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.

 Vaktathi wrote:

Thus, HP's should be the primary kill method the majority of the time, and Chimeras, along with pretty much anything that isn't a very heavy tank, are hideously vulnerable to them.

Sorry mate, this is not true. It is your math that is off.

What is are the probabilities for killing a vehicle by stripping 3HP vs exploding it? Assume three heavy weapons (ac and las).

AC vs......p(3HP stripped)...........p(explodes!)..........LAS vs....p(3HP stripped)...........p(explodes!)
AV10.......0.240412153................0.22536063..........AV10.......0.047089335................0.36147548
AV11.......0.128044317................0.15551243..........AV11.......0.028527949................0.297668038
AV12.......0.049672929................0.080492843.......AV12.......0.015625.......................0.22974537
AV13.......0.009545262................0.............................AV13.......0.00735168................ 0.157578875
AV14.......0......................................0.............................AV14.......0.002679184................0.081039952

As you can see, apart from an AC firing at a 3HP av10 facing vehicle, the probability of exploding is greater in every single case.


Automatically Appended Next Post:
Just note that these are the numbers for three heavy weapons, and the results may change as more heavy weapons add their weight of fire, but I imagine (but not certain) that the conclusion will be the same. Also, another disclaimer: I'm not sure if this is even the right methodology to use to answer the question.

Another exercise for someone who wants to try it out is to ask the question, which outcome is more likely to happen first (ie 3HP stripped or explodes! result)? I haven't ran the numbers myself, but my initial thoughts are that the exploding result will still win out.

Where are you getting those numbers just out of curiosity?

They don't make any sense, 3 lascannons against AV10, that auto-inflict HP damage on a hit, are 8-fold less likely to kill it through HP damage than through an explodes result that only occurs on 1 in 3.6 hits (1/3 possibility on a 5/6 chance to pen)? Even taking into account the chance to kill it on the first or second hits, that's nowhere near accounts for that difference.

I'm failing to see how a result that requires occurs once ever 7.2 shots (bs3 lascannon vs AV10, 1/3 chance to explode on 5/6 chance to pen on 1/2 chance to hit=7.2 shot average) is 8x more likely to be responsible for destruction as one that occurs once every 6 shots (requires 3x auto-remove HP with S9 vs AV10 on each hit with a 50% chance to occur=6 shot average)

 Vaktathi wrote:
Where are you getting those numbers just out of curiosity?

Yeah, those numbers are completely wrong. Just to point out the obvious one, the chance of killing an AV 10 target by HP removal with three lascannon shots is 0.125, the same as the probability of hitting with all three shots.

 Vaktathi wrote:
Where are you getting those numbers just out of curiosity?

Find the probability of a single shot from ac/las causing HP loss or explodes result. Plug those results into a binomial model to find the probabilities for 6/3 shots, for ac/las respectively, to cause at least 3HP stripped, or at least one explodes! result.

 Vaktathi wrote:

They don't make any sense, 3 lascannons against AV10, that auto-inflict HP damage on a hit, are 8-fold less likely to kill it through HP damage than through an explodes result that only occurs on 1 in 3.6 hits (1/3 possibility on a 5/6 chance to pen)? Even taking into account the chance to kill it on the first or second hits, that's nowhere near accounts for that difference.

Most of the time when you hit with a las vs av10 you are penning. You've got three shots, so find the probability that at least one will hit and yield an explodes! result.

 Vaktathi wrote:

I'm failing to see how a result that requires occurs once ever 7.2 shots (bs3 lascannon vs AV10, 1/3 chance to explode on 5/6 chance to pen on 1/2 chance to hit=7.2 shot average) is 8x more likely to be responsible for destruction as one that occurs once every 6 shots (requires 3x auto-remove HP with S9 vs AV10 on each hit with a 50% chance to occur=6 shot average)

Try it out yourself mate. Its because when you use averages, it tends to overestimate how many shots you need to be reasonably certain of achieving a certain result.

 Peregrine wrote:

 Vaktathi wrote:
Where are you getting those numbers just out of curiosity?

Yeah, those numbers are completely wrong. Just to point out the obvious one, the chance of killing an AV 10 target by HP removal with three lascannon shots is 0.125, the same as the probability of hitting with all three shots.

If you want to know the the probability of three las stripping three HP, you need to use a binominal model. This is not the same as the probability of hitting with three shots, because you need to consider when a pen causes hp strip as well.

 Barksdale wrote:
What you are calculating here is the expected number of hp stripped. If you want to know the the probability of three las stripping three HP, you need to use a binominal model.

No, you're just doing the math wrong and over-thinking it. A lascannon that hits AV 10 automatically strips a hull point, so the probability of stripping three hull points with three shots is the probability that you hit with all three shots. And the probability of hitting with all three shots (at BS 3) is 0.5^3 = 0.125.


(Actually it would be slightly higher since you could get a penetrating hit and an immobilized result, then another immobilized result and strip the third HP even if one shot misses. But that's rare enough to ignore the possibility.)

Right, but what about the case when a pen causes a hp lost? Most of the time you are penning with a las vs av 10, not just glancing. That leaves much more oppurtunity for an explodes! result.


Automatically Appended Next Post:
And remember, you need to get three hp strips, but only one explodes result. That is where the large difference in probabilities comes in.

 Barksdale wrote:
Right, but what about the case when a pen causes a hp lost? Most of the time you are penning with a las vs av 10, not just glancing. That leaves much more oppurtunity for an explodes! result.

And remember, you need to get three hp strips, but only one explodes result. That is where the large difference in probabilities comes in.

You're missing the point. You said that the probability of three lascannon shots killing AV 10 by stripping 3 HP is 0.047. This is completely wrong because the probability must be at least 0.125, since there's a 0.125 chance of hitting with all three shots and any time you hit with all three shots the target will be wrecked.

 Vaktathi wrote:
Where are you getting those numbers just out of curiosity?

They don't make any sense, 3 lascannons against AV10, that auto-inflict HP damage on a hit, are 8-fold less likely to kill it through HP damage than through an explodes result that only occurs on 1 in 3.6 hits (1/3 possibility on a 5/6 chance to pen)? Even taking into account the chance to kill it on the first or second hits, that's nowhere near accounts for that difference.

I'm failing to see how a result that requires occurs once ever 7.2 shots (bs3 lascannon vs AV10, 1/3 chance to explode on 5/6 chance to pen on 1/2 chance to hit=7.2 shot average) is 8x more likely to be responsible for destruction as one that occurs once every 6 shots (requires 3x auto-remove HP with S9 vs AV10 on each hit with a 50% chance to occur=6 shot average)

You are right I did not apply the logic to both sides but it is not weighed evenly.



Okay these are the odds to get a single result from a single shot. So if we get three HP results then a chimera is dead(leaving a road block but oh well) while two results leave a living chimera from HP but perhaps not explodes. So we see that there is a 60% chance to explode AV 10 with a lascannon with 2 hit shots while a 0 percent chance to strip 3 hp with 2 shots. Vice versa 3 HP will kill a chimera where you have 0% chance to ever explode it.



These are the odds to explode the vehicle by the time you are statistically likely to get 3 HP of damage.

You can see that the odds to Explode with a lascannon before HP striping is excellent with the AC damage being nearly elusively HP stripping. So can I ask you a question? Do you use autocannons or lascannons? If you use autocannons then In know why you see the results you do.

This says nothing of the fact that every pen-non-explodes result stops the tank from being fully effective. So explodes is important.


I agree that HP stripping + non scoring vehicles made AC a deadly weapon capable of stopping a chimera rush flat...except if the AC targets AV14 or gets blown away before it can fire. So yeah mech vet charge of the chimera (its a herd cause they get eaten alive by predators if they get bogged down and they get sacrificed for their rider's god) is dead. Mechvet needs a really good distraction and screening to close, then it can do its thing. Though sadly vendetta vet spam is better in a lot of ways, the only way it is not as good is turn 1 strike if your enemy gets close and artillery screening.

