The 'Proven X' Quality

By Nimsim, in Dark Heresy House Rules

For all this talk of being math wizards neither of you have put forth a better way to analyze the efficacy of the Proven trait.

Lies, now? That's just sad, even by your standards (none), Nimsim.

Forums poster Lrod-Iniquitsor Fgdsfg, please try to write at least one topical post without an insult.

It doesn't have to be amazing, but has to be not-useless and average damage is the right metric here. Proven(2) is useless more than 90% of the time due to the DoS rule and for Proven to bring frag grenades to the point where they can consistently hurt people you need Proven(5)!
I've seen at most Proven(3), and that was on the Lascannon. Which is... neat against infantry, but against tanks, which is what the Lascannon is for, it's not enough to deal damage.

Have you figured out how basic statistics work, yet?

I think everyone except you have already figured out that you need an entry-level course on statistics. Just because you can make statistics out of something, doesn't mean that it's relevant to the argument. What you are doing is comparable to people arguing over average U.S. income instead of median income.It's misrepresentative.Statistics are useless without the skills to interpret them.

Please use this math to inform your arguments about the Proven quality, unless you are fgdsfg, who is currently devise an alternate mathematical system to probability which he is using to interpret these numbers.

Lies, now? That's just sad, even by your standards (none), Nimsim.

lol

Wow, Nimsim. You'rve utterly failed to understand any of my points or even the point of statistics.

1. I have a PhD in chemistry. I use stats on a regular basis far above the level you're displaying here.

Just because I can use them doesn't mean I should. And just because I haven't doesn't mean you should cast aspertions on my intelligence.

2. Minimum damage has zero to do with the statistics you're quoting. Proven(2) may not increase the statistical damage of a weapon by much, but it decreases the chances of getting a 1 by 100%. That's some pretty powerful stats right there.

You can use stats to prove that the average damage doesn't do much, but for all your brains you have completely failed to understand the point of the trait.

3. I still really don't understand your obsession with making all traits do something amazing.

Some traits do somthing small, and when that happens it's just a nice thing to have happened, rather than some game-breaking powergaming must-have weapon quality. Some people actually play this game for fun, you know?

1. When talking about raw numbers as determined by probability generators (dice), statistics are the best way to represent them. It doesn't take more than a high school education to do this kind of math, which is why I'm surprised that so many people neglect to do it. I cannot believe that I'm having to argue about whether to use statistics to model the effect something has on a die roll.

2. I've already been over this, but that increase to minimum damage is only going to come into effect 10% of the time for proven 2. It's not like adding +1 flat damage, because the maximum damage isn't actually increasing. The proven 2 quality does not exist in a vacuum where only 1s are rolled. In order to take into account the times that Proven 2 won't do anything, you need to look at the average. If you only look at the increase to minimum damage, you ignore all the times that this quality does nothing. When running the math for Tearing, I don't just assume that the second die is always going to be higher than the first, because that skews the value of Tearing. I get that the point of the trait is to increase the minimum, but the increase is so small in proportion to how often it will occur for Proven 2, that it might as well not even exist. There's a reason why modifiers in this game are +10s and occasionally +5s rather than +1s.

3. This is a case of having the trait do something better than a result that could be rounded down to 0. I'm not trying to make it amazing. Proven 7, is about on par with just adding +2 to damage. I don't think most people would argue that +2 damage is "amazing" or game-breaking, given that weapon mods and talents exist that grant this. Please stop mis-characterizing my desire for Proven to be implemented as a useful quality rather than a negligible one. Nice attempt to imply I don't want to have fun with the game because I suggested a house rule to make a mostly useless quality better.

If you want a really accurate representation of what any grenade does, you basically need to do this: Concussive (2), Toxic (1).

Why should all grenades have toxic? Drop a flashbang right next to you and don't take cover. Odds are you'll keel over, because of the blast alone, without any other form of incendiary or shrapnel. Once you get to actual damage rolls, depending on the type of grenade, pick your proven or add some penetration. Decent armour can and will stop the shrapnel from a frag blast, sometimes, so I don't see an issue with fizzes against light carapace and up. But that's really only half the damage of any explosive device.

For all this talk of being math wizards neither of you have put forth a better way to analyze the efficacy of the Proven trait.

Well you see, Proven 7 would add 6 damage to the minimum and that's basically the same as adding 6 damage to the weapon, so it's overpowered. Also, given that rolling a d10 can potentially have 9(!) extra damage over the minimum, it should probably be looked at as well. /sarcasm

Lies, now? That's just sad, even by your standards (none), Nimsim.

Forums poster Lrod-Iniquitsor Fgdsfg, please try to write at least one topical post without an insult.

Hey, I'm making digs at him as well. To be fair, I'm wondering at this point if he's just purposefully trolling.

Lies, now? That's just sad, even by your standards (none), Nimsim.

Forums poster Lrod-Iniquitsor Fgdsfg, please try to write at least one topical post without an insult.

Hey, I'm making digs at him as well. To be fair, I'm wondering at this point if he's just purposefully trolling.

Fair point. It's probably that you're ribbing amounts to making fun of him for what he saying regarding using math to look at game mechanics and his are name calling that irked me.

I guess what I'm saying is if you're going to insult someone at least make it funny for the rest of us.

If you want a really accurate representation of what any grenade does, you basically need to do this: Concussive (2), Toxic (1).

Why should all grenades have toxic? Drop a flashbang right next to you and don't take cover. Odds are you'll keel over, because of the blast alone, without any other form of incendiary or shrapnel. Once you get to actual damage rolls, depending on the type of grenade, pick your proven or add some penetration. Decent armour can and will stop the shrapnel from a frag blast, sometimes, so I don't see an issue with fizzes against light carapace and up. But that's really only half the damage of any explosive device.

Do you have an article or something on this? I sort of remember reading something about being near explosions causing terrible things with the innards, but I think that had something to do with heat and pressure, which I don't think frag grenades have as much of. That and a flash bang is meant to be non-lethal, with the primary effect being a flash that blinds for several seconds and a sound that deafens and affects inner ear liquid, leading to balance issues. Flash bangs are about as hard to kill someone with as a blank bullet, although it is possible.

It doesn't have to be amazing, but has to be not-useless and average damage is the right metric here. Proven(2) is useless more than 90% of the time due to the DoS rule and for Proven to bring frag grenades to the point where they can consistently hurt people you need Proven(5)!

I've seen at most Proven(3), and that was on the Lascannon. Which is... neat against infantry, but against tanks, which is what the Lascannon is for, it's not enough to deal damage.

Hey, I never got why grenades don't all have Concussive. Toxic seems very off to me though.

I'm not arguing for realism though, I'm arguing that it feels wrong in a game with the aesthetics and at least seeming design goals of WH40k if grenades don't consistently hurt people. So give them Proven (4) for a 16% chance of no damage against Joe Guardsman and his flak armour.

On top of that, I'm arguing that there is no case where Proven(<4) makes sense at all, because it is too weak by far. I might as well not have it. If you're straight up asking me whether I want it or not, of course I'll say yes, but that's like saying 'Hey, if you buy 100 bullets, you get one extra for free!' Sure, it's nice, and if that one bullet ever saves my ass, great! But an extra 10 or a full extra reload would be a more reasonable special offer...

Mmn, not for nothing, but doesn't [Tearing] also not increase damage part of the time? Say a player rolls a 10 on a 1d10 weapon. Tearing does absolutely nothing in that case. If the player rolls a 9, Tearing only helps if the extra die is a 10. Then rolling an 8 needs Tearing to roll a 9 or 10, and so on. If your argument is that Proven does nothing a good portion of the time (not disagreeing with you there), then doesn't Tearing have that problem too? While I won't change my opinion that Proven (7) is overkill, I want to know your thoughts on Tearing not being helpful 100% of the time.

Well, Tearing actually increases the average damage by a good bit more, starting with 1.65 for a 1d10 roll, adding a bit more for each extra d10 of damage. It also increases the chance of righteous fury, staring with an almost doubling of the 1d10 chance of 10% to 19%. So tearing adds 1-2 damage AND adds to the chance or righteous fury.

Proven 2-6 don't even add as much average damage as Tearing does, AND don't add to righteous fury. It's only at Proven 7 that more average damage is added than Tearing. The reason I'm comparing these two is because they both are meant to affect the damage rolled on a d10, and are meant to be universally rather than specifically useful. Tearing makes a decent baseline comparison ability, as it is one used for the iconic Bolter, is flavorful, and is fun to use.

Here are a bunch of outputs on anydice:

http://anydice.com/program/490e

That's not really the answer I was looking for. [Tearing] is useless at least 10% of the time on a 1d10 weapon if the user die rolls a 10, similar to how [Proven (3)] is useless 80% of the time if the user rolls a 3 or higher. I wanted your thoughts on how Tearing isn't always useful, though I'm not denying that Tearing does increase average damage. How often will Tearing's extra die roll lower than the original 1d10 for a weapon?

EDIT: Not fully awake right now, so my logic's kind of fuzzy.

Edited by Asymptomatic

Well, you're correct in saying that Tearing may occasionally be of no use, but it could give a benefit of anywhere from +1 to +9, and this benefit occurs more than the 10% of the time that Proven 2 works. The reason I'm still using the average damage is that it combines the chance of benefit/nothing versus how big that benefit is. I honestly don't know the formula for calculating how often 1 die would roll lower than the other, but that really doesn't matter to me, because the average damage takes it into account. So yeah, my answer to "do I care that tearing is sometimes useless" is that the benefit is far more in proportion to the chance of having no effect than Proven 2 through 6 is.

What Nimsim said. And again, 10% chance for Proven(2) to do something is overestimating its usefulness!

Concussive doesn't always work either, nor does Accurate always give a benefit. They are all more often useful than Proven and when they do work, the benefit is much larger.

Edit: Actually Accurate always gives a +10 when aiming, which is a really nice benefit, even when you don't get the extra damage.

Edited by Myrion

The Tearing question was more for my curiosity than a serious argument. I'm still firmly of the opinion that Proven is fine as written, but I also want to find an alternative for people who think Proven is weak or useless.

Tearing is only ever 'useless' in the case where you roll doubles (which on 2d10 is 10%), BUT it let's you roll an extra die in the first place, so the statistics get a little more complicated.

Maybe one of our resident math wizards could figure out how often rolling a second die yields a result greater than the result of the first die.

Tearing is only ever 'useless' in the case where you roll doubles (which on 2d10 is 10%), BUT it let's you roll an extra die in the first place, so the statistics get a little more complicated.

Maybe one of our resident math wizards could figure out how often rolling a second die yields a result greater than the result of the first die.

From my rusty knowledge, rolling a 10 makes Tearing useless 10% of the time or 100% of 10%. Rolling a 9 makes Tearing useless 9% of the time or 90% of 10%. 10% of 10% of the time, or 1%, Tearing rolls a 1 when the original die rolls a 1. So right there is a 20% useless rate, if my logic isn't flawed. I'm not confident, but going by that logic, doesn't Tearing have a significant chance of not doing anything?

So if we define 'not useless' as 'rolling higher on the second die than you did on the first die', then, I think what you do is you look at each face of a d10 and figure out what the chance of rolling higher than that face on another die, then average that out over all 10 faces. If you take 0.1 * (10-{d10 roll}) for all faces of a d10 (because there are 10-X faces higher than X), take the average of those 10 numbers, you get 0.45, or 45%, meaning that 45% of the time Tearing gives you a higher result than a straight d10. Doesn't say anything about how much higher, which I guess is what the average damage is for. Also only really applies to 2d10.

Someone please correct me if I'm wrong.

So if we define 'not useless' as 'rolling higher on the second die than you did on the first die', then, I think what you do is you look at each face of a d10 and figure out what the chance of rolling higher than that face on another die, then average that out over all 10 faces. If you take 0.1 * (10-{d10 roll}) for all faces of a d10 (because there are 10-X faces higher than X), take the average of those 10 numbers, you get 0.45, or 45%, meaning that 45% of the time Tearing gives you a higher result than a straight d10. Doesn't say anything about how much higher, which I guess is what the average damage is for. Also only really applies to 2d10.

Someone please correct me if I'm wrong.

I think you're good, that makes sense. So if Tearing actually only has a 45% "success" rate, how does that compare with Proven (5) that makes half of all rolls irrelevant? Again, this is an experiment rather than an actual argument.

Edited by Asymptomatic

This is also if you're looking at the dice as two distinct occurrences rather than being rolled together. If you roll then together, one of the dice will always be higher than the other unless doubles are rolled, and it's impossible to tell whether the second die or first due was higher. I think that's actually the correct way to calculate Tearing/take the best of Xd10.

Would it help to make proven both stack with itself, and with the exchange-DoS-for-one-damage-roll by adding its level on top ?

E.g.:

You roll 3 DoS with your attack with an proven (2) weapon.

You roll 1d10 damage and every roll that is lower than 3+2=5 gets a 5 automatically ?

This is also if you're looking at the dice as two distinct occurrences rather than being rolled together. If you roll then together, one of the dice will always be higher than the other unless doubles are rolled, and it's impossible to tell whether the second die or first due was higher. I think that's actually the correct way to calculate Tearing/take the best of Xd10.

Fair enough. That is another way of looking at it.

Would it help to make proven both stack with itself, and with the exchange-DoS-for-one-damage-roll by adding its level on top ?

E.g.:

You roll 3 DoS with your attack with an proven (2) weapon.

You roll 1d10 damage and every roll that is lower than 3+2=5 gets a 5 automatically ?

I didn't even realize that DoS-for-die substitution was even a thing until I read the Combat chapter more carefully. When did that become a thing, during <Only War>? Personally, I think you have a reasonable "buff" for Proven there, but I don't think it's me that has to be satisfied.

ADDENDUM: Seems to me like every game other than the first <Dark Heresy> had the substitution mechanic. My knowledge is hella flawed then.

Edited by Asymptomatic

Tbh, I share your dissatisfaction with the current state of proven.

It is of almost no use, and should get some kind of interesting feature.

Maybe it helps to write the devs. They are quite open to suggestions.

The DoS-exchange exists since Rogue Trader IIRC. At least, thats where I saw it the first time.

But its a very hidden rule that is hard to find.

Edited by GauntZero

Yes, on a single die Tearing "only" has 45% chance to help you. However, that's all Tearing promises to do: Give you better odds on rolling high and on scoring Righteous Fury. It does that very well!
Proven promises to make your gun more consistent and to make even your worst rolls meaningful due to the increased minimum damage. It fails at doing that, not least because more than 90% of the time it fails to do anything.

It's an interesting idea for a buff, although I'd have to run the numbers...

Maybe it would already be enough to increase the proven-numbers.

Increase all Proven by 2 and it should at least have a little use.