Necromunda combat stats

thisisalie

Ganger
Feb 6, 2012
62
44
18
I do enjoy your write ups enyoss.

I totally agree true grit needs a buff. It also has some problems versus multiple wound weapons or multiple shooters, where it becomes a liability.

Example - Losing 2 wounds: Looks still in favour of true grit not being detrimental. 3/36 of leaving you pinned minus some face (Yay!), 4/36 chance of giving you a flesh wound for no useful reason (see - 1 flesh wound and down). If I'm down I want as few flesh wounds as possible.

True grit - 11/36 OOA, 4/36 2 flesh wounds, 12/36 1 flesh wound and down, 9/36 down.
Normal - 11/36 OOA, 1/36 2 flesh wounds, 8/36 1 flesh wound and down, 16/36 down.

3 wounds. I may regret having true grit about now. Not sure getting 3 flesh wounds and pinned is good... Sure I can run away, but the chance of putting me OOA with my own talent bugs me. To avoid this being a tactics discussion I'll just ignore it. 24/216 of results become +2 flesh wounds and down and 6/216 of results become +1 flesh wounds and down. Given that 125/216 of these results are not OOA, having 24/125 of these giving me 2 more flesh wounds than the default seems awful.

True grit - 91/216 OOA, 8/216 3 flesh wounds, 36/216 2 flesh wounds and down, 54/216 1 flesh wound and down, 27/216 down.
Normal - 91/216 OOA, 1/216 3 flesh wounds, 12/216 2 flesh wounds and down, 48/216 1 flesh wound and down, 64/216 down.

This just gets worse for true grit as it begins handing out more flesh wounds for little gain as wounds inflicted increase. This maths is also valid for people shooting downed true grit people. (shoot them it's worthwhile)


My suggestion: True grit's should just prevent flesh wounds on 1 and flesh wound on 2, that way they never increase the amount of flesh wounds gained.

1 - Pinned only
2 - Flesh wound and pinned
3-5 - Down
6 - OOA
 
Last edited:

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
I do enjoy your write ups enyoss.

I totally agree true grit needs a buff. It also has some problems versus multiple wound weapons or multiple shooters, where it becomes a liability.

Example - Losing 2 wounds: Looks still in favour of true grit not being detrimental. 3/36 of leaving you pinned minus some face (Yay!), 4/36 chance of giving you a flesh wound for no useful reason (see - 1 flesh wound and down). If I'm down I want as few flesh wounds as possible.

True grit - 11/36 OOA, 4/36 2 flesh wounds, 12/36 1 flesh wound and down, 9/36 down.
Normal - 11/36 OOA, 1/36 2 flesh wounds, 8/36 1 flesh wound and down, 16/36 down.

3 wounds. I may regret having true grit about now. Not sure getting 3 flesh wounds and pinned is good... Sure I can run away, but the chance of putting me OOA with my own talent bugs me. To avoid this being a tactics discussion I'll just ignore it. 24/216 of results become +2 flesh wounds and down and 6/216 of results become +1 flesh wounds and down. Given that 125/216 of these results are not OOA, having 24/125 of these giving me 2 more flesh wounds than the default seems awful.

True grit - 91/216 OOA, 8/216 3 flesh wounds, 36/216 2 flesh wounds and down, 54/216 1 flesh wound and down, 27/216 down.
Normal - 91/216 OOA, 1/216 3 flesh wounds, 12/216 2 flesh wounds and down, 48/216 1 flesh wound and down, 64/216 down.

This just gets worse for true grit as it begins handing out more flesh wounds for little gain as wounds inflicted increase. This maths is also valid for people shooting downed true grit people. (shoot them it's worthwhile)


My suggestion: True grit's should just prevent flesh wounds on 1 and flesh wound on 2, that way they never increase the amount of flesh wounds gained.

1 - Pinned only
2 - Flesh wound and pinned
3-5 - Down
6 - OOA
Cheers matey :).

Your point about True Grit has much more impact than the minor rules tweak I suggest too. Interesting to see it nailed down with some concrete numbers!