Does anyone know how to format a table on this board?

There's also the matter that you're looking at the odds of stripping 3 hull points with just three shots -- requiring that all three of the shots glance or penetrate the target. When you choose such a small number (i.e. BARELY large enough for there to be even a chance of wrecking the vehicle), of course the number of vehicles killed by hull points will be small. What we're interested in is how those ratios act as the number of shots approaches key values -- like the amount of anti-tank fire you can expect to field in a single turn, or over the course of the game.

Surely you can see that.

Corollax wrote:
There's also the matter that you're looking at the odds of stripping 3 hull points with just three shots -- requiring that all three of the shots glance or penetrate the target. When you choose such a small number (i.e. BARELY large enough for there to be even a chance of wrecking the vehicle), of course the number of vehicles killed by hull points will be small. What we're interested in is how those ratios act as the number of shots approaches key values -- like the amount of anti-tank fire you can expect to field in a single turn, or over the course of the game.

This. For example, the supposed chance of killing AV 12 with the three LCs is about 25%, so your most likely scenario is having to fire six LCs (or three of them twice on consecutive turns) or ten LCs and a melta gun or whatever. So in reality you're dealing with larger numbers of shots, and the probability of death by HP loss goes up significantly.

 Peregrine wrote:

 Barksdale wrote:
Right, but what about the case when a pen causes a hp lost? Most of the time you are penning with a las vs av 10, not just glancing. That leaves much more oppurtunity for an explodes! result.

And remember, you need to get three hp strips, but only one explodes result. That is where the large difference in probabilities comes in.

You're missing the point. You said that the probability of three lascannon shots killing AV 10 by stripping 3 HP is 0.047. This is completely wrong because the probability must be at least 0.125, since there's a 0.125 chance of hitting with all three shots and any time you hit with all three shots the target will be wrecked.

I think that you are not clear about what these numbers are telling you. There is no reason why the probability of killing an av10 facing vehicle by stripping HP should be at least 0.125. That is only the probability of all three of those shots hitting. Alot of those hits are going to be pens, some of which strip a hull point, but at least one of which is very likely to be an explodes! result.

Consider if only two of those las hit. Then there is no way to strip 3 HP. The only other way to destroy the vehicle is by getting at least one explodes! result, or by getting two immobilized results.


Automatically Appended Next Post:

 Peregrine wrote:

Corollax wrote:
There's also the matter that you're looking at the odds of stripping 3 hull points with just three shots -- requiring that all three of the shots glance or penetrate the target. When you choose such a small number (i.e. BARELY large enough for there to be even a chance of wrecking the vehicle), of course the number of vehicles killed by hull points will be small. What we're interested in is how those ratios act as the number of shots approaches key values -- like the amount of anti-tank fire you can expect to field in a single turn, or over the course of the game.

This. For example, the supposed chance of killing AV 12 with the three LCs is about 25%, so your most likely scenario is having to fire six LCs (or three of them twice on consecutive turns) or ten LCs and a melta gun or whatever. So in reality you're dealing with larger numbers of shots, and the probability of death by HP loss goes up significantly.

This is a very real possibility, but I'm not entirely convinced that it is true. I'd have to run the numbers.

I would encourage you to do so. The wrecked/exploded ratio most certainly varies as a function of the number of shots fired. As the number of shots fired increases, that ratio will approach an asymptote. I haven't run the numbers, but I'd expect the function to approach that asymptote rather quickly. Right now, you're pulling off only the very edge of the binomial curve (where all three shots glance/penetrate). Naturally, your numbers are lower. =P

The specific value of that asymptote will depend upon the expected ratio of penetrating and glancing hits and the net increase on the damage table. Obviously, these factors will vary according to the difference between Armor Value and Weapon Strength and the presence of such special rules like rending and ordnance.

