GK Metrics

[jakehunter52]

Ok, so after reading this article on the 40k metrics system, my mind was pretty blown (possibly because I love working with numbers) and I was wondering if any of you guys were tracking this too. I guess it was done by a guy named Nikephoros and he even has a Crowe list. However, I have been trying to run the numbers myself and I can't seem to find out how he came up with the values he did (mine seem to be on the lower end unit for unit) and perhaps some other mathematicians would like to join in on the fun.

[Captain Idaho]

I know there are people about who like this stuff, but I think it is spoiling 40k. You can't judge an army or even unit based purely upon it's damage output in comparison to something else. What if it is faster and/or resilient than it's damage out put? What if it takes up a force organisation slot you don't need, thus allowing you to have your cake and eat it in a list? What if it multiplies your force even with a poor damage out put, allowing you to bulk out a list with something you need?

 

There are so many factors to 40k and people got lazy. Instead of practicing with units and armies to learn what works for them, they want a winning formula.

 

Sorry if this is your thing and I don't want to insult you, but people need to be told. Learning the hard way in 40k is actually the fun way, because it means playing games! When you get experienced enough you can see the merits and rough potential of a unit without having to crack out pointless statistics. You then try it out and adjust it for anything you didn't anticipate. Dice don't perform averagely or in fractions and they don't care what your statistics or previous dice rolls are. Use your experience instead.

[jakehunter52]

I appreciate your honest criticism of this system and some of it is well warranted. A person's objective should not be to find out what the latest and greatest list is by breaking down the numbers and ridiculing those that don't meet up to the standard. The games should not start and end with a sheet of paper with a bunch of arbitrary numbers on it that supposedly to represent the culmination of your army's value. I do think that lists need to be written up and then played to see even if you like the way it feels, if you are comfortable with your options or you wish to take some other routes. I hope that is not what this is taken as.

 

However, I do think saying things "Dice don't perform averagely or in fractions and they don't care what your statistics or previous dice rolls are" are short sighted and wrong. I have studied enough statical modeling and economic trends to be convinced that a law of averages is present and over the long term, will provide consistent results. And I think too many people confuse "statistics" with un-average dice rolling. Why did you lose at that critical point? Because of a couple of bad rolls? Well, I would bet on you to lose. It's like investing your 401k in small businesses and expect it stay at a reasonable average ALL THE TIME. You don't diversify yourself enough and you allow yourself to fall trap to a couple of bad rolls. That's why redundancy is the word in constructing any solid performing list....and why people hate it so much. Because it limits options as well, but that's another discussion.

 

And this numerical system is no way to find an "ultimate" list. You will fail if you try that. Upon further reading, I see that it is simply a fair comparison of the firepower between two army lists. It still gives you raw data to see what your competition is capable of and using that to see a) how you match in firepower and :D what strengths and weaknesses you both have. It will it pretty handy to know just how well those Fire Dragons can destroy your LR and priority that over taking out the larger group of Dire Avengers.

 

And for those who are interested, here are a Psyrifle Dread stats:

 

1.98, .56, 8.89, 0

[thade]

"Dice perform average". Sure they do. Most of the time. Statistically all the time, but I think we all feel at times that's not the case. :P

 

Regardless, it doesn't matter. Because your expected damage output a turn will vary wildly depending on more factors than you can reasonably account for with math. Terrain, positioning, relative movement, suboptimal moves on the part of yourself or your opponent, your opponent making strange list choices that put out less damage but can soak more damage than you can put out...you have a number to judge a unit by, but realize that the number will be no more or less reliable than purely your own experience. I regularly draw and beat meta'd lists with my silly, fluffy, obviously under-tuned lists. I am not alone in this regard. At the end of that game when the guy with the suboptimal list wins, can you blame the maths?

 

No, I think not.

 

Maths are handy, but they are not the end-all-be-all of list building or target priority. Use them as a tool but not as your only tool. That is my advice. <3

[Captain Idaho]

Much respect jakehunter52, you seem to know your stuff. I know what you mean about statistics, but I don't like them as in the hands of someone who doesn't have the appropriate understanding of wargaming will end up making the wrong decisions in list building or just be very 1 dimensional (like the plague of Netlists we have at the moment).

 

I think we are on the same page with regards to how we play on the table though; we both utilise redundancy in our tactical choices (as well as in a list).

[Grimtooth]

Mixed martial arts has a few fight metric systems to determine how well match-ups will go and who is the best or going to be the best. Even some of the highest rated are still getting knocked out by those with lower ratings.

 

Numbers work in a vacuum but tend to break down in the real world.

[thade]

Numbers work in a vacuum but tend to break down in the real world.

QFT

[BLACK BLΠFLY]

Wow I dont think anyone toots their own horn more than Kirby. I found the article to be a classic case of TL;DR. As stated above math hammer in a vacuum is pretty much worthless.

 

G ^_^

[Grimtooth]

I just get the feeling that it is like,

 

"Here is a bunch of math to support COMMON SENSE."

[Babel_Triumphant]

Honestly, I feel like I don't need an article to tell me that missiles are good at blowing up light vehicles and that longfangs are better than devs in a purely unit:unit comparison. Environments, lists and playstyle are so varied in the game that I feel like mathhammer is pretty limited in most cases. I will concede that it's not so bad for running comparisons of similar units fulfilling similar functions, though.

[Something Wycked]

This kind of measurement only establishes a baseline- under the perfect circumstances, what can I expect the unit/army list to kill? That's all it measures. It won't tell you how the unit/list will actually perform- not only can it not predict how your dice will roll, but everything else that goes into a battle (range, cover, casualties, etc) cannot be accounted for with this metric.

 

But, again, its a good baseline for the worth of units.

[Toasterfree]

Numbers work in a vacuum but tend to break down in the real world.

 

But, again, its a good baseline for the worth of units.

qft

[antique_nova]

"Dice perform average". Sure they do. Most of the time. Statistically all the time, but I think we all feel at times that's not the case. ;)

 

Maths are handy, but they are not the end-all-be-all of list building or target priority. Use them as a tool but not as your only tool. That is my advice. <3

 

Pfft tell that to my dice. Average dice rolls my arse. I have seen myself roll more 1s and 2s than i have ever seen numbers of humans. ;)

 

Also, that it the best piece of advice, an advice that i have learnt some time ago, that i have ever seen for any player who wants to improve - especially on a extremely competitive level.

 

thanks

antique_nova

[templargdt]

I haven't run that metrics system yet, but I think there is some value added to a stats analysis of the game. Yes there are tons of factors that can alter a decision set, but the fact is there are choices that are more efficent than others and math can help you realise what those choices are.

 

That said, I'm only running probabilites in my head during a game when I'm at a tournament. When I go to my buddies man-cave and play on his game table, I'm not there to show off my skill playing with toy soldiers.

[Crynn]

I find the system rather bland. My 1500 blood angel list performs horribly by that system but it's speed means i hit flanks, side armour get better cover saves, move out of range of units, shield keeps theings alive, armour saturation helps keep marines alive as do priests. My list performs fantastic in RL but by that measure is horrible. I think it is a very basic analysis that experienced players can pretty much ignore like ramseys said, logic. For a gunline army maybe, for anything that require a decent tactician to play, it just doesn't equate nearly enough variables to be a valid metric number system in my mind. I am a researcher by profession so I love numbers and even I think this system is rather lacking.

 

Regards,

 

Crynn

[Toasterfree]

there is a lot of value to this system. it adds something more to the math hammer. but its not going to give you what is the end all and be all. this is only going to come from experience playing the game, and from personal flavor. sure by the system long fangs are awesome. if they are all missiles. but take two out and make them lascannons (giving them some DLRPG) and whoah, suddenly missile fangs just arent so epic anymore.

 

 

better yet, why not check them numbers, test the unit out at you FLGS, and then see what works. not every math hammer, metric is going to work everytime.

[jakehunter52]

I find the system rather bland. My 1500 blood angel list performs horribly by that system but it's speed means i hit flanks, side armour get better cover saves, move out of range of units, shield keeps theings alive, armour saturation helps keep marines alive as do priests. My list performs fantastic in RL but by that measure is horrible. I think it is a very basic analysis that experienced players can pretty much ignore like ramseys said, logic.

 

there is a lot of value to this system. it adds something more to the math hammer. but its not going to give you what is the end all and be all. this is only going to come from experience playing the game, and from personal flavor. sure by the system long fangs are awesome. if they are all missiles. but take two out and make them lascannons (giving them some DLRPG) and whoah, suddenly missile fangs just arent so epic anymore.

 

I think you have a valid point Crynn. Metrics doesn't effectively measure an army's durability and tactical mobility because it assumes that is positively correlated with points cost and offensive ability is not as much. And yet the opposite is proven to be true: Most top-tier lists are mechanized because transports are immune to a majority of small arms fire, even though a footslogger list would perform better in metrics. People are looking at this as a quick way to set benchmarks and see who can reach them. And that is not what this is.

 

And at the same time, I don't think it is worthless but quite valuable. It proves quick access to useful statistics that can be used to make choices. Toasterfree's example is perfect, it demonstrates what we all know in real life: That sometimes units look really good on paper but miss a vital variable. AND METRICS REFLECTS THAT! Which is why it requires experienced handling to be useful and cannot be used as a "short-cut".

[thade]

Pardon me for meta-commenting on this thread.

 

I have been really impressed of late with the directions these kinds of threads have been taking. Over my time here (what, just a couple of years, give or take?) I've seen a shift. Where it used to be that the majority of posters in a thread like this would blindly support math-hammer - whether they had a real grasp on the statistics or not - more and more people seem to be posting that "Math-hammer is a good tool, but it's not the only tool in my kit."

 

I'm proud of all of you. *sniff*. So...so proud... *sniff*

 

As you were. <3

[Toasterfree]

i never use math hammer to decide if i am going to take a unit or not. if i am having some issues with a unit performing where i want it, then i mathhammer it to get it to where i think it should be. if that doesnt work then i go back to the drawing board. what i look for is melee results, but never thought about my shooting results also. will this be what i use to decide what units to take? no. but it will be a deciding tool to help with strategy on the table. that is where tools like this really shine. is AFTER a unit is placed down....

 

"well if shoot these guys about X could die and then if i assault then Y will die and leave me out in the open...."