1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Hemos actualizado nuestra PolĂ­tica de Privacidad acorde con la nueva RGPD. +Info // We've updated our Privacy Policy to comply with the GDPR. +Info
    Dismiss Notice

Is the game getting too complex?

Discussion in 'Access Guide to the Human Sphere' started by Space Ranger, Oct 30, 2018.

  1. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    3,739
    Likes Received:
    4,219
    It's way more complex than you're both arguing. Not least because SWC is a thing. Investing in a BS13 2W ARM4 HMG is a great SWC investment compared to 2 BS12 ARM1 Sptifires.

    It's a matter of list composition and table design (tables with sparse DZs start making fewer, stronger pieces more viable) as much as points.

    There's a few things that do skew in favour of 'not HI' troops though:
    Mimetism is undercosted.
    Armour is overcosted.
    Hacking/EM vulnerability isn't costed in.
    Middling CC (14-16) is overcosted.
    More orders is the only solution to some problems (Xenotech anyone).

    I think overall the points system is pretty good: but it's reached the point where the bits where it less accurately reflects tabletop reality is increasingly obvious.
     
    Phlyk, theradrussian, Lesh' and 6 others like this.
  2. kinginyellow

    kinginyellow Well-Known Member

    Joined:
    Dec 19, 2017
    Messages:
    541
    Likes Received:
    393
    I am not sure what you are referencing here. I was talking about thorkitai. When I was talking about aro defense, I meant from a strategic point of view. A model with a gun who shoots as aro is what I meant as aro defense, not particular intest in that post of shooting at arm or bts. If the post was confusing, apologies.



    You are correct and yet I have found many games where people are spending orders to fight other more significant models. Some people do focus down the netrods, which does hurt my game, but a lot others spend the entire order pool setting up to kill my TR bot or my sniper, or other significant models or push forward towards the objectives. A netrod is not tanky due to any effect it has, but it doesn't have any guns and cant win a mission and so most opponents I find instead target my important models.

    If they shoot at netrods it's generally because they landed in a truly horrible position and the enemy has a spare order, or they were moving past it and had a short skill free.

    But most games I play, unless I am getting stomped, I end the game with the same netrods I started with, or maybe lose 1 out of the 3 I bring.

    So tankyness of the netrod is due to how ignorable it is.
     
  3. Sabin76

    Sabin76 Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    1,150
    Likes Received:
    1,389
    You are right, of course. Perhaps a better comparison would be using a generic attack piece (Wu Ming?) against both and see the difference in how many orders it takes to remove them? Still leaves a lot of variables off the table, but I think it's informative nonetheless. The idea is to show that the opportunity cost of taking an elite model is already factored into the cost (at least somewhat), which is something @Mask doesn't seem to think is true. My main point, if you remember, was to explain why a skill like mimetism should cost more on elite models than on scrubs.

    Wu Ming HMG vs Order Sergeant spit within 24" both in cover (showing that the HMG had to maneuver to get at the OS because nobody leaves their spit on overwatch down lanes greater than 24"):

    66.66% vs 7.63% or 1.5 orders to remove the order (on average) with a 0.58% chance of losing the Wu Ming in 2 consecutive orders.

    If the Wu Ming was able to catch the spit outside of 24", however:

    78.46% vs 4.48% or ~1.27 orders to remove the order (on average) with a 0.20% chance of losing the Wu Ming in 2 consecutive orders.

    Wu Ming HMG vs Suryat HMG both in cover and in +3 bands:

    55.02% vs 9.45% or ??? orders* to remove the order (on average) with a 0.89% chance of losing the Wu Ming in 2 consecutive orders.

    *I've been trying to figure this one out for over an hour now and I can't quite parse it. I think the worst-case scenario for the Suryat is ~2.55 orders, however. Perhaps someone with a better background in probabilities can tackle this. What I think are the relevant stats:

    55.02% chance to do 1+W to the Suryat in each order.
    35.30% chance to do only 1W to the Suryat in each order.
    19.72% chance to do 2+W to the Suryat in exactly 1 order.
    19.42% chance to do 2+W to the Suryat in exactly 2 orders.
     
  4. toadchild

    toadchild EI Aspect

    Joined:
    Feb 22, 2017
    Messages:
    2,497
    Likes Received:
    4,729
    I would recommend going with expected values. Calculate how many wounds you expect to do per turn (capping at unconscious) and then divide the number of wounds by that to get the expected number of turns to disable them.
     
    Phlyk and Section9 like this.
  5. Section9

    Section9 Well-Known Member

    Joined:
    Nov 24, 2017
    Messages:
    5,791
    Likes Received:
    9,124
    IIRC, that is the 'correct' way to do it per the mathemagicians.

    My dad (PhD in Math) did an expected value analysis on the Powerball lottery, came up with the answer of "don't buy a ticket until the jackpot is over 400mil".
     
    toadchild likes this.
  6. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    3,739
    Likes Received:
    4,219
    It also helps that this is really intuitive:
    33% odds of doing a wound per order is expectation of 3 orders to take UNC.

    So: 35% odds of 1W + 20% odds of 2W + 4% odds of 3W = 0.87W / order. So 2.3 orders to take UNC. (Although I'm assuming away more than 3 wounds for easy maths, so it's marginally lower).
     
    #686 inane.imp, Feb 2, 2019
    Last edited: Feb 2, 2019
    toadchild likes this.
  7. Mask

    Mask Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    96
    Likes Received:
    78
    I think skills shouldn't cost more on elite troops (which is anyway fun and helpful to invest swc in) because when I choose to field one I'm investing not only it's cost but I'm "paying" also with the overall capacity to field "more orders" and "more threats". Obviously less scary threats but nonetheless more in numbers and with the possibility to cover more "vectors".

    To put that with an empirical example.
    To bring 300 points on the table I can choose 15 troops with a medium cost of 20 points or 10 troops with a medium cost of 30. (of course 6 swc for both and of course I choose oversemplificated example to put that in concept :kissing_heart:)
    In almost every case I found more competitive the first. The more I blend towards the second the more I think I'm loosing much more than I gain.
    I loose orders. I loose the possibility to cover more "vectors" on the battlefield. I loose some "speed bumps" for my opponent which are of course going to die during the game but wisely positioned (spread in tactical position) could be an order sink for my opponent. More than a single elite troop could be.
    And now the point.
    If elite troops doesn't "pay more" for skills and cool things we could end up with a list where I can have the 10 elite troops mentioned earlier ... Aaand... Some points left to put some more threats. Maybe I can reach 13 orders?? Then my "perceived competitiveness" of the example could put the elite list in the "worth buying" chest.

    Mask
     
    #687 Mask, Feb 2, 2019
    Last edited: Feb 2, 2019
  8. Section9

    Section9 Well-Known Member

    Joined:
    Nov 24, 2017
    Messages:
    5,791
    Likes Received:
    9,124
    Point of order: That's spelled "median", not "medium", and you should be using "mean" or "average" there. Median means the price of the majority, so if you have 9 units costing 20 points and one costing 100, the median cost is 20 points; while the mean/average cost is 30.

    But yes, orders matter. Nonetheless, CB has (rightly, IMO) concluded that because some skills become more combat-effective as other stats increase, they need to be more expensive.
     
    inane.imp and meikyoushisui like this.
  9. Sabin76

    Sabin76 Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    1,150
    Likes Received:
    1,389
    Yeah, it wasn't hard for the 1W model at all. My brain just didn't want to figure out how to account for the multiple wounds. But wouldn't it be 16% on the 2W since you are taking out the 3+W part? Which would make it 0.79W / order or roughly 2.54 orders to take out the Suryat.

    In other words, the Suryat is about twice as order intensive to kill with a dedicated attack piece as well as being over twice as likely to give a wound back for your efforts.

    Again, I'm not trying to argue that LI lists are superior, or even on equal footing to order spam. I'm not trying to argue that HI are totally worth their overall cost. What I am arguing is that a skill that is demonstrably better should cost more, and a skill like mimetism is demonstrably better on a high BS platform than a low BS one.
     
    Section9, inane.imp and toadchild like this.
  10. toadchild

    toadchild EI Aspect

    Joined:
    Feb 22, 2017
    Messages:
    2,497
    Likes Received:
    4,729
    Lotteries are of course further complicated by the fact that if multiple people pick the winning number (which is more likely to happen the higher the jackpot goes and the more tickets are bought) then they split it, causing the expected return per ticket to drop precipitously.
     
  11. Mask

    Mask Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    96
    Likes Received:
    78
    Then we have to agree to disagree.
    I'm firmly convinced that high bs platform (high PA cost) bring also some downside that'll need to be considered in the costing algorithm.

    Mask
     
  12. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    3,739
    Likes Received:
    4,219
    Yeah, I forgot how the Dice Calc displays stats when I reworked it to account for the 3+W.
     
  13. inane.imp

    inane.imp Well-Known Member

    Joined:
    Jan 28, 2018
    Messages:
    3,739
    Likes Received:
    4,219
    Then why doesn't a HMG cost more on a BS13 platform than on a BS11 platform? That's the logical extension of your argument. Partly this is accounted for with SWC, but when you have situations like the Wildcat HRL and a Kamau HRL that seems less well defined.

    There's also other synergies that aren't costed in: why should a Prowler and a Bolt pay the same amount for Bio-immunity? Bio-immunity on a Bolt is far more effective than on a Prowler due to the different BTS. And that's leaving aside things like Camo + Infiltration.