At the very least, you should be able to conclude that the cause of vehicle destruction is a function of not just weapon type but also weapon quantity. If you missed something that basic, you should probably wait before drawing any conclusions from the data you're calculating. Even if it's calculated correctly, it might be describing a scenario that doesn't generalize to the game as a whole (such as an army that only brought 3 lascannons for anti-tank and fired them precisely once). Get a stronger feel for the data first. Run some simulations. Python is excellent for this, but just about any programming language will work.

EDIT-1: If you think about it, the reason the wrecked/exploded ratio is not constant is very simple -- the discrepancy between the value for the given quantity of firepower and the asymptotic value reflects the average percent of damaged vehicles on the field. If you've fired 30 shots and destroyed 8 vehicles, neglecting the missing hull point on the 11th isn't very significant. If you've fired 3 shots and managed to rip 2/3 hull points off the one vehicle you shot at, it wouldn't be very accurate to say you haven't accomplished anything.

EDIT-2: As an unrelated note, you appear to be discounting results where a penetrating hit achieved an explode result on a vehicle with one hull point left. Whether or not the vehicle exploded, it was still killed by HP loss. You need to count such events under both categories. (That explode result didn't let you allocate any of your excess firepower elsewhere, after all!)

Whats bugging the heck out of me is that the more I see IG in the competitive scene, the more and more we're getting turned into a "henchmen" army. We're incredible allies for literally almost everybody, but when we're on our own we really struggle at taking the fight to the other guy. We lack a really tough scoring unit that can be used aggressively. Platoons can be incredibly hard to remove, but only when holding a spot. The moment they move up, they die horrible horrible deaths.

I've also noticed that more and more lately, I've just been going for "crush the opponent so he can't kill my scoring units" instead of just playing the objectives like a normal player would. Technically, the "kill everything so we can just hold the only objective at the end of the game" is still playing the objective, but it feels like a really weird way to go about it. At least we have the firepower to pull it off, but with the Tau on the way I'm worried our reign as the kings of dakka might be in trouble...

Perhaps that's the future of 6th ed IG? We have to literally pound the opponent into submission and then just waltz onto any objective that looks fairly safe. Sounds like a kind of boring way to play, but who knows, it's kind of how IG players played before anyways

BTW: Haven't read must past the OP yet, so this is mainly me responding to the first post, forgive me if all this has already been mentioned.

 MrMoustaffa wrote:
Whats bugging the heck out of me is that the more I see IG in the competitive scene, the more and more we're getting turned into a "henchmen" army. We're incredible allies for literally almost everybody, but when we're on our own we really struggle at taking the fight to the other guy. We lack a really tough scoring unit that can be used aggressively. Platoons can be incredibly hard to remove, but only when holding a spot. The moment they move up, they die horrible horrible deaths.

I've also noticed that more and more lately, I've just been going for "crush the opponent so he can't kill my scoring units" instead of just playing the objectives like a normal player would. Technically, the "kill everything so we can just hold the only objective at the end of the game" is still playing the objective, but it feels like a really weird way to go about it. At least we have the firepower to pull it off, but with the Tau on the way I'm worried our reign as the kings of dakka might be in trouble...

Perhaps that's the future of 6th ed IG? We have to literally pound the opponent into submission and then just waltz onto any objective that looks fairly safe. Sounds like a kind of boring way to play, but who knows, it's kind of how IG players played before anyways

BTW: Haven't read must past the OP yet, so this is mainly me responding to the first post, forgive me if all this has already been mentioned.

IG is a hammer, man. With such ineffective objective-holders, IG has become all about going for the tabling.

At least for me.

MrMoustaffa wrote: We have to literally pound the opponent into submission and then just waltz onto any objective that looks fairly safe. Sounds like a kind of boring way to play, but who knows, it's kind of how IG players played before anyways

TheCaptain wrote:

IG is a hammer, man. With such ineffective objective-holders, IG has become all about going for the tabling.

