Warhammer 40K Tabletop VIII: "You're Gonna have To Face It, You're Addicted To Math"

[Razaele]

Comp score?

Where I'm from, those don't exist.

[Cheesegear]

Comp score?

Composition Score. Basically, is your army fair? If not, you start a tournament with less points than some whose list is fair. Not everyone uses them. I don't know why. It keeps things fair.

It's incredibly useful in competitive areas such as mine. A common phrase after a friendly game in my area is "How would you rate that list?"

[Razaele]

Composition Score. Basically, is your army fair? If not, you start a tournament with less points than some whose list is fair. Not everyone uses them. I don't know why. It keeps things fair.

I guess the idea never came to our country. I've brought it up before, but all I got was a bunch of raised eyebrows and strange looks.

[Cheesegear]

I guess the idea never came to our country. I've brought it up before, but all I got was a bunch of raised eyebrows and strange looks.

Maybe the thread title should change to

"Why do I have to re-write a Comp: 0 list?"

[Ninja Chocobo]

No it isn't. Because it's your opponent that decides who takes the wounds.

Not for Dangerous Terrain they don't. One in six isn't great, sure, but it's better than you'd get with any other weapon.

Daemonhunters...No. Just...No. A regular Grand Master is way better than Sten.

Explain how. Also, if nothing else he lets you take a second GK HQ. And thus a second GKT retinue.

Orks; Grotsnik, Wazdakka and Zagstruk. And G*. Damn. Zogwort.

Grotsnik isn't very good. Feel no Pain on a unit is nice, as is giving Meganobs invul saves. What is less nice is being led around by a Rhino or Landspeeder or similarly crappy vehicle.
Since it's not Rage, it can't be circumvented by getting in a transport.

[Incomp]

A regular grey knight grand master is waay better than Stern, because his profile is better (significantly, mind you) and the grand master is only slightly more expensive. That reroll ability Stern has...meh. It'll help, like having a master-crafted weapon helps; it's a benefit for sure, but I doubt that it would change the course of the game.

EDIT: As for comp scores, I'd never heard of them before reading this thread. They do sound like a pretty good idea, I have to say. However, my area isn't hyper-competitive.

[Cheesegear]

Not for Dangerous Terrain they don't. One in six isn't great, sure, but it's better than you'd get with any other weapon.

It's not better than a D-Cannon or Tempest Launcher. Or Fire Prisms.

Explain how. Also, if nothing else he lets you take a second GK HQ. And thus a second GKT retinue.

How is a Grand Master better than a Brother-Captain? Really? The numbers on the page say so.

Stern also doesn't have access to an Icon of the Just. Which is kind of important on a special character.

Anyway, you should already have two GKT squads. You really don't need a third. Especially if it's costing you an extra 120 points over normal to do so.

Grotsnik isn't very good. Feel no Pain on a unit is nice, as is giving Meganobs invul saves. What is less nice is being led around by a Rhino or Landspeeder or similarly crappy vehicle.

'Leading Around' a unit is harder to do than you think. Especially when the Ork opponent figures out what you're doing, and immediately sets the rest of army onto the task of busting that unit.

[Ceridan]

Maybe the thread title should change to

"Why do I have to re-write a Comp: 0 list?"

So how does the Comp score system work? I assume a Comp: 0 in the above is horrible broken, so what is the scale?

[Cheesegear]

So how does the Comp score system work? I assume a Comp: 0 in the above is horrible broken, so what is the scale?

There is no scale. That's what confuses a lot of people. It's a handicap, and it is directly proportional to everyone else. I can take the same list to any given tournament and be given anywhere between a 1 and 4. It all depends on how the TO and friends feel about my list in comparison to everyone else at the tournament.

There is no 'system' to define your Comp Score before your TO gives it to you.

Comp: 0 is a Forced Rewrite. You're not allowed in the tournament with your current list. There is no point in you showing up because you know that you will win.

[Ceridan]

There is no scale. That's what confuses a lot of people. It's a handicap, and it is directly proportional to everyone else. I can take the same list to any given tournament and be given anywhere between a 1 and 4. It all depends on how the TO and friends feel about my list in comparison to everyone else at the tournament.