    That being said, I'm actually not opposed to scaled skills: but it should be offset by scaled discounts. Isolated is a prime example: the effectiveness of Isolated scales with the cost of the target. This lack (no built in discounts) is a large part of the reason 'vanilla' HI (Suryats, MB, Orcs etc) are often maligned: they get hit by the scaling on BS, PH and ARM but lack any compensation (compare instead the WM Chain Rifle SMG profile, which gets the benefit of an extremely cheap weapon loudout).

    Given the same stats and skills a HI should be cheaper than an MI or LI. Hacking and EM vulnerability needs pricing in.
     
    Mask, Sabin76 and AdmiralJCJF like this.
  14. Sabin76

    Sabin76 Well-Known Member

    Joined:
    Nov 23, 2017
    Messages:
    1,150
    Likes Received:
    1,389
    I actually think it should, but as you admitted, the SWC cost is double for an HMG on something like a vanguard vs a suryat. Should it be more or less? Again, I don't have the answer to that. I never tried to say how much more a skill should cost on one profile over another, just that it should cost more. I've never bothered to dig into the cost formula, myself. I do know that some skills/loadouts seem to be criminally underpriced and that synergies are not considered and I agree that both of those things needs to be looked at.
     
  15. Section9

    Section9 Well-Known Member

    Joined:
    Nov 24, 2017
    Messages:
    5,791
    Likes Received:
    9,124
    That's mostly because you need an actual statistician to crunch those numbers to come up with an appropriate costing formula. I've had two semesters of statistics, which is only just enough to actually understand how they work (I finally 'got' stats about 4 weeks into 2nd semester). I am nowhere near good enough to derive a scaled cost formula for a game. I'm not sure my Dad is, either (and as I said, he has a PhD in Math, but not statistics).

    My guess is that CB has made higher BS a little more expensive than it's literal % improvement across all weapons, since players will tend to put their better guns on high-BS platforms. So the cost for a BS13 HMG is really carried in the BS13, instead of in the HMG. Simpler math.



    That I totally agree with. It's really obvious in MO and IA.
     
    Phlyk, Sabin76 and RogueJello like this.
  16. toadchild

    toadchild EI Aspect

    Joined:
    Feb 22, 2017
    Messages:
    2,497
    Likes Received:
    4,729
    Personally, I think trying to make a bulletproof points pricing formula is a fool's errand, and effort needs to be put into making sure the specific loadouts have a cost that makes sense for what they offer within the context of their army. That's not to say that I don't think specific pieces of equipment can't be under- or over-costed, but rather that it's not worth the effort to try and calculate out every last variable to come up with a single number that is objectively the correct cost.
     
  17. Section9

    Section9 Well-Known Member

    Joined:
    Nov 24, 2017
    Messages:
    5,791
    Likes Received:
    9,124
    It's very difficult, that's for sure.

    The most insane cost formula I know of is the one from Star Fleet Battles, which includes range-to-target as a variable. Get one of the 'cheap' Kzinti Strike Cruisers up close to someone and the formula says it goes up about 20-30% in cost! (because all the short-range defensive phaser-IIIs can now be used to shoot up the opposing ship, instead of being out of range and unable to do damage)


    This is why I suspect that CB figured the cost increase for BS based on the effectiveness of a heavy weapon, not across the board. It's mathematically simple and makes sense for what they offer in the army.
     
  18. RogueJello

    RogueJello Well-Known Member

    Joined:
    Nov 24, 2017
    Messages:
    444
    Likes Received:
    434
    There is no such thing as a completely objective points value for most skills, stats and equipment. You'll find that either the skill, stat, etc is useful for a particular game or it is not.

    A good example is MSV, which gets better the more enemy models have CH skills, but is completely worthless when used against a foe with no CH troopers. You can extend this to most terrain skills, and even some of the weapons.

    While a combi rifle is usually going to do about the same thing every time, AP weapons gain in utility as enemy armor increases. Similarly Shock is much more damaging against foes with some level of Valor, or troops with Doctor/Paramedic skill. EXP is much more damaging to models with multiple wounds because it's more likely to cause multiple wounds per successful hit.

    So the best you can hope for is to be in the right general area, and not have things be too odd when comparing across factions.
     
    Mask, inane.imp and toadchild like this.
  19. Wombat85

    Wombat85 Well-Known Member

    Joined:
    Nov 25, 2017
    Messages:
    413
    Likes Received:
    506
    Something I have wanted to make a video on is the objective surface for the game, but to be honest im not sure there is enough demand for such a video for the time it would take to create.
     
  20. toadchild

    toadchild EI Aspect

    Joined:
    Feb 22, 2017
    Messages:
    2,497
    Likes Received:
    4,729
    But the point is that you can't know all of that at list-building time in the case of Infinity.You are necessarily making choices blind, and weighing them against some sort of average expectation of what your opponent is likely to bring, and what sort of terrain layout you're likely to see. A classic extreme example is a super densely packed table, where 5 point warbands with chain rifles are king, and long range weapons like snipers and HMGs are frankly terribly.
     
    RogueJello and Mask like this.