In the past I've thought a workable solution would be to let the owning player choose when to activate the skill. That's a bit of a cop out in two ways though:
  1. It really grates thematically - I just can't see a guy who is so hardened he'll go on fighting while holding his gut-sausages in (i.e. FW with other Down results) choosing to whimper on the floor rather than hold in said gut-sausages (Down with other Down resutls).
  2. The fact a skill can regularly become more competitive if you pretend you don't have it seems wonky.
So I think your suggestion is better in that it gets round both of those problems. I have to admit that I haven't really given much thought to how powerful it is though - if it's considered stupendously game breaking I guess you could always say the skill just lets you ignore any downsides from Flesh Wounds (i.e. similar to what you have already, but re-combining the '2' result back in with the Down category).

My gut feeling is that as it's a negative passive skill, i.e. a situational skill that only kicks in to ameliorate your predicament when things are going badly, it can afford to be skewed slightly towards the powerful end if it comes to it.

Then again, I would say that - I'm playing Cawdor in the next LAGGNOG campaign :D
 
  • Like
Reactions: TakUnderhand

thisisalie

Ganger
Feb 6, 2012
62
44
18
My gut feeling is that as it's a negative passive skill, i.e. a situational skill that only kicks in to ameliorate your predicament when things are going badly, it can afford to be skewed slightly towards the powerful end if it comes to it.
So I have a similar feeling about most of the ferocity table. Most of the table gets weaker as the game progresses... Berserk charge gets weaker with attacks advances (see earlier in this thread for enyoss' maths), Killer Rep vs enemy LD increases, True Grit and Nerves vs Offensive talents (Specialist (Plasma guns/Autoslugger), Gunfighter, Crack shot), Iron will gets weaker with LD increases.

Only Impetuous remains as good as it was. Nerves can get better with the right advances (Mostly wounds but also init) but wounds are uncommon (1/18)...

I feel that the defensive talents should be stronger than the offensive ones, as the offensive talents have the added flexibility of who you use them on.
 
Last edited:
  • Like
Reactions: enyoss

thisisalie

Ganger
Feb 6, 2012
62
44
18
maths for New Grit:

1 - Pinned only
2 - Flesh wound and pinned
3-5 - Down
6 - OOA

2 wounds.

New grit - 11/36 OOA, 1/36 pinned, 2/36 flesh wound, 1/36 2 flesh wounds, 6/36 1 flesh wound and down, 15/36 down.
True grit - 11/36 OOA, 4/36 2 flesh wounds, 12/36 1 flesh wound and down, 9/36 down.
Normal - 11/36 OOA, 1/36 2 flesh wounds, 8/36 1 flesh wound and down, 16/36 down.

3 wounds

New Grit - 91/216 OOA, 1/216 pinned, 3/216 1 flesh wound, 3/216 2 flesh wounds, 1/216 3 flesh wounds, 9/216 2 flesh wounds and down, 42/216 1 flesh wound and down, 66/216 down
True grit - 91/216 OOA, 8/216 3 flesh wounds, 36/216 2 flesh wounds and down, 54/216 1 flesh wound and down, 27/216 down.
Normal - 91/216 OOA, 1/216 3 flesh wounds, 12/216 2 flesh wounds and down, 48/216 1 flesh wound and down, 64/216 down.

So has very little difference in resisting large amounts of wounds to not having it. For less wounding hits, it is similar to a 1/6 armour save on the wound received but only when wounds are reduced to zero 0. If downed, causes the fighter to recover slightly faster (the value of this is highly variable) and reduces the chance of injury by (up to) 33% (Getting shot, engaged in melee etc. reduce this number). This 33% is about half as effective as tough as nails as it only works on downed fighters who are probably about half of injury rolls (for every 4 down, 1 is OOA but this doesn't include melee or recovery rolls. Furthermore, down fighters have a 50% chance of not needing to roll on injury tables).

This would now be slightly more powerful than crack shot (unless the crack shot guy has a multi wound weapon....), and similar usefulness to iron jaw (iron jaw scales well with toughness and/or wounds (1/9 advances) and badly vs weapon strength and range). Dodge is better than this on 2+ wound gangers vs ranged attacks.
 
Last edited:

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
An unexpected detour...

So, I've recently been working on ways of play testing different reinforcement and deployment set-ups in the Hit & Run scenario.

And this is quite tricky – there are 18 different combinations for the initial deployment set up, and the first turn can go to either attacker or defender. So to get a real-life feel for the initial scenario conditions, you have to play at least 36 games right off the bat – and that's before you even factor in the huge array of potential outcomes which hinge on the dice rolls in that first turn.

A quick back of the envelope calculation suggests that from the moment the scenario is selected, the system diverges very quickly into approximately 1000+ potential states by the end of just the first turn. Good luck playtesting that one thoroughly :D. Then you have turn two...

And this train of thought set me on an interesting (to me at least... please resist commenting on that) trajectory, whose final destination is the answer to one question:

“Using just the probabilities of various outcomes during the games and post-game sequences of a Necromunda campaign, is it possible to simulate the ebb and flow for any given scenario (or game) of Necromunda probabilistically?”

To be upfront about it - I think the answer is probably no :D.

But you know what?... I think you can get close. It's fraught with assumptions and idealizations, but I think it's possible to get something which, in certain circumstances and with varying sized pinches of salt, can roughly simulate the macro-statistics of 1000s of real-life games.

Sure, for any individual fighter it's impossible to do. But for the macro-state of a gang during a game, taking into account probabilities associated with the in-game and post-game actions of each individual fighter rather than the actualities of what they do, I think it's possible.

I'm not sure how useful it is, given all the assumptions and things you necessarily have to leave out, but ho-hum. It should be feasible though, however remotely, to at least use these macro-statistics to assess the impacts of e.g. initial defender numbers or reinforcement setups in each turn of a hypothetical game,

And so bit by bit, I've built an engine to do this. To get started though, it's worth going back to square one: advance probabilities...
 
Last edited:

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
Advance likelihoods

The first building block for the engine is the probability of getting various advances at different stages of a campaign. This adds on existing work done in this thread (https://yaktribe.games/community/threads/advances-probabilities.5449/), which looked at the average increase to each profile characteristic when a fighter has reached 401+ experience.

I'll present something slightly different here.

Instead of the average increase to each profile characteristic, I'll look at the probability of getting 0/1/2/3 advances in that characteristic. And instead of looking at 18 or so advances (up to 401+ experience), I'll look at how the probability of rolling up 0/1/2/3 advances varies as the total number of advances a fighter has rolled up, e.g. 3/6/9/12/15/18.

This should give us an idea of how likely a fighter is to have +1BS (for example), given how far along they are through the campaign.

I'm guessing that @Crav37 has these stats banging around somewhere as well by the way, although they're not posted in that thread.

So, here are plots which show the probability of having a certain number of BS, T, W, I, LD advances, for a fighter who was hired in at ganger level. The other characteristics which I'm not showing can just be mapped to the ones already shown using WS = BS, S = T, A = W.

CharacteristicAdvancesProbs_RandomOverflow_BS_Ganger.png


CharacteristicAdvancesProbs_RandomOverflow_T_Ganger.png


CharacteristicAdvancesProbs_RandomOverflow_W_Ganger.png


CharacteristicAdvancesProbs_RandomOverflow_I_Ganger.png


CharacteristicAdvancesProbs_RandomOverflow_LD_Ganger.png



Due to complications which come about due to advance 'overflow', i.e. when you cap out a characteristic and any further advances to the same characteristic have to go somewhere else, I chose to simulate these numbers instead of calculating the probabilities exactly.

This involves levelling up a fighter to the number of advances given (e.g. 6), and dealing with advances as you would in a real game – if you've capped out T you would increase W instead, and if both are capped out you can go on to increase any other currently uncapped characteristic of your choice.

In terms of this engine, it does that last bit by just picking a random uncapped characteristic and increasing it. I have other versions which target desired characteristics rather than selecting at random, (e.g. choosing to increase W before I), but the situation arises so infrequently it doesn't really change the outcome enough to worry about it.

And once you've done that for one fighter, you then do it again, and again... 10,000 times... and then calculate the statistics.

Anyway, some choice observations from the plots:

BS

Early campaign, when a ganger has just three advances, there's over a 60% chance they are still BS3. By the time they reach 6 advances, that drops to 40% - there's another 40% chance they'll be BS4, around 15% chance of BS5, and around 5% chance of BS6.

Another way of thinking about that is that for two opposing gangs with ten ganger level fighters each, in the mid-campaign when each fighter has hit 6 advances, of those twenty fighters combined over both gangs:

8 will have BS3
8 will have BS4
3 will have BS5
1 will have BS6

By the time your fighters are getting to 9 or more advances, they are more likely to have BS4 or BS5 than they are BS3, which is encouraging I guess.

When I combined this with the sister probability distribution for WS advances throughout a campaign, I discovered that my lack of any WS or BS advances over 20-30 total advances across the gang in my last campaign was as unlikely and unfortunate as it felt at the time. Stupid dice.

T/W/I/LD

These are a bit different to the BS/WS advances, as until you get to a very mature stage of the campaign (i.e. fighters with 10 advances or more... that's 150 experience gained by the fighter) the single most likely number of advances to have for these characteristics is zero. That is, although the probability of having one or more advances is still more likely than having zero, the probability of having zero is higher compared to the probability of having one and only one, or two and only two etc..

This is because the likelihood of getting these advances on any given roll is much smaller than the chance of getting BS or WS. Simples.

Anyway, that's that done. So now to move onto how to use this information in our game simulation engine...
 
Last edited by a moderator:

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
Ambush Revisited...

Time for another detour *sigh*.

Things have been busy round these parts, and despite having lots to update on I find myself with little time to actually write it up. What time I do have spare has been spent fashioning (over-elaborate) rules for and playing in a fresh campaign – the LAGGNOG “Crouching Spyrer Hidden Bastard” affair.

And so far, in contrast to how my last campaign started, it has been going surprisingly well.

My last game was against Spyrers – a blind fight ambush, with me defending. And to my surprise I managed to win, in no small part due to getting pretty lucky and having two of my defending groups deploy as surprise drops away from the centre. Going way back to my first post on the Ambush scenario in this thread, this gave me a 70% chance of taking the first turn, compared to a 20% baseline chance which I would normally expect when defending with 12 fighters. So far so lucky.

But, more importantly, the second group I got to deploy rolled a 6 and was therefore a surprise drop I could place away from the central group. Which meant that no matter what happened with the later groups, thanks to the Ambush rules allowing you to deploy groups within 4” of any previously deployed fighters, I could now deploy all of my remaining groups away from the chumps getting ambushed in the centre of the board.

This was pretty critical to my success since there were only five Spyrers in the opposing gang, and although they had the drop on the losers stuck in the middle of the table, they were easily surrounded once I could deploy elsewhere.

So even when I got my second surprise drop deployment, which tipped the odds of getting the first turn firmly in my favour, I was already on the road to victory thanks to the good fortune of getting my first surprise drop so early in the deployment sequence.

Of course, my opponent decried my victory as simple luck. However, a man of my skill, Trump-esque ego and insecurity could never suffer such a slight – and so I hit the maths to determine exactly how much of it was down to luck, and how much could have been planned for after all.

The thing to note here is that from the perspective of out-manoeuvring the attacker, all that matters for the defender is how early in your deployment sequence your first surprise drop occurs (if you get one at all). We can then assume that any sane player will position that first surprise drop in such a way as to enable all remaining groups to deploy off them in safety, or good counter-ambushing positions.

So the question we need to ask is:

“For a gang which can be split up into N groups of two models, what is the probability my first surprise drop will be the first group, second group, third group etc.?”

Which then allows us to answer the question we (I) really care about:

“For a gang which can be split up into N groups of two models, what is the probability that I will get to deploy M of these groups away from the central group?”

The answer to this second question is given below for various gang sizes:

AmbushE_PDF.png


There are a couple of observations from this.
  1. Firstly, unless your gang is massive (22+ models, not shown), the single most likely outcome is that you don't get to deploy any groups away from the centre, i.e. the most likely value on the x-axis is zero.

  2. Secondly, and this is a really counter intuitive result, the second single most likely outcome after deploying no groups at all away from the centre is that you get to deploy all of your groups away from the centre instead! Crazy eh? I'm betting most people didn't realize that was the case.
But, really, we don't care so much about these single outcomes. If you tell me I don't get to deploy 5 groups away from the centre, but 6, of course I'd still be happy. So it's much more useful to ask the related question:

“For a gang which can be split up into N groups of two models, what is the probability that I will get to deploy at least M of these groups away from the central group?”

Which is answered by this figure...

AmbushE_CDF.png


So in the game I played, for my defending gang of 12 models exactly how lucky was I to deploy 4 or more groups away from the centre? Well, reading up from 4+ on the x-axis to look at the point marked by the box in this figure, there was about a 30% chance of that happening.

I wouldn't call those super long odds to be honest. Lucky? Probably – if you define “lucky” as defying 2:1 odds, then definitely. If you define lucky as defying 3:1 odds, then definitely not.

So, the moral of my tale:
  1. As defender: when positioning your surprise drops, plan accordingly to deploy all your remaining groups off those models – even if you don't get the first turn, you will easily out-manoeuvre the attacker.

  2. As attacker: ambushes can go sideways for the attacker very quickly. For many defending gangs, the odds are in favour of having at least two defender groups counter-ambushing your attackers. Plan accordingly!

  3. I was lucky – but not as lucky as I (and my opponent) thought. Still, every little helps. I'll leave observations on the implications for my playing skill to more humble Underhivers... :D
 

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
Don't get it right, get it written...

... as my old man used to say :D.

So, as I mentioned before I've been working towards some kind of scenario simulation tool thingy to assist with play-testing. Mainly for fun (nerd!), but also because it was useful when designing and calibrating some of the scenario tweaks for the latest LAGGNOG campaign we're running.

Basically, I'd code in all the various rules for each scenario, and then look at how the tool predicts attacker and defender fighter count probabilities turn-by-turn (e.g. how likely am I to have one attacker left in turn 3, or two attackers left etc.). Attackers get decimated too quickly? - just tweak the initial attacker numbers, or defender reinforcement mechanics, and rerun until the distributions of attacker and defender counts turn-by-turn look more reasonable.

And the advantage of doing that? Simple really - it estimates the outcome from hundreds of thousands of games to minimize the effect of anecdotal bias in the rules design process.

The plan was to write it up bit by bit as it's really an amalgamation of several different components, but @ineptmule has prodded me into putting it on the site now, so I thought I might as well plonk what I have written up in this thread.

It's too detailed (read: boring) to put in a post, so for the eager beaver reader I've attached a pdf. It's a bit stuffy and pretty OTT, and really aimed at no-one but the maths geeks out there, but it does the job :D.
 

Attachments

Last edited:

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
Combat Skills – Rules of Thumb

*** Feint here refers to the October 2016 NCE rules version - first fumble counts as a critical hit ***

Simple takeaways:
  • Combat skills are a force multiplier for fighters who can already hold their own in combat (duh).

  • They are significantly less useful on low WS/A fighters, for whom simply getting the charge is more important (which is bad news for Goliath, Escher and Cawdors gang leaders who take Combat skills on their Juves).
  • Deflect functions best as a defensive skill - the best performer when you are at a disadvantage.
  • Feint functions best as an offensive skill - the best performer when you are at an advantage.
  • Feint almost completely nullifies the downsides from taking a chain when it comes to win-loss-draw probabilities (not shown here though).
So I finally had chance to analyse all the combat skills and weapons (e.g. chain and sword) to rank them from best to worst.

Simple as that might sound, the lovely rich nonlinear game mechanics of Necromunda make it quite a complicated task. A hugely enjoyable complicated task, but a task nonetheless (I'm going to miss this system if N17 takes off :().

For this post I'll just stick to the rules of thumb from the analysis. In my next post I'll show how each skill and weapon impacts your win-loss-draw percentages. And more importantly (for combat nerds like me anyway), I'll introduce the concept of Gearing - this is the ratio of your big wins to your big losses and is really where the combat skills shine.

But enough, you say! (or, at least, I say to mysefl) To the rules of thumb!...

Rules of Thumb:

To keep things simple, the opponent here always has A2 – so either a shooting fighter who has A2 on their profile, or a dedicated melee armed fighter with A1 on their profile. The latter covers a large percentage of dedicated melee fighters in the game, so it's a decent baseline to work off.

Here's a little table which summarizes which skills are the best performers, depending on whether your WS is higher/lower than your opponent, and the number of attacks you have.

Candidates for the table? Deflect, Feint, Counter Attack, Charge!, Parry (just the one), Parried, Chain (with and without Feint), Knife only...

WS Lower than opponent
Number of Attacks 1-2: Deflect / Charge!
Number of Attacks 4+: Deflect / Feint

WS Equal to opponent
Number of Attacks 1-2: Deflect / Charge!
Number of Attacks 4+: Feint

WS Higher than opponent
Number of Attacks 1-2: Feint / Counter Attack
Number of Attacks 4+: Feint

Funny how the not-a-skill skill, that is simply charging in and getting +1WS, outperforms everything apart from Deflect when you have lower or equal WS and 1-2 attacks.

But when you start to get lots of attacks, already have a decent WS or are stuck in an ongoing fight, that's when the Combat table really starts to pay off.

Next up is the full analysis and the most important thing of all for combat skills – Gearing! What more could you want?...
 
Last edited:

thisisalie

Ganger
Feb 6, 2012
62
44
18
Agility table needs a nerf you say? Sprint/leap change WS by at least 2 due to not being charged. The merits of being the charger gets even more interesting if you have weapon choices and talents - e.g. flails/clubs/swords/great weapons, as if it's your turn you get to decide order of resolution for weapon declarations and talent uses. This makes being the charger really important for any counter attacking and feinting characters if they go against other character that can force re-rolls.

Example: I roll 6,3,1. My opponent parries my six resulting in 3,1,1. I have feint so one 1 is good, but that second I don't want, so I re-roll the one that has not been re-rolled. 4,3,1 - adding +5 to my score. If my opponent made me do re-rolls first, mathematically the best re-roll is the 3 as the 1 is adding +1 to my combat score. Lets keep it identical and say I get a 4, my opponent parries my 6 and I roll a 1. 4,1,1, - +4 to my combat score.

Example 2: I roll 4,4. My opponent has 2 valid parries. I have counter attack. If I chose resolution, he would declare first, as his best move is to not parry. I would then re-roll as it's worth +2/6 on average. If he chose resolution he would make me choose to roll first. By maths, I should re-roll as the odds are in my favour. Lets say I do, but I roll a 2 or 1. My opponent should now force me to re-roll the other. If I'd rolled a 3, it's neutral on average to force a re-roll, but initiative and other specifics may make it worthwhile - it's now my opponent's choice.

Example 3: Combat master - I will choose to go from weakest to strongest in fighting order. This is important twofold - 1) Me downing someone may cause panic. 2) The +1WS and +1A. If you chose order, you'd pick the opposite.

These scenarios may seem niche but there are quite a few niche scenarios that picking resolution order is very important for and they add up and only a few are mutually exclusive.
 
Last edited:
  • Like
Reactions: enyoss and spafe

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
You're absolutely right @thisisalie, interactions of combat skills between players does become pretty important late in a campaign.

I'm not sure it always comes down to who's turn it is though. For example, in my turn I might not be able to parry as my opponent's dice is lower than my own - so I'm forced to pass on the opportunity. My opponent might then use deflect on me to reroll my highest dice and if the reroll result is low I would then activate my parry.

I really don't have a feel for how big the advantage Is for the person who's turn it is, but having first refusal on using skills must give at least some advantage to them even if all else is equal.

I also really don't have the capacity to analyse it - it must be hopelessly complicated! I do like a challenge though :D.

You make a very good point on the double merit of charging versus being charged. My thought process though is that, often, no-one really plans on getting charged. It's what you do when your plans go suboptimally really (although obviously it happens a lot :D).

The reason I include charge along with the other skills though is that it can help me plan which of my fighters can take on other fighters as long as they get the charge. If my opponent wants to get someone into combat with me then, mostly, rather than calculating how much of an edge that would give them I prefer to introduce them to my heavy Bolter. Dakka. Dakka. :)

It's a good point though and definitely worth seeing how important it is relative to the other skills, so I'll include it in the analysis in my next post :). Any more suggestions or pointers are welcome!
 
Last edited:
  • Like
Reactions: spafe

thisisalie

Ganger
Feb 6, 2012
62
44
18
I've realized that current player declares which weapons he is using before the other player and that there is no resolution order for deflect/parries. I always thought current player decided the order that everything was going to resolve. In the above example I thought current player could go:

Player: Opponent, resolve any parries.
Opponent: No parries.
Player: Okay I now deflect you.

Otherwise you can end up with weird standoffs where the first person to act loses... I must've imagined it. Well you still get to choose fighter order in mass combat...
 
  • Like
Reactions: spafe and enyoss

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
Yeah, I put up an example a few pages back on the whole weapon choice and skill activation standoff. I also brought it up in the NCE thread around the same time, and I think that's where the 'current player selects weapon first' thingy came from (although that wasn't my suggestion, just the selected solution!).

It's definitely a disadvantage to have to declare first much of the time. At least for weapon choice. I really can't get my head around all the possible combinations for when it's good/bad for skill, but your example is definitely one of them! :)
 

thisisalie

Ganger
Feb 6, 2012
62
44
18
Knew I'd seen it somewhere. Kudos for that.

It's good to go last for talent activation and weapon declarations. The later you activate your talents, the more information you have and the better decisions you can make. I cannot think of any times when activating talents first is good for you. Deciding combat resolution order is always good.
 

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
Some in depth analysis going on here. So far above my head that it hurts just trying to get to grips with it.



With that said, many thanks for this. Running an Escher gang getting a run down on the best use if invaluable. Cheers!
Cheers matey :).

Now it's getting to the end I'm actually trying to streamline the results from here on. I've got a couple more posts nearly ready to put up which should hopefully have some (relatively) easy graphics for understanding how good all the skills are next to each other.

Then I'll do a dedicated post for Disarm, then one for skill combinations. Then done! :D
 

enyoss

Gang Hero
Tribe Council
Jul 19, 2015
1,115
1,690
163
SL
Combat Skills – The Definitive Guide

That's a pretty grandiose post title. But then, the post has some pretty grandiose content, if I do say so myself :D.

Here are side-by-side comparisons of all the win-lose-draw probabilities for most of the major skills, abilities (e.g. parry), and weapons (e.g. chain) you might come across.

We'll also introduce the concept of Offence-to-Defence (O2D) Gearing here as well. This is simply a ratio of the number of big wins you expect per big loss. Here, I've chosen big win to be an outcome where a fighter wins and scores 3 or more hits, and a big loss is an outcome where the same fighter loses and has 3 or more hits scored against them.

Our Escher protagonist is once again Fighter 1, and we view everything from her perspective. We'll be applying all of these skills and weapons to her, and varying her number of attack dice from one to five. Her opponent is Fighter 2, who always rolls two attack dice – i.e. a fighter of moderate close combat ability.

When devoid of skills or equipment, we assume that Fighter 1 has the same WS as Fighter 2 – we'll call this situation the BASELINE. Any skills or weapons will be added to this baseline, one at a time, and we get a corresponding change in win-loss-draw probability.

Most of the time the skills or weapons are actively employed by Fighter 1, e.g. PARRY, CHARGE!, or CHAIN (even if the 'active' employment of a chain here is just to stop her being parried, at the cost of fumble penalties).

We'll also consider two active situations for Fighter 2 - when they parry Fighter 1 (i.e. Fighter 1 is PARRIED), or when they are charging Fighter 1 (CHARGED).

So here are the comparisons:

ProbSummary_AllSkills_F1WSbonus0_A1-All_A2-2.png


Here, pink bars show the percent chance Fighter 1 has of winning, yellow bars the percent chance the combat is a draw, and blue bars show the percent chance Fighter 1 has of losing. For each situation, e.g. BASELINE, these percentages change as Fighter 1 rolls from one attack dice (left-most bar) to five attack dice (right-most bar).

Numbers in each coloured section of the bars give the win/draw/lose percentages rounded to the nearest whole number.

If Fighter 1 has higher initiative than her opponent the yellow bars all turn pink (i.e. she wins draws). If she has lower initiative than her opponent the yellow bars all turn blue (i.e. she loses all draws).

Now for O2D Gearing:

Off2DefGearing_AllSkills_F1WSbonus0_A1-All_A2-2.png


In the second plot, we see how O2D Gearing changes for the same situations. Colours here indicate increasing attack dice for Fighter 1 - dark green is just one attack dice, pink is five attack dice. For example, the score of 9.1 when Fighter 1 rolls five attack dice and has the Deflect skill, indicates that they will inflict a win with 3 or more hits on their opponent 9.1 times more often than they suffer a loss of 3 or more hits against themselves.

Key Observations:
  • As said way back on page-1 of this thread, it's pretty easy to see that for attack dice the single biggest jump in effectiveness is when you go from A1 to A2, both in terms of increasing how often you win and decreasing how often you lose. By A4-5 the additional benefits from the extra attack dice are minimal in terms of win-lose percentage.

  • However... this is less true for offensive-to-defensive gearing. For this metric, extra attacks retain their effectiveness in ramping up the number of big wins you get for each big loss.

  • Feint leads to crazy O2D Gearing ratios. With four attacks, which isn't that hard to get if you take ferocity skills, on average you will get 17(!) big wins per big loss. That's pretty nuts.

  • High initiative typically gains you a 15-20% increase in win rate through grabbing the draws. So high initiative really matters!

  • Using a Chain with Feint (and to a lesser extent Counter Attack) has almost no impact on your win-lose-draw percentages but it will bring your O2D Gearing ratio down to be comparable with the other skills for Feint (but lower than other skills for Counter Attack).
  • Parrying increases the chances you get a drawn combat, in addition to increasing your win probability. Good news if you have high initiative, not so good if you have low initiative.
  • Deflect decreases the chances you get a drawn combat. I have no idea why that works the way it does, given the result for Parry above :p.

  • A single additional point of WS, e.g. as demonstrated here through CHARGE!, is as good or better than most combat skills.

  • Combat skills should therefore really be seen as a way to exceed the statline limits on WS, rather than being a more powerful advance than WS in themselves.
So there you have it, the canonical guide to (most) combat skills. How the skills interact with one another is for another (shortly to arrive) post.

Astute readers will also notice that Combat Master and Disarm are missing.

The first can be inferred from plots like the above in a limited sense, although to do so properly we need more figures which account for Fighter 2 having A3 and above. I might get around to that one day, but until then take my word for it – Combat Master is mean when it comes to the bonus it gives. In the second round of combat against a Fighter 2 who normally rolls two attack dice, it swings the win probability for Fighter 1 from less than 25% without Combat Master to over 65% with it. Very nasty, if situational.

Disarm is a slightly different beast, and I will be covering that in another (shortly to arrive... as well) post. The effectiveness of that skill depends on a lot of things, namely the chances you pull it off and what you have to gain should you do so.

In the meantime though, the above should lift the lid on the majority of what the Combat table has to offer. Hopefully quite useful if you ever find yourself picking between skills or deciding the relative strengths of those combat fighters already on your roster. Enjoy :)
 
Last edited: