The methods, not the final result.

If all you had said was "The odds of getting an Explodes on 2 hits was about 1/3", I wouldn't have any sass for that, because that's accurate.

But you said you arrived at that by simply adding 1/6+1/6=1/3, which is entirely NOT how you do it. By that logic, 6 hits would give you a 100% chance, when it's actually only about 2/3rds. (66.51%, to be more exact.)

I'm not a fan of bad math.

 JNAProductions wrote:

SemperMortis wrote:

 AnomanderRake wrote:

SemperMortis wrote:
...Right, so from now on instead of saying I have a 1/3 chance to hit something I'll go ahead and do this "33.33333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333" because we are trying to be as precise as possible right?

Depends. What's the margin you're working with? Is there another value you're trying to compare it to that's 33.3333...4% or 33.3333...2%?

As a general rule of thumb when dealing with 40k I would say that rounding things to the nearest 1%-2% is almost always enough to make an accurate decision; rounding things to the nearest 17% like you did when trying to mock JNA earlier is going to produce inaccurate or misleading results if you want to try and make any actual decisions based on your math.

So rounding 1-2% is absolutely fine. SO when I said 1/3rd and made my comment you made your rebutall.....so what is 11/36? 30.5% or roughly 2.8% difference. So you are now saying 2% is ok but 2.8 is just too imprecise to round to and therefore we have to have a discussion on why rounding up that extra .8% is just way to far and ruins the point someone is trying to make. Or conversely you just like arguing over nonsense.

The issue I had was that your math was completely and utterly wrong. The methods, not the final result.

If all you had said was "The odds of getting an Explodes on 2 hits was about 1/3", I wouldn't have any sass for that, because that's accurate.

But you said you arrived at that by simply adding 1/6+1/6=1/3, which is entirely NOT how you do it. By that logic, 6 hits would give you a 100% chance, when it's actually only about 2/3rds. (66.51%, to be more exact.)

I'm not a fan of bad math.

So in essence you are angry that I used common sense to get the correct answer instead of using a statistics formula which gives me almost the exact same answer? Sorry but I don't give a damn about .8% nor do I care about using statistics beyond the bare minimum for the point of a D6 based wargame. Now, if I was calculating something that mattered and wasn't based on a simple D6 engine then by all means lets break out the statistics formulas. Until then though, calling out someones math for being .8% off in a D6 game is just nitpicking for the sake of nitpicking.

Again, the methodology is completely flawed. Common sense can't be used to replace mathematics.

For reference, your "common sense" odds versus the actual odds on...

1 Hit
16.67%
16.67%

2 Hits
33.33%
30.56%

3 Hits
50%
42.13%

4 Hits
66.67%
51.77%

That's a difference of just under 15 percentage points at four hits. Or nearly an entire face on a d6.

If you only want to put in the minimum effort and are happy with wrong answers, that's fine-just don't expect everyone to be satisfied with that.

I don't generally look at grabbing a Hammer unless the model can put out 3 attacks or more.

It is just a general points ratio. Adding a power sword to a sgt is 2 points per attack. Adding a Fist/Hammer is 6/8 per attack. But on a Vet Sgt it is 4/5 a much better ratio. On characters even with the 5 pt increase on hammers they usually have 2+WS so it isn't as big a deal.

Power fist type weapons are okay on characters with 4 or more attacks but they are unreliable on models such as a sergeant since they hit on 4+. Even if you have a squad five terminators it’s still not that good.

 JNAProductions wrote:
Again, the methodology is completely flawed. Common sense can't be used to replace mathematics.

For reference, your "common sense" odds versus the actual odds on...

1 Hit
16.67%
16.67%

2 Hits
33.33%
30.56%

3 Hits
50%
42.13%

4 Hits
66.67%
51.77%

That's a difference of just under 15 percentage points at four hits. Or nearly an entire face on a d6.

If you only want to put in the minimum effort and are happy with wrong answers, that's fine-just don't expect everyone to be satisfied with that.

