Wolf Claw vs. Frost Blade

[Grey Mage]

Of course math hammer works. Its silly to assume that it doesn't.

 

 

Try playing 10 bloodclaws vs 10 grey hunters.

 

Math hammer says the Grey hunters will hit more often with their shooting and get hit less often in CC by most other troops.

 

 

Mathhammer is why people use powerfists and TDA.

 

 

A large part of this game is based on math and being able to estimate how things are going to go down based on probabilities.

I have no problem with the idea that mathhammer can predict trends, but its not 100% accurate, and those trends are not the final word in what happens- the dice roll is.

 

I find alot of it is simply the lack of a standard deviation every being figured for most of these numbers.

[Inquisitor Kravin]

My post was two part:

 

First: condense the information, a long post containing info arrayed like that is confusing (and so meaningless). We just need to see the result.

 

Secondly: the pillars, thats then answer I was waiting for, your stacking the odds, which you can't do (my point entirely) each dice roll/coin flip is a seperate event, entirely unrelated to the previous event, there is no reason that every dice roll won't be a six, or every coin flip a head. The coin/dice does not remember the previous result, and so each outcome is random. Give it a try, I guarantee you won't find the splits you think you will (or maybe you will! its all chance!).

 

First point: Anyone with any training and/or experience in a mathematical or scientific discipline appreciates the importance of demonstrating one's working. Only by showing your working can your reader confirm that your calculations are correct or bring to your attention any errors that you have made, enabling you to ensure the validity of your conclusions. I thought that his post was well presented and clear.

 

Second point: I'm not sure what to say to this simply because you seem to fundamentally misunderstand the concept. The long existence of casinos as successful businesses is just one of a huge range of things around us that you are denying are possible. To me your denial of the existence of probability is on a par with trying to deny the existence of gravity or the Sun. The calculation of independent probability does not make any assumption that the die or coin "remembers" its previous results. It is counting on exactly the opposite, the fact that each roll of the dice is totally independent. If somebody tells you that because a coin came up heads last time it will come up tails this time, they are talking nonsense. However, I would happily bet money against you that if we flip two coins and on a result two heads you win and on the result a head and a tail I win and that we will play the game ten times. Yes, on any given time we don't know what will happen but overall I will most likely come off better.

 

A single gretchin killing a whole unit of terminators in close combat doesn't prove probability wrong. It just consoles the Space Marine player that he didn't make a mistake and in future he shouldn't be put off repeating the move.

 

By your logic, the best army in the game is Imperial Guard because they get to roll the most dice and therefore will do most damage, since by your logic a stormbolter and a lasgun are equally effective.

[nurglespuss]

I'm sorry you misunderstood my post, I thought i made it resonably clear, and you seem to have picked up some concepts.

 

I never denied the existence of probability, it doesnt imply that anywhere in what I've written....

 

Also, a casino is a terrible example, the house always wins, why? because the games are fixed. Not physically altered, as in weighted dice etc. but fixed so that the odds are most definately in favour of the casino (vastly) and your winnings are minimised as much as possible.

 

Finally, back to the first point, your incorrect. I deal with scientific papers/journals etc. daily, and have written a few, you never present information like that,at any time (apart from early stage workout) information like that is presented in the appendicies only, a summary table presenting the results is all that is required, if the workings are in doubt, these are requested to be shown.

 

-------------

edit:

This conversation looks like its more about who can sound clever than it is about math hammer.....

But thats just the opinion of this Bloodclaw...

also spazmonkey is entirely correct.

[DapperDave]

Also, a casino is a terrible example, the house always wins, why? because the games are fixed. Not physically altered, as in weighted dice etc. but fixed so that the odds are most definately in favour of the casino (vastly) and your winnings are minimised as much as possible.

 

Finally, back to the first point, your incorrect. I deal with scientific papers/journals etc. daily, and have written a few, you never present information like that,at any time (apart from early stage workout) information like that is presented in the appendicies only, a summary table presenting the results is all that is required, if the workings are in doubt, these are requested to be shown.

 

Ok, this is kinda funny now... you're saying it's possible for casino's to "fix" odds so that certain outcomes are more likely... but that combinations of wargear have no effect on probable outcomes. :unsure:

 

Also, what science discourages the showing of your work that supports your conclusion?

 

Bill Nye would be ashamed.

[nurglespuss]

LOL most science, go pull up any science paper you like (run a quick PDF search on google scholar) find me one cluttered with workings out, and data as you saw before... can't? didn't think so, its what appendicies are for.

 

What you will find are tabulised data, graphical representations and statistics (the result, the statistic name/method name/modifications and the significance, nothing else, no one needs to see it. If they do, they go to the relevant appendix.

 

I hope that helps.