At least for me.

Am I the only one who has a problem with this? It's really not interesting to play in my opinion. Comes down to dice rolling far too much.

 Griddlelol wrote:
Am I the only one who has a problem with this? It's really not interesting to play in my opinion. Comes down to dice rolling far too much.

Sure, but the question here is the best way to win, not the most fun army type/strategy for each individual person.

And coming down to rolling dice is a good thing for winning. It takes player mistakes out of the picture and lets your superior list optimization win the game for you. You just have to make sure that average dice give you a win.

 Peregrine wrote:

 Griddlelol wrote:
Am I the only one who has a problem with this? It's really not interesting to play in my opinion. Comes down to dice rolling far too much.

Sure, but the question here is the best way to win, not the most fun army type/strategy for each individual person.

Indeed, but is trying to wipe your opponent off the table the safest way to win? If you don't roll average at key points that strategy is out the window, and with such low numbers (because even 60 dice will have huge variants from their statistical average) that chance isn't non-significant.

Feels like gambling rather than planning.

 Griddlelol wrote:
Indeed, but is trying to wipe your opponent off the table the safest way to win? If you don't roll average at key points that strategy is out the window, and with such low numbers (because even 60 dice will have huge variants from their statistical average) that chance isn't non-significant.

Sure, the limited margin for error when you're depending on tabling (or at least killing all the scoring units) is a concern, but that's an entirely different objection from "it's not interesting to play" which is about personal enjoyment rather than how effective the strategy is.

So if foot lists are dead, and mech lists are dead, what the hell are you guys fielding?

tybg wrote:
So if foot lists are dead, and mech lists are dead, what the hell are you guys fielding?

6+ flyers.

Corollax wrote:
I would encourage you to do so. The wrecked/exploded ratio most certainly varies as a function of the number of shots fired. As the number of shots fired increases, that ratio will approach an asymptote. I haven't run the numbers, but I'd expect the function to approach that asymptote rather quickly. Right now, you're pulling off only the very edge of the binomial curve (where all three shots glance/penetrate). Naturally, your numbers are lower. =P

First off, the model uses the standard to hit and to glance/penetrate probabilities. It does not assume all three shots glance/penetrate.

Right, so vehicle destruction is obviously a function of weight of fire. For those interested (if anyone) the ac will switch to HP stripping being more likely (compared with explodes! result) at 4 ac for av 11 (~25% vs ~20%), 5 ac for av 12 (~18% vs ~13%). For the las, a HP becomes more likely at 12 las for av11 (~76% vs ~75%), 14 las for av 12 (~72% vs ~ 70%), 15 las for av 13 (~58% vs ~57%and av14 (~35% vs ~34%). Personally, I take the explodes! route any day of the week.

tybg wrote:
So if foot lists are dead, and mech lists are dead, what the hell are you guys fielding?

Dead doesn't mean unusable, it really means either "this is not as good as it used to be" or "it's not competitive any more."

Mech is still strong, and so is foot (when backed up with allies and FW). They're just not what they were in 5th. I've not tried a full on air-cav yet, money is a huge limiting factor for that...

 Vaktathi wrote:
In every conceivable sense mech has become drastically less useable and survivable under 6E. Transports have less utility and cannot hold or even contest objectives with embarked scoring units, and tanks in general are easier to kill than they've ever been before typically.

However, changes in the environment mean that transports can still be very useful. For instance, 6e's increased prevalence of Barrage weapons, Torrent weapons, and Thunderfire Cannons means that transports are still very important for protecting your guys. Unfortunately IG gets the short end of the stick in that side armor 10 on Chimeras makes the Chimeras themselves very vulnerable to many of the weapons used in this context.

 Griddlelol wrote:

MrMoustaffa wrote: We have to literally pound the opponent into submission and then just waltz onto any objective that looks fairly safe. Sounds like a kind of boring way to play, but who knows, it's kind of how IG players played before anyways

TheCaptain wrote:

IG is a hammer, man. With such ineffective objective-holders, IG has become all about going for the tabling.

At least for me.

Am I the only one who has a problem with this? It's really not interesting to play in my opinion. Comes down to dice rolling far too much.

The decrease in cover from 4+ to 5+ has made very difficult to get across the table for anyone. In 6th edition more of my games are determined within the first few turns than they ever were in 5th. For most armies the most important dice roll is going to be who gets to place more objectives. Then the other guy has to come to you and get horribly gunned down.

I think that's why you see most tournaments vary the book missions to some degree so that both people are forced to go forward. Even then, most games come down to not who brought more guns to blast the other army to bits to prevent them from holding objectives.

I agree, most of the time, the game's result turn around who got the first turn and took care of the opponent's heavy hitting untis, leaving the weak troops being shot up/assaulted/retreating

Inquisitor Jex wrote:
I agree, most of the time, the game's result turn around who got the first turn and took care of the opponent's heavy hitting untis, leaving the weak troops being shot up/assaulted/retreating

Yeah...no. Getting the second turn gives you the chance to capture/contest objectives. First turn damage can be quite easily mitigated by deployment. Unless of course the ridiculous Seize the Initiative screws up an aggressive deployment.

 Griddlelol wrote:

MrMoustaffa wrote: We have to literally pound the opponent into submission and then just waltz onto any objective that looks fairly safe. Sounds like a kind of boring way to play, but who knows, it's kind of how IG players played before anyways

TheCaptain wrote:

IG is a hammer, man. With such ineffective objective-holders, IG has become all about going for the tabling.

At least for me.

Am I the only one who has a problem with this? It's really not interesting to play in my opinion. Comes down to dice rolling far too much.

Yeah it's bugging me too. If this strategy ends up being as boring as it sounds I may shelve my IG for a while, or just use them for silly games.

I'm not exactly gracing Nova, Feast of Blades, and Adepticon with my presence every year so competitiveness to me only matters as long as the list is interesting to play.

That said, shoot the gak out of everything does feel like a very Imperial Guardsy thing to do...

We rain fire down in the emperor's name, as it were.

Has anyone tried allying in dark angels with a libby/land raider to run alongside the chimeras/leman russ' to give them 4++ saves? I have been meaning to do this but I never bought a leman russ for my dark angels. (or you could run those flimsy coffins the space marines get but seems like it would die, fast)

 ansacs wrote:
We rain fire down in the emperor's name, as it were.

Has anyone tried allying in dark angels with a libby/land raider to run alongside the chimeras/leman russ' to give them 4++ saves? I have been meaning to do this but I never bought a leman russ for my dark angels. (or you could run those flimsy coffins the space marines get but seems like it would die, fast)

I'd love to try that. I every time I head to a store considering a different build with allies I end up coming home with another guard unit -_-


Automatically Appended Next Post:

 MrMoustaffa wrote:

Yeah it's bugging me too. If this strategy ends up being as boring as it sounds I may shelve my IG for a while, or just use them for silly games.

Now this sounds like whining, so there's no point in hiding it:

But I don't wanna!

After seeing the Tau rumors, I'm going to play a mech list this weekend because I think it will be the last time in an RTT until our new book arrives, whenever that is.

I'm going to try to blow everything off the table strategy, using vets in Vendettas to take the objectives. Might not work, but it will be fun to try. I'll try and go second most games, and not take an astropath to get the planes in a bit later if possible.

CCS in chimera, lascannon
4 vet squads, two with flamers in Vendettas, 2 with Melta in chimera
Exterminator with HB/sponsons
2 LRBT in a squadron
Manticore

4 Storm Eagles
2 battlecannons
7 lascannons
2 autocannons
6 melta guns
3 heavy flamers
6 heavy bolters
3 multi-lasers
6 flamers
assorted lasguns and laspistols

Weight of Metal for the Emperor!