Here is the best part, you failed the basic assumptions check, because you assume I was using common sense in a specific way, for me I know statistics well enough to put two and two together. And I am well aware of the statistical likelihoods of a specific result. You do know the old adage, when you assume you make .....


Automatically Appended Next Post:

 Primark G wrote:
Power fist type weapons are okay on characters with 4 or more attacks but they are unreliable on models such as a sergeant since they hit on 4+. Even if you have a squad five terminators it’s still not that good.

Couldn't agree more. And even on characters they are iffy since the cost/benefit ratio isn't that great to begin with. In the case of orkz it is significantly better to equip your warboss with the relic Big Choppa then a PK.

 Imateria wrote:

SemperMortis wrote:
YAY Mathhammer time.

7th edition a PK cost 25pts and was so worth taking that i NEVER equipped anything else and took as many as I could. Why? Because in 7th edition a PK Nob was S9 (+1 strength on charge) so he was AUTO glancing most vehicles and had a 5/6th chance to penetrate most of them, on the pens he was +1 because of AP2. So a regular Nob was usually hitting on 3s and penetrating on 2s so usually 2-3 pens and at +1 on the damage chart he could usually do some serious damage, easily 3 HPs but he had a 1/3rd chance to explode the vehicle.

Now for 8th edition. That same Nob is now swinging with a minimum of 3 attacks, usually 4+ he now hits on a 4+ against vehicles so 2 hits, those 2 hits are S10 Vs most vehicles being T6-7 so they wound on a 3+now instead of an auto wound. They now also get an armor save Vs that PK attack. So generally 1 will go through now and do 2 dmg to a vehicle which usually has about 10+ wounds. So yes the price dropped by about 50% but the effectiveness of the weapon has dropped by more then 75%. I have 1 true PK in my tournament list and that is because I can't field the Painboy without a PK, if I could you can bet I would save those points and give him NOTHING. The only other PK weapon I have is on Ghaz and that is because he hits on a 2+ always and does 3dmg each, plus he has 6 attacks.


So is this the edition of Power Fists/Klaws? No, for Orkz in my opinion, they aren't worth taking.

I really hope you weren't blowing up vehicles in 7th on a 5+ with an AP2 weapon as an explodes result was a 7 on the table, so you needed 6's. Also, if a combat unit charged you (say my Incubi), I'd challenge your Nob and probably kill him bepfre he gets to attack since I5 beats I1/2 any day.

Spamming pks in MSU boyz squad was a very effective option for anti tank in 7th. Challenge was an issue but if you had 5-6 MSU of boyz plus the rest of the army it wasn't an issue. If you had bigger squads like 19+mek in a battle wagon or 30 plus a painboy on foot it wasn't an issue either, you could refuse the challange with the other character that wasn't good in melee.

SemperMortis wrote:

Here is the best part, you failed the basic assumptions check, because you assume I was using common sense in a specific way, for me I know statistics well enough to put two and two together. And I am well aware of the statistical likelihoods of a specific result. You do know the old adage, when you assume you make .....

Well then if you know why you keep pretending you don't? You are talking pure BS here so no wonder you get called out on it

You people derailed and entire conversation about whether you think a PF is worth it on squad leaders for a whole page to argue about math and try to show off the elective statistics and logic classes you took in school.

We could all be reading a book right now.

cmspano wrote:
You people derailed and entire conversation about whether you think a PF is worth it on squad leaders for a whole page to argue about math and try to show off the elective statistics and logic classes you took in school.

We could all be reading a book right now.

And this post in turn contributes... How?

Here-a PF on a 2 attack Sergeant does the following wounds to the following models. In each case, there is 1 hit.

MEQ-25/36
Primaris-125/108
TEQ-25/27
Custodian Guard-2/3
Rhino-10/9
Leman Russ-5/6
Land Raider-2/3

All I know is, PF on my berserker champion kills deamon princes and other big vehicles, regularly.

Berserkers have a lot of attacks so they are a snow flake compared to other champions (CSM) and sergeants (SM).