 

Ok, this is kinda funny now... you're saying it's possible for casino's to "fix" odds so that certain outcomes are more likely... but that combinations of wargear have no effect on probable outcomes

 

Ahhh now misquoting me there, probable out comes and possible outcomes. it does not affect the 'actual' probability (that remains the same) but it does present a broader range of 'attractive possible outcomes' which is not the same as directly increasing the probability.

 

A good example is the bag of beans: if i have a bag of beans with 50/50 red and blue, you have an equal chance of grabbing red/blue.

If I have a bag of beans with 2 red/8 blue beans, this increases the probability of slecting a blue bean.

If I have 3 bags of of the same combination, I have also increased the possibility, but not the probability.

However, if i combined the bags (6 red 24 blue) I have increased the chance of selecting red, but not as much as I have increased the sheer number of blues, therefore, even though the proportions are the same, the actual numerical superiority of blues reduces the probability of selecting a red.

[imperious]

Well, I suppose I see your point, but I guess I just don't see why you've made it. Does the math used in this thread stand up to the rigors of a scientific treatment? Naw, probably not. Is it, however, a useful reference point in discussing whether Squad A is better than Squad Br? I think it is, yes. Obviously you disagree, but frankly, I don't give a damn and I'm fairly sure a there are a good bit of people on this and other forums who don't either, if the popularity of the method is anything to go by.

 

Now if you've got a better way of showing the probabilities or even a completely different approach for comparing different squads or wargear combination's, then by all means, I would love to see it (non-sarcastic). But if your just here to wave your arms about and shout about how meaningless it is, then I just don't see how that could possibly be viewed as constructive. You're more then welcome to be right, but let us play our numbers games in peace.

 

Now can we please get back to talking about our toys?

[Freman Bloodglaive]

To hit 3+,4+,5+

Toughness 3,4,5 versus strength 4WC or 5FB

 

3+T3

S4WC, 2/3 x (1-(1/3 x 1/3)) = 16/27

S4WC, (1-(1/3 x 1/3)) x 2/3 = 16/27

S5FB, 2/3 x 5/6 = 10/18 or 15/27

 

Victory to Wolf Claws, doesn't matter whether you re-roll hits or wounds.

 

3+T4

S4WC, 2/3 x (1-(1/2 x 1/2)) = 6/12 or 9/18

S4WC, (1-(1/3 x 1/3)) x 1/2 = 8/18

S5FB, 2/3 x 2/3 = 4/9 or 8/18

 

Wolf Claws, re-rolling to wound.

 

3+T5

S4WC, 2/3 x (1-(2/3 x 2/3)) = 10/27

S4WC, (1-(1/3 x 1/3)) x 1/3 = 8/27

S5FB, 2/3 x 1/2 = 2/6 or 9/27

 

Wolf Claws, re-rolling to wound.

 

4+T3

S4WC, 1/2 x (1-(1/3 x 1/3)) = 4/9 or 16/36

S4WC, (1-(1/2 x 1/2)) x 2/3 = 6/12 or 18/36

S5FB, 1/2 x 5/6 = 5/12 or 15/36

 

Wolf Claws, re-roll to hit.

 

4+T4

S4WC, 1/2 x (1-(1/2 x 1/2)) = 3/8 = 9/24

S4WC, (1-(1/2 x 1/2)) x 1/2 = 3/8 = 9/24

S5FB, 1/2 x 2/3 = 2/6 or 8/24

 

Wolf Claws, any flavour.

 

4+T5

S4WC, 1/2 x (1-(2/3 x 2/3)) = 5/18 or 10/36

S4WC, (1-(1/2 x 1/2)) x 1/3 = 3/12 or 9/36

S5FB, 1/2 x 1/2 = 1/4 or 9/36

 

Wolf Claws, re-roll to wound.

 

5+T3

S4WC, 1/3 x (1-(1/3 x 1/3)) = 8/24 or 18/54

S4WC, (1-(2/3 x 2/3)) x 2/3 = 10/27 or 20/54

S5FB, 1/3 x 5/6 = 5/18 or 15/54

 

Wolf Claws, re-roll to hit.

 

5+T4

S4WC, 1/3 x (1-(1/2 x 1/2)) = 3/12 or 9/36

S4WC, (1-(2/3 x 2/3)) x 1/2 = 5/18 or 10/36

S5FB, 1/3 x 2/3 = 2/9 or 8/36

 

Wolf Claws, re-roll to hit.

 

5+,T5

S4WC, 1/3 x (1-(2/3 x 2/3)) = 5/27 or 10/54

S4WC, (1-(2/3 x 2/3)) x 1/3 = 5/27 or 10/54

S5FB, 1/3 x 1/2 = 1/6 or 9/54

 

Wolf Claws, any combination.

 

Trends. Below toughness 8 a Wolf Claw is always better than a Frostblade.

 

Against toughness 3, and equal or better WS (if you find a toughness 4 model that requires 5+ to hit do the same) take the re-roll to hit option. Against anything else re-roll to wound.

 

As already said, in power armour take the frostblade because you get another attack from your free bolt pistol. In terminator armour take the wolf claw. I pair mine with storm bolters.

[Wolf Lord Cuneglas]

Nurglespuss....uh...sorry?

 

Thank you for reconfirming with me Freman Bloodglaive. I don't understand how this thread jacking occured....

 

Bottom line, either play how you want, or play with the "odds" in your favor.

[bravelybravesirrobin]

This post will be On-Topic the post below will be on the slightly weird mathhammer topic.

 

On-Topic

 

The pair of wolf claws beats everything hands down but it is also the most expensive option.

 

The Frostblade however is not that much worse than a pair of wolf claws as Bloddlgaive's math demonstrates it is mere fractions that separate them. A Frostblade with a bolt pistol is much cheaper than the pair of wolf claws and so probably gives more bang for your buck or the best possible killing ability for the cheapest cost.

 

but if you can't get that cheap extra attack from the bp due to being in termie armour or weilding a storm shield then the wolf claw is not only superior to the frostblade but also cheaper again.

 

 

 

 

 

 

Off Topic

 

Nurglepuss we started this conversation in another thread and I really dson't understand where you are coming from with this. I suspect that is probably my problem as I don't have a background in mathematics but with all your lols and open statements without explaining the process you are really failing to communicate your argument here.

 

Try stripping out any jargon and explaining in plain terms why you think Mathhammer to be useless. You've provided many examples which I think I grasp but I fail to see how those examples relate to 40k as it is played.

 

For example your bean example. 3 bags with 4 red beans and 8 blue beans. I follow that the probability of picking a blue bean is higher (8/12 or 2/3) and that this means the possibility of picking a blue bean is higher. I also follow that adding more bags doesn't change the probability (it remains 8/12 or 2/3). I also follow you that if I add all the bags together it still doesn't increase the probability because the odds are evenly numbered (i.e. with 12 red beans and 24 blue beans it's still 2/3)

 

But 40K doesn't just add the same events together, it removes beans and changes the odds. It is based on chains of events.

 

So if I roll 10 dice and know that a 4,5, or a 6 is a hit then I have odds of 1/2 and I can expect roughly half my dice to hit. I know less could hit or more but most of the time half will hit. I will then remove 5 dice and roll them but this time maybe I need a 3+ to wound so I would expect 2/3 of my dice to wound. Meaning that out of every 10 dice I can expect to cause about 3 or 4 wounds.

 

That just seems like perfect common sense to me. Obviously there is some massive fallacy I haven't seen there but in practise there hasn't been any. I have used Mathhammer to predict if marines will beat orks in cc, if orks in cover can outshoot marines, etc for years and the conclusions I and others reach are usually born out on the tabletop.

 

I also put it to you that if Mathhammer doesn't work then how come grots don't usually beat marines in close combat? How come missle launchers aren't the ultimate tank busting weapons? Because when you say mathhammer doesn't work that's what you seem to be implying to me.

[Freman Bloodglaive]

Hi Robin, nice to see you're still around.

[nurglespuss]

But if your just here to wave your arms about and shout about how meaningless it is, then I just don't see how that could possibly be viewed as constructive. You're more then welcome to be right, but let us play our numbers games in peace.

 

Fair enough, I shall leave it there.

 

Just to finish on a positive though:

 

@ Freeman Bloodglaive: thats a much beter way of presenting these data, its still long, but its much more readable and you can properly compare results.

 

Thanks.

[bravelybravesirrobin]

Hi Robin, nice to see you're still around.

 

Nice to know people remember me.

 

Shiny new codex has brought me back to the fold after a few years out of the hobby and a few playing Orks. But I was always just waiting for an excuse to bust out my Wolves again..

[DanTheManGum]

My post was two part:

 

First: condense the information, a long post containing info arrayed like that is confusing (and so meaningless). We just need to see the result.

 

Secondly: the pillars, thats then answer I was waiting for, your stacking the odds, which you can't do (my point entirely) each dice roll/coin flip is a seperate event, entirely unrelated to the previous event, there is no reason that every dice roll won't be a six, or every coin flip a head. The coin/dice does not remember the previous result, and so each outcome is random. Give it a try, I guarantee you won't find the splits you think you will (or maybe you will! its all chance!).

 

EDIT: I see the discussion was already resolved.

[RADU LYKAN]

just something to add to peoples thoughts

in power armour a frost blade plus bolt pistol is better than a wolf claw plus bolt pistol

in power armour 2 wolf claws is better than frost blade and bolt pistol

in power armour frost blade and plasma pistol is better than 2 wolf claws (as long as you get to shoot on the way in)

 

in terminator armour the wolf claw wins :lol:

[Wildfire]

in power armour frost blade and plasma pistol is better than 2 wolf claws (as long as you get to shoot on the way in)

 

Even setting aside the cost of plasma pistols (way too much), I disagree with this statement. You're assuming you'll always get the charge, which is simply not true.

[Freman Bloodglaive]

in power armour frost blade and plasma pistol is better than 2 wolf claws (as long as you get to shoot on the way in)

 

I respectfully disagree. Although the strength 7 shot is good, there's a non-zero chance of killing yourself and that's not a good thing.

 

1/6 x 1/3 = 1/18

[Inquisitor Kravin]

Sorry for the threadjacking, I'll come back on topic below.

 

Nurglepuss:

-I found the acronyms in the post that sparked this debate perfectly self-explanatory and I think that to dismiss it as meaningless because you apparently couldn't deduce that TDA meant Terminator Armour, FB meant Frost Blade etc. was unfair to the poster of that information.

 

-Yes, a scientific paper would be better structured and presented but then it's generally about communicating concepts and the calculation is just procedural, you would expect to see at least some description of the methodology and line of argument rather than just an isolated statement of conclusions. Without supporting information enabling the reader to analyse the workings themselves the results are of dubious value.I think that to expect so much from a forum post is unrealistic. Without wishing to demean the poster their work was comparable to a school maths lesson exercise and I think that their working was presented accordingly.

 

-You have to consider audience. Science papers are aimed at a specialist audience. This forum is not, well it is, but the specialism is 40k Power Armoured armies, not maths.

 

-In the casino example I was perhaps too vague, I was specifically thinking of roulette and dice games where each round is an independent event but the house wins because they can reliable calculate the long term averages and offer odds accordingly. Precisely what we are thinking of doing when we assess different weapon loadouts with mathhammer.

 

-It seems I did misunderstand your original intent. Since you declared a post "meaningless" and said "mathhammer doesn't work" I mistook you for one of the people I've encountered before who states, "My terminators lost to guardsmen - therefore that proves that probability is wrong". Apologies for that.

 

However, if i combined the bags (6 red 24 blue) I have increased the chance of selecting red, but not as much as I have increased the sheer number of blues, therefore, even though the proportions are the same, the actual numerical superiority of blues reduces the probability of selecting a red.

 

[bold added by me for emphasis] I have to disagree, the probability remains the same provided the ratio remains the same. If scaling up or dowwn affected probability then the end result would depend on your starting point and the whole principle of sampling would collapse.

 

However, (the bit where I agree with Nurglepuss), whilst I think that you were far from clear in your first post with your blanket statement that "mathhammer doesn't work" I do agree that it needs to be kept in perspective and that Probability and Expectation alone are very crude measures. Each individual encounter is an independent event and any possible outcome may happen but I don't build a new army for every game. I plan an army, build it and use it repeatedly so the long-term trends matter.

 

This means that the usefulness of such calculations varies:

what does mathhammer say about Logan Grimnar fighting a Keeper of Secrets? Not a lot since it's a rare event and both sides just take their chances.

what does mathhammer say about the effectiveness of massed bolter fire against MEQ? A lot since it happens many, many times and overall your performance will be close to average.

 

ON TOPIC

 

I'm favouring the Wolf Claw. I already have masses and masses of power armour of various sorts. What I don't have much of in any existing army is Terminators so I'm going down the "Wolfwing" route. The extra attack for the additional weapon is a moot point with Terminators. Therefore, the only advantage of the Frostblade is that it can hurt Toughness 8 and penetrate AV10. Whilst this is not insignificant I intend to have at least one Powefist, Chainfist or Thunderhammer in each squad so the main purpose of the rest of the squad will be cutting through enemy troops and for this I favour the greater reliability of the Wolf Claw.

[nurglespuss]

OFF TOPIC: I think thats a fair post Kravi, I can see that, you can see, where I was coming from. We don't have to agree, but I appreciate your candour.

 

ON TOPIC: Personally I think it better to maximise the effect your going for (if a model has a specific purpose) e.g. I have a battle leader with jump pack and wolf claws, partly because I think he looks cool :P and partly because he's leading a unit of skyclaws who are rarely (if ever) going to get to shoot.

 

Other wise, i do tend to go for plasma. Again (disapointingly) my characters excellent balistic skill rarely gets to be used, but when it does, I know I stand a decent chance of hitting something, plus again the plasma pistol looks cool. Its a shame master crafting isn't available to space wolves.