Page 2 of 2 FirstFirst 12
Results 41 to 52 of 52

Thread: "Victim Stats"?

  1. #41
    Annihilator Nyxu's Avatar
    Join Date
    Feb 2015
    Posts
    738

    Default

    on the def vs arm matter, there's some shifty and misleading stuff going around with the values for two reasons:

    Reason the first:
    We're operating on something that resembles normal distribution. (2d6 is actually triangular, but bear with me.)
    (courtesy of Mit.edu)

    The takeaway here is that steps toward the mean become progressively more impactful, steps away are progressively less.
    Example: You're RAT 7, shooting a DEF 10 target. You need 3s to hit. The likelihood of you hitting is 97.22% If we change this model to be DEF11, you'll hit 91.67% of the time. That one point of defense is worth 5.55%. Same target, but it's got concealment. 5's to hit, you're hitting 83.33% of the time. While just going up to DEF 11 is worth 5.55, going up to DEF 12 is a whole 13.89% increase in miss rate (Meaning that the second point of DEF was worth 8.34%). Same attacker, same defender, but cover? 7s to hit, hitting 58.33% of the time. Cover increases the miss rate by 38.89%, meaning that the third and fourth points of DEF are worth 25%

    Same shooter, much more slippery target. DEF 14 base. Hitting 58.33% of the time. One point nets you 16.66% additional survivability- pretty serious. Concealment gets you up to DEF16, which takes 9s to hit, you're hitting 27.78% of the time. It's rare, but it's less significant an accuracy hit than going to DEF15, as it only adds 13.89% to the miss rate (Concealment is worth 30.55% here) . DEF18 takes 11s to hit, 8.33% chance, so while cover on a DEF 14 model vs RAT7 strips away a whopping 50% of the hit rate, the third and fourth point are worth less than the first two you get from concealment.

    As shown, every step away from 2d6=7 is less meaningful than the last, and every step towards 2d6=7 is conversely more meaningful.
    Addendum: RAT7 is used because it represents both "accurate" guns and "average" guns aiming.. that and it hits DEF10 on 3s, unlike RAT6 or RAT5.

    How does this impact armor?
    ARM values comfortably go higher, but so do POW/P+S values. This impact is a lot more evident on single wound models like infantry, as any success leads to a disabled model, whereas multi-wound models also measure degrees of success. It'd take a lot more math to determine whether or not the impact is greater on multi-wound models than on single-wound models.

    Since POW7 is not the measuring stick of weapons, we need to shift the distribution over a few points. If POW 10 is considered the "baseline", as we'd done with RAT 7, then ARM 13 is equivalent in value to DEF 10, ARM 15 to DEF 12, and ARM 17 equivalent to DEF 14. Extending to 2d6=1 and 2d6=13, ARM values cease to match with DEF, as DEF0 is missed by a roll of 2, and DEF100 is hit with a roll of 12. ARM0 is damaged by any roll, and ARM100 damaged by no rolls at all.

    Reason the second:
    Conditional defensive benefits do not have equivalent conditional armor benefits. Conditional damage benefits do not have equivalent accuracy benefits.
    In short: Concealment, Cover, Elevation, and attacking over an obstacle exist. These give bonuses to DEF based on terrain. Charging gives boosted Damage rolls outside of Focus/Fury manipulation or special rules. No such universal rule grants boosted accuracy, save for aiming.

    tl;dr DEF and ARM of equal value are not of equal value. in a RAT7 POW 10 world, ARM provides as much protection as 3 points less of DEF. Numbers courtesy of http://anydice.com/program/a5e8

    Addendum to tl;dr: In a RATX POWY world, ARM provides as much protection as (Y-X) points less of DEF
    Last edited by Nyxu; 01-11-2017 at 10:50 PM. Reason: oh jeez I'm my own page.

  2. #42

    Default

    Single wound.
    Below def 13. (mat/rat 6 hits with more than 0.5 probability, essentially you are more likely to be hit than not)
    Below arm 17. (pow 10 damages with more than 0.5 probability, essentially you are more likely to get damaged than not)

    Good stats on single wounders are def 17+ or arm 20+ after all modifiers, you're safe with those against common threats.
    Last edited by Smooth Criminal; 01-12-2017 at 12:15 AM.