Comp: 0 is a Forced Rewrite. You're not allowed in the tournament with your current list. There is no point in you showing up because you know that you will win.

Ah. So if I had a 4 list and played aginst your 1 list what would the handicap do? Is it something mechanical or is it that someone could just say; he beat me because I had a list that was crap?

Oh, and can you give an example of a Comp: 0 list, and where would that DE Wytch army you mentioned in one of your battle reports rate?

[Cheesegear]

Ah. So if I had a 4 list and played aginst your 1 list what would the handicap do? Is it something mechanical or is it that someone could just say; he beat me because I had a list that was crap?

No. That's not what it's for. It only counts at the end of the day.

I've said once or twice before, going into a tournament with a 'killer' list and coming out with all Major Victories is pretty easy. However, if some guy with a less-powerful list also gets all Major Victories, he ends up winning the day because he's actually a better general (or lucky) because he doesn't have a broken list.

Ultimately, Composition is an extra 1-10 points added on to your final score. And is used fairly often to determine final winners.

Oh, and can you give an example of a Comp: 0 list, and where would that DE Wytch army you mentioned in one of your battle reports rate?

Comp 0? 60 Scouts with Sniper Rifles earned me a Forced Rewrite...
The Wych Cult army has already been mentioned in one of my tournament-reports before with a Comp Score of 3.
That should give you an idea of just how awesome the Comp 1 and 2 lists were...

[Ceridan]

Ah, that make sense. So, can you give an example of a Comp: 0 list, and where would that DE Wytch army you mentioned in one of your battle reports rate?

[Ninja Chocobo]

It's not better than a D-Cannon or Tempest Launcher. Or Fire Prisms.

Better for picking out a character.

How is a Grand Master better than a Brother-Captain? Really? The numbers on the page say so.

A Grand Master is way better than a Brother-Captain, yes.
Stern however is only 1W and 1A behind the GM, for a non-trivial amount of points.

Stern also doesn't have access to an Icon of the Just. Which is kind of important on a special character.

Less important given his Retinue. Also given his re-rolls.

'Leading Around' a unit is harder to do than you think. Especially when the Ork opponent figures out what you're doing, and immediately sets the rest of army onto the task of busting that unit.

If they're on foot, they're only moving 6+d6" a turn, and they can only Assault after doing so one of those turns. If they're in a vehicle, it can be Stunned or Immobilised pretty easily. Leading them around with a single unit is hard, yes. But what happens if you drop an empty Drop Pod 11" behind them? They move, Assault, kill it, and are an extra 11-d6" away from your lines. It's pretty easy to manoeuvre in such a way that you're controlling the placement of the closest unit and the second-closest unit.
Also, they have to Assault a unit. Which means they have to disembark from a vehicle. And if they have to move to re-embark, they can't.

[Cheesegear]

But what happens if you drop an empty Drop Pod 11" behind them?

That's something I've never considered doing. I wasn't aware that you could even have empty Pods. However, now that I think about it, you can.

[Ninja Chocobo]

That's something I've never considered doing. I wasn't aware that you could even have empty Pods. However, now that I think about it, you can.

..Wasn't it you who was talking about having five Pods, two of which were empty, for the purposes of Drop Pod Assault?
..Maybe I've got you confused with our local power-gamer.

[Cheesegear]

..Wasn't it you who was talking about having five Pods, two of which were empty, for the purposes of Drop Pod Assault?

No, it wasn't. But thanks for the idea. But, I still need unit/s in my army capable of taking a Drop Pod as a Dedicated Transport in the first place. You can't take them on their own. And, considering that I like Bikes and Scouts...

Still, the Grotsnik-factor is kind of irrelavent. Since the question was 'Who uses Special Characters'. And my answer was that people in my area do. Not that I thought he was any good and somesuch.

Also, I believe the tactic is to let Grotsnik run around by himself, not to actually put him in a unit where he can wreck it. The only reason he gets taken is to get Invulnerable saves for anyone.

[Ninja Chocobo]

No, it wasn't. But thanks for the idea.

Also, if you drop it empty, you can get into it. The wording runs "Once the passengers have disembarked..."

Still, the Grotsnik-factor is kind of irrelavent. Since the question was 'Who uses Special Characters'. And my answer was that people in my area do. Not that I thought he was any good and somesuch.

Misunderstood you there, based on the fact you seemed to have opinions on other characters.

Also, I believe the tactic is to let Grotsnik run around by himself, not to actually put him in a unit where he can wreck it. The only reason he gets taken is to get Invulnerable saves for anyone.

Yeah, that's a much better idea.
Still, for his kill point and slot there's better things available.

On Stern's re-roll: The fact is, it's so very versatile. Reserves? Scatter? Terrain? Saves? Leadership? You can re-roll any of these, for any squad. The opponent gets a re-roll, too, but chances are the roll they use it on won't be half as critical.

[Cheesegear]

Also, if you drop it empty, you can get into it. The wording runs "Once the passengers have disembarked..."

Have you tried doing that? And did you get punched in the head?

Misunderstood you there, based on the fact you seemed to have opinions on other characters.

The opinions were pretty much why those characters got taken. I do talk to people about their army lists and find out why they take what they take.

[Ninja Chocobo]

Have you tried doing that? And did you get punched in the head?

I played Orks.
I mention it merely as an "Oh ho ho ho isn't RAW silly" point.

[Razaele]

Guys, could I please have an opinion on my 500 point army list?

[lord_khaine]

Eldar have the Wonder Twins. They're better than the other HQ options you can take of the same calibre. So there's no reason not to take them.

Who are you talking about here?

[Kzickas]

Who are you talking about here?

I would assume one is elrad since a rerollable 3++ save, 3 wounds and the ability to kill any powerfists with mindwar is very nice

[lord_khaine]

I would assume one is elrad since a rerollable 3++ save, 3 wounds and the ability to kill any powerfists with mindwar is very nice

Ahh yeah, i was a bit unsure about if the wording on Mind War would allow it to pick out models from a unit.

[LCP]

You don't need statistics. You can surely, without fail, say that 'on average' a Khorne Berzerker will kick the crap out of a Kroot. It's simple numbers on a page, guys. 7 > 4. Everyone knows already.

That’s nowhere near as useful, though. Knowing “my unit stats are bigger than yours” does not quantify the kind of success you can expect.

'Average' dice rolls...Don't belong in 40K. Any decent statistician will tell you that normal distribution doesn't exist with samples under 30. Have you ever rolled 30 dice at once? Unless you're Orks, Guard or 'Nids, probably not. However, that's one of the reasons why good Guard and 'Nids list can consistently win. Because the dice will actually roll what the theory says.

Furthering this, a good statistician will tell you that samples sizes should be somewhere in the two to three hundred range...

Firstly – combat rolls aren’t normal distributions! Each individual set of rolls (to hit, to wound, to save) is a binomial distribution, and the final ‘success’ (i.e. unsaved wounds) distribution is what comes out of the end when you pass those three distributions through each other (which isn’t even binomial – check out the distribution I generated, it has an asymmetry due to the number of unsaved wounds having a lower-end cap at 0). Take your number of events (rolls) high enough, and sure, the binomial distributions will approximate normal distributions, but why would you want to do that?

The variance on a binomial distribution is given by np(1-p). For ten rolls, not thirty, assuming your success probability is 0.5 (i.e. needing a roll of 4+ for whatever you’re doing) that gives a standard deviation of ~1.6, which means you can expect your total number of successes to lie within ~1.6 of the expectation value with ~68% confidence. 68% confidence that your number of successes will lie between 4.4 and 6.6 - that’s far from useless. For rolls where you need a 3+ or a 2+, the spread is tighter still. Propagating these errors through the various stages of to hit, to wound and to save is more work than I’m prepared to do right now, and would probably increase them a bit, but it still gives the lie to your claim that dice quantities less than thirty are completely futile subjects for statistical analysis.

Also, I’d like to know where you read/heard these numbers of 30 and 200-300, since they are almost meaningless without context. When ‘any decent statistician’ was talking about whether a normal distribution ‘existed’ – were they talking about whether it could be called continuous to within a good approximation? Because that’s irrelevant to this discussion, it’s a quality of the normal distribution that doesn’t matter here. What kind of precision were they requiring in fulfilling the criteria of the distribution? A science or engineering context is probably going to be demanding orders of magnitude more accuracy than you require when you’re talking about casual gaming with dice. I’m currently on a summer placement at my uni’s Physics department, and if you applied the criteria for ‘good statistics’ from the experiment I’m running – looking at the statistical distribution of flaws on an optical fibre – you’d be demanding that math-hammer grant its users oracular superpowers. And that's not what we're after - math-hammer only has to be at the very shallow end of the accuracy pool to be useful in this very casual context.

So, yeah, [citation required]. My statistics are a bit rusty, so I might well have made a mistake, but if I have I’d be interested in hearing it.

[dsmiles]

There is no scale. That's what confuses a lot of people. It's a handicap, and it is directly proportional to everyone else. I can take the same list to any given tournament and be given anywhere between a 1 and 4. It all depends on how the TO and friends feel about my list in comparison to everyone else at the tournament.

There is no 'system' to define your Comp Score before your TO gives it to you.

Comp: 0 is a Forced Rewrite. You're not allowed in the tournament with your current list. There is no point in you showing up because you know that you will win.

This, I don't get. If it's entirely subjective how can it be a rule at wargame tournaments? (I can understand paintjob getting/losing you points - in a separate category. My paitjob has nothing to do with the playability of my army.) Seriously, a rules-legal list is a rules-legal list. If you don't know how to make an effective army list (like me) don't participate in tournaments (like me) until you do know how. It's like M:tG. A deck is a deck. If your deck sucks, you lose more than you win. It's a fact of life: Games are not fair. They favor the skilled, and there is no 'level playing field.' [/rant]

[† Dran †]

Maths hammer is a great way of using theory to prove that you should win. Play the game and I bet that you will still lose.

I can write a list that should be nigh unstoppable and that should win me every single match I play, but we use dice in this system and the dice gods are crazy creatures.

I don't need Citation to prove that im right, I know I am cause I've played so many games. I've seen a unit of hamminators die in one round of combat against grots. GROTS. Tell me how that happened? I've also seen a single warlock solo a unit of 10 assault marines and win.

Now im not going to argue that there's not an average to dice, but just because there is doesn't mean that in this game im going to get the average.

[Cheesegear]

Games are not fair. They favor the skilled, and there is no 'level playing field.'

40K does not favour the skilled. There is no skill in dice rolls. There is no skill in the ability to have more money than everyone else and the ability to buy six Land Raiders.

The internet, being what it is, now allows you to net-list like all get out. You post a list, and 'the internet' will change it for you. The internet will give you the best possible outcomes for your list. There is no skill in following others' opinions.

Ultimately, without composition scores, you end up having six out of the top eight lists being Space Wolves or Blood Angels, no-one should even bother turning up if they aren't a power-gaming git. And that's fine. That's what 'Ard Boyz tournaments exist for.

However, there's a line on how much of a complete douchebag you're allowed to be. And Composition Scores raise that line. Composition Scores want you to bring a different army to what people have already seen eight thousand times before. Composition Scores want the tournament to not be 'Space Wolves vs. Blood Angels, and other lists are just in the way'.

Composition Scores encourage diversity by fairness. Maybe you haven't read Page 2 of the rulebook?

[dsmiles]

However, there's a line on how much of a complete douchebag you're allowed to be. And Composition Scores raise that line.

*wiggles fingers*

This is not the line you're looking for.

[lord_khaine]

Originally Posted by Cheesegear
You don't need statistics. You can surely, without fail, say that 'on average' a Khorne Berzerker will kick the crap out of a Kroot. It's simple numbers on a page, guys. 7 > 4. Everyone knows already.

Statistisk become usefull though, when you are 4 berserkers and 9 Kroots, or when some of the stats are higher, while some are lower on the models you compare.

[Cheesegear]

Statistisk become useful though, when you are 4 berserkers and 9 Kroots, or when some of the stats are higher, while some are lower on the models you compare.

Yeah. Sure. But you can't actually use Statistics in the game. At least, you can't do in-depth calculations at the board, whether or not you should Assault if you've got 4 Berzerkers or 5. At the time, there's probably nothing else for the Berzerkers to do anyway, so they may as well Assault.