  3. #43
    Destroyer of Worlds Stormpuppy_Infantry's Avatar
    Join Date
    Jul 2012
    Posts
    9,786

    Default

    The chance of 2d6 is....


    Roll Hit Critical
    3+ 97.22..% 13.88..%
    4+ 91.66..% 13.88..%
    5+ 83.33..% 11.11..%
    6+ 72.22..% 11.11..%
    7+ 58.33..% 8.33..%
    8+ 41.66..% 8.33..%
    9+ 27.77..% 5.55..%
    10+ 16.66..% 5.55..%
    11+ 8.33..% 2.77..%
    12+ 2.77..% 2.77..%
    You create your own destiny, and your lot in life is not fixed at birth.
    Truth is truth, regardless how bitter it is.
    Anti armor ranged damage per point cost for Haley1
    Stormblade Infantry, fragile but versatile infantry.
    Clockwork Oracle, Warmachine&Hordes probability calculator for Android.

  4. #44
    Destroyer of Worlds Cyel's Avatar
    Join Date
    Jun 2014
    Location
    Warsaw, Poland
    Posts
    1,327

    Default

    Quote Originally Posted by DrakeRaids View Post
    Victim stats means a model dies more than half the time to a Mat/Rat 5 Pow 10 which is generally considered the lowest common denominator.
    In that case 12/15 is fairly far from victm stats as it has a 58% chance to hit further decreased by the requirement of rolling a 6 to wound (72%)

    0,58 x 0,72 =0,41 so only 2 in 5 attacks result in the target models dying.

    To get the fairly good chance of 50% (closer to 52 actually) you need stats of 12/13. It still means those RAT5 POW10 troops need 2 attacks to kill one such model.
    There's no such faction as "Menoth". You can't "play Menoth"

    There's the Protectorate.

  5. #45
    Annihilator Ephraim's Avatar
    Join Date
    Feb 2008
    Location
    Minneapolis, MN
    Posts
    648

    Default

    Quote Originally Posted by Cyel View Post
    In that case 12/15 is fairly far from victm stats as it has a 58% chance to hit further decreased by the requirement of rolling a 6 to wound (72%)

    0,58 x 0,72 =0,41 so only 2 in 5 attacks result in the target models dying.

    To get the fairly good chance of 50% (closer to 52 actually) you need stats of 12/13. It still means those RAT5 POW10 troops need 2 attacks to kill one such model.
    I had the same concern earlier in the discussion, when that criterion was first advanced. I think that using 65% survival against 5/10 or 50% survival against 6/12 makes more sense.

    I had a particular matter in mind when I proposed this question. Namely: Is it worthwhile to cast Death Ward on Revenant Crew of the Atramentous. I concluded that it does since it raises their survival rate against 5/10 from 59% to 65%. Considering that the one list I use them in fields 20 revenants and they can regenerate their numbers, I can see a 1/20 improvement in survival rate making a difference in my board presence.

  6. #46

    Default

    Don't just think of it as a 1 in 20 chance. -5% chance to be killed is nearly useless if you go from 100% to 95%, but very good if you go from 5% to 0%.

    Instead, think of it this way: After 100 shots, you would have had 41 dead and instead now you have 35 dead. 35/41 is about 85%, so that's a 15% reduction in casualties. Sure, the bonus is still only meaningful 1 attack in 20. But the 20 counts troopers that didn't need saving because their preexisting defenses were sufficient.
    Normal voice. Mod voice.

  7. #47
    Destroyer of Worlds
    Join Date
    Nov 2009
    Posts
    5,767

    Default

    I hear the term "victim stats" get used from time to time but I'm not really sure what different people mean by it. I understand, conceptually, that it means a model that will die to any directed effort (and sometimes as collateral damage). Numerically, what qualifies?
    victim stats is like DEF12 and ARM that isnt high enough to stop any damage.

    In that case 12/15 is fairly far from victm stats as it has a 58% chance to hit further decreased by the requirement of rolling a 6 to wound (72%)
    12/15 is definitely victim stats.

    If POW 10 is considered the "baseline"
    POW10 is not baseline. not in the current meta which is geared towards cracking warjack ARM.

    DEF12 with ARM15-16 might have been semi-effective in the past when POW10 was more baseline. But its not very good with everyone gearing up with higher POW weapons. Its victim stats nowadays.

    The lowest POW weapons before buffs in most of my lists is POW13. Because anything lower that that is just not useful vs warjacks.
    Last edited by Soulblighter; 01-12-2017 at 09:02 AM.

  8. #48
    Annihilator Nyxu's Avatar
    Join Date
    Feb 2015
    Posts
    738

    Default

    I'll have to, on a day when I'm horribly bored and have nothing more entertaining to do, go through and catalogue all the various pow/*AT numbers, possibly weighted by points cost (is that a decent measurement? Something like 20/PC per attack?)

    Anyway, point being it would let us look at a couple different metrics to see what the average baseline stats are. And by baseline I mean without buffs. Also, how would maths-minded folks suggest tackling weapon master? I'm leaning towards a separate category, since they do have a more bell-shaped curve and as such change the marginal value of additional points of ARM. (Amusingly, this also means that a 5pt ARM swing is not necessarily worth the same as a 2/3 swing, and in some cases is worth less than a 2/2)

  9. #49
    Annihilator Nyxu's Avatar
    Join Date
    Feb 2015
    Posts
    738

    Default

    Quote Originally Posted by Nyxu View Post
    I'll have to, on a day when I'm horribly bored and have nothing more entertaining to do, go through and catalogue all the various pow/*AT numbers, possibly weighted by points cost (is that a decent measurement? Something like 20/PC per attack?)

    Anyway, point being it would let us look at a couple different metrics to see what the average baseline stats are. And by baseline I mean without buffs. Also, how would maths-minded folks suggest tackling weapon master? I'm leaning towards a separate category, since they do have a more bell-shaped curve and as such change the marginal value of additional points of ARM. (Amusingly, this also means that a 5pt ARM swing is not necessarily worth the same as a 2/3 swing, and in some cases is worth less than a 2/2)

    Aaand it's just occurred to me that ARM valuation is on an entirely different curve than def valuation, as the value of arm against a given attack profile has marginal benefits based both on DEF and ARM curves: a 50% resist rate for ARM is worth more at lower DEF values than at high DEF values- infinitely high DEF makes ARM against fully rolled attacks worth only 2.78% of what would be calculated without respect for the Attack roll.
    Also I'm pretty sure I wrote about this at some other point.

  10. #50
    Annihilator Ephraim's Avatar
    Join Date
    Feb 2008
    Location
    Minneapolis, MN
    Posts
    648

    Default

    Quote Originally Posted by Nyxu View Post
    Aaand it's just occurred to me that ARM valuation is on an entirely different curve than def valuation, as the value of arm against a given attack profile has marginal benefits based both on DEF and ARM curves: a 50% resist rate for ARM is worth more at lower DEF values than at high DEF values- infinitely high DEF makes ARM against fully rolled attacks worth only 2.78% of what would be calculated without respect for the Attack roll.
    Also I'm pretty sure I wrote about this at some other point.
    Should these considerations also take into account Anatomical Precision?

  11. #51
    Annihilator Nyxu's Avatar
    Join Date
    Feb 2015
    Posts
    738

    Default

    Quote Originally Posted by Ephraim View Post
    Should these considerations also take into account Anatomical Precision?
    For auto-success the relevant stat has marginal value of 0.

  12. #52
    Destroyer of Worlds jandrese's Avatar
    Join Date
    May 2006
    Location
    Herndon, VA, US
    Posts
    19,833

    Default

    Quote Originally Posted by vaildin View Post
    only until you get knocked down.
    And the ARM is only more valuable until the other guy has the charge die...
    NoVA players: Join the Nova Nomads.
    Quote Originally Posted by BarskeMannen View Post
    Lamentation, Boundless Charge and charging already accomplishes this to some extent. The most important part of the tech is to charge specifically their caster while looking your opponent dead in the eye and flaring out your nostrils. Establish table dominance from turn 1 and onward.

Page 2 of 2 FirstFirst 12

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •