MATHS: Modelling some proposed changes to Close Combat Part 1

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
Right as promised here are the results of some of the proposed changes to close combat.

These have all been applied to the scenario of an Escher Juve equipped with a Stilletto Knife and an Auto Pistol charging a Goliath Leader, equipped with furnace plate.

Note I've only considered the weapons and wargear relevant to the attacker and defender respectively, nor have I modelled the leaders reaction attacks should he survive. This purely focuses on the percentage chance of the juve taking the leader OOA with her attacks.

This has been chosen as a highlighted example of a low value fighter charging a high value fighter who has a better WS.

This is not ment to be the definative case (there are others I want to model) but is I think a useful illustrative case.

The scenarios are as follows.

1a (baseline case), current N20 CC rules.

1b, proposed change, defending fighters with a higher WS than the attacker gain 1 parry.

1c, as 1b, but parry is changed to a WS test on the defenders WS to ignore one hit.

1d Escher juve hitting on 5+ rather than 4+ (simulating both a -1 to hit modifier (such as fighting over a barricade) or the application of an opposed WS table, where the juve would be hitting on a 5+ as a result of that table).

1e The goliath leader has an armoured undersuit, thus his save increases to 4+.


RESULTS

Scenario 1a
21.78% chance of the Goliath Leader going OOA.

This was a little surprising to me as I had thought the odds of this happening would be a little higher (nearer about 25%), I don't think the odds as they stand show quite as bad a situation as some yaktribers thought it was.

Overall this shows that there is a decent chance of taking out a better, and tougher melee combatant. Just over 1/5th of the time, but most of the time your juve will try and then get splattered for her audacity.

Scenario 1b
17.09% chance of the Goliath Leader going OOA.

So this is an improvement in survivability of 4.67%. Which is pretty nice.

Parry is nice when its forcing a medium WS fighter to re roll a hit.

Also shows that a weapon with parry is a reasonable investment for a goliath, as both the Chainsword and chain axe are good weapons in the hands of a Goliath that cost a reasonable price.

Note the programme would always part toxin hits if it could in preference over autopistol hits because they are the more deadly.

Scenario 1c
15.40% chance of the Goliath Leader going OOA.

So this is an improvement in survivability of 6.36%. Which is getting quite good.

Making parry a WS test for the defender has a greater benefit here, but would be worse for a defending fighter with a WS less than the attacker.

Scenario 1d
14.73% chance of the Goliath Leader going OOA.

So this is an improvement in survivability of 7.03%. Which is best of all the scenarios. The effective -1 to hit the defender is a big deal.

Scenario 1e
16.28% chance of the Goliath Leader going OOA.

So this is an improvement in survivability of 5.48%. Which is getting quite good.

The extra point of armour works well in this scenario.


Conclusion

Obviously I'm interested in others thoughts on these numbers but my own take.

1) The odds as they stand right now seem pretty reasonable just under 4 out of 5 times your Juve will fail.

2) Higher WS defending fighters gaining a parry (under the current N20 parry rules) gives the lowest increase in survivability, at just under 5%, or about 6.4% increase if you change to the alternative parry method.
These are good because parry has counters.

3) to hit modifers if situational (e.g. defended obstacle) have the strongest modifier at about 7% more survivability in this case. An opposed WS table potentially apply this all the time might be too much? As the defender is surviving more than 5 out of 6 attempts.

4) getting more armour is unsurprisingly worth it in this case.


Here is my results spreadsheet.


I'm really interested to see how these changes work to the fight between to elite combatants next.
 
  • Like
Reactions: Mr. M

Casualty

Ganger
Dec 14, 2016
77
83
23
Portland OR USA
Ex cellent. This provides super clarity. Gracias, Merci !

I know this isn't an all request radio station, but ...before you reset the scenario.....

Please run 1d again with a -2 modifier (hitting on 6's). This new test would be 1g. (This would show the cumulative effect of stacked modifiers.)

Please run a series where the Juve attacks the Goliath leader with one less attack. This wil be 1h. (Number of incoming attacks is another very easy variable that we could easily control in the rules.)

And because I am super pushy..... and no one around here has set good boundaries....include these new tests when the two elite combatants fight as well.

Could you run all these tests where the Goliath leader attacks the juve?


THANK YOU. I KNOW YOU HATE ME. I'M SORRY.
 
Last edited:

Casualty

Ganger
Dec 14, 2016
77
83
23
Portland OR USA
Could you run the tests where 2 identical juves fight?

And then identical Goliath leaders?



Something is wrong with me.

I took too much math in university. My parents were sadists.

I AM REQUIRED TO FORM A FULL 3D STATISTICAL MODEL BEFORE I CAN MAKE ANY DECISION.

MODS!!!! BAN ME.
 

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
I can look into those, it does take sometime.

I'm gonna pass on the goliath leader hitting back, he will (with any reasonable CCW loadout have a very good chance of killing the juve. I'll concentrate on the other scenarios listed first as I think they provide a better insight into the suggested problem areas.
 

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
Please run 1d again with a -2 modifier (hitting on 6's). This new test would be 1g. (This would show the cumulative effect of stacked modifiers.)

Easy enough.

Please run a series where the Juve attacks the Goliath leader with one less attack. This wil be 1h. (Number of incoming attacks is another very easy variable that we could easily control in the rules.)

Again easy enough but as a 1 attack on profile model, reducing the juves attack would involve removing the extra attack for charging or the extra attack (with the pistol) for 2 cc weapons, and I think that would be a controversial and unnecessary change to combat. I'll model it anyway (one with no pistol attack, on with no second stiletto knife attack).
 

Casualty

Ganger
Dec 14, 2016
77
83
23
Portland OR USA
I think the majority of the feedback and concern I 've read from players indicates that most folks would like to reduce the odds of all charges resulting in OOA results. Those same folks therefore would like to see an increase in the odds that there will be reaction attacks.

Your first round of math illustrates that simple modifiers can significantly push close combat in those directions.

To put it in perspective, that juve , in the current rules only takes out that leader about 21 percent of the time. With a simple -1 to hit modifier, you reduce that to about 14 percent.

Most folks will read what I just typed, and think " big deal, 7 percent isn't much." But the reality is that we're not comparing percents...we're comparing relative proportions. We're comparing 21 percent in proportion to 14 percent. You've actually increased the leader's chances of surviving by 33 percent.

Huge.
 
Last edited:
  • Like
Reactions: Al_Weeks

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
Ok I quickly did some of the other scenarios.

1f -2 to hit penalty.

7.81% chance of OOA.

This is as expects a further 7% less chance of OOA over the minus 1 to hit penalty.


1g one less toxin attack

12.3% OOA chance

1h no auto pistol attack

21.07% chance of OOA

A 0.69% chance in survivability.

Unspurisingly the auto pistol adds very little in this scenario because it has to get at least 2 hits on the firepower dice to have a chance of doing the 2 unsaved wounds required for an injury dice to be rolled.
 

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
I think the majority of the feedback and concern I 've read from players indicates that most folks would like to reduce the odds of all charges resulting in OOA results. Those same folks therefore would like to see an increase in the odds that there will be reaction attacks.

Your first round of math illustrates that simple modifiers can significantly push close combat in those directions.

To put it in perspective, that juve , in the current rules only takes out that leader about 21 percent of the time. With a simple -1 to hit modifier, you reduce that to about 14 percent.

Most folks will read what I just typed, and think " big deal, 7 percent isn't much." But the reality is that we're not comparing percents...we're comparing relative proportions. We're comparing 21 percent in proportion to 14 percent. You've actually increased the leader's chances of surviving by 33 percent.

Huge.
So certainly in this case my gut feeling is that the odds of the leader going OOA under the current combat system are kind if OK as is in this situation.

Modifers to hit for good positioning (e.g. defending a barricade feel acceptable to me, as tactical placement should be rewarded in game)

Think it also shows how underrated parry is by people against medium melee threats.


However I need to do the other scenario of two combat elites charging each other.

As it think that is where the real issue lies (I suspect you'll get odds of 60%+ of taking your opposing elite combatant out of action on the charge and I'd be comfortable with those coming down a bit.
 
  • Like
Reactions: Orngog

JawRippa

Gang Hero
Mar 31, 2017
1,069
1,380
133
Saint-Petersburg, Russia
I think that realistically toxin juves missiles are supposed to be used when assisted in close combat by a fellow ganger (who charged previously, but didn't fight as a part of charge to avoid triggering reaction attacks). So really it should be +1WS, which would bring chances of a kill up a decent notch. Overall I agree that juves with toxin are not the worst offender - duels of champion vs champion is, since whoever charges first wins.

Could you take an extreme cases of a WS2+ 3A base, +1A for dual weapons champions with 2W and 2D and have a mirror match where one of them charges another? What are the chances that defender will get to react at all in these cases? And then do the same when they have paired weapons. I'm pretty sure than with alternative attacks we would see a meaningful change when -1 modifier will slightly bring down the lethality, but not enough.
 

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
Could you take an extreme cases of a 3A base, +1A for dual weapons champions and have a mirror match where one of them charges another? What are the chances that defender will get to react at all in these cases? And then do the same when they have paired weapons. I'm pretty sure than with alternative attacks we would see a meaningful change when -1 modifier will bring down the lethality, but not enough.

Working on the Death Maiden Charging a stimmer and Visa versa right now. See the close combat thread, I'd love to know what you think the odds are from a gut feeling.

The juve with an assist will take the numbers up to about 28-9% chance of an OOA on the charge, which again doesnt feel too bad considering you've had to work to apply that assist in the form of another fighters actions.
 

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
I actually think the worst offender in terms of cost vs leathality might be a boneing sword armed corpse grinder ganger.
 

JawRippa

Gang Hero
Mar 31, 2017
1,069
1,380
133
Saint-Petersburg, Russia
Working on the Death Maiden Charging a stimmer and Visa versa right now. See the close combat thread, I'd love to know what you think the odds are from a gut feeling.

The juve with an assist will take the numbers up to about 28-9% chance of an OOA on the charge, which again doesnt feel too bad considering you've had to work to apply that assist in the form of another fighters actions.
From gut feeling, they both are lethal as hell. Probably enough for retaliation attacks to never occur in 60% of charges, no matter who charges first.
 

Casualty

Ganger
Dec 14, 2016
77
83
23
Portland OR USA
The stacking of negative "to hit" modifiers (ie: -1 to hit and then -2 to hit) is showing a lot of promise. If we get a range on the 2 fighting elites of somewhere between 5 to 10% OOA reduction per negative modifier...we're probably golden. The elites test will prove whether it has real feet or not.

Fingers crossed.

I said probably....
 
  • Like
Reactions: Ledward

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
Ok so I ran all the scenarios for the death maiden attacking the stimmer.


TL;DR

They are deadly as expected, but there are appreciable boosts to survivability for each scenario that gives a different boost to survivability.

The to hit modifiers (or opposed weapon skill test which is the same as a -2 to hit) make a decent difference in the survivability of the stimmer, but genesmithing him to T5 is probably the best survivability increase (especially for its measly 10 cred cost)

Baseline Scenario:
Death Maiden, 2 Stilleto Swords charges Stimmer with furnace plate.

79.3% chance of OOA

Brutal, and I'd agree this is in the too much of a sure thing scenario.

Scenario 1b
A -1 to hit penalty

70.73% chance of OOA, an 8.6% reduction in leathality

Not bad, but still quite very likely OOA.

Scenario 1c
-2 to hit, also the equivalent of opposed WS table result of hitting on a 4+

58.5% chance of OOA.

This is getting more like it, 60% chance is what I said it felt like the lower end to me the death maidens chances on a charge should be like.

Scenario 1d

Armoured undersuit for goliaths (a 4+ base save(

70.44% chance of OOA, extra armour gives and appreciative boost.

Scenario 1e
Genesmithed to T5

62.85 chance of OOA, this is a 16.44% reduction in leathality. It's worth it by a long way.

Scenario 1f
Chemsynth added to death maiden

89.37% chance OOA
If she procs it she gains a 10% increase in leathality.

I need to mod the code to test for if she activates it, so this will probably come down to I think around 86% chance OOA overall.


Summary.
Between some modifiers (or opposed WS tables) and modest extra investments in protective gear the stimmer can weather the storm quite well.

I'm looking forward to doing the reverse and having the stimmer attack the death maiden. See if those 2 parry's help much.(especially if we used the proposed parry is a WS test for the defender to cancel the hit

I suspect the stimmer will be a fair more lethal, but we already know he's less likely to charge so that feels ok
 

JawRippa

Gang Hero
Mar 31, 2017
1,069
1,380
133
Saint-Petersburg, Russia
We've established the fact that a melee monster charging another melee monster would result in no retaliation attacks most of the time.

What about alternating attacks instead of modifiers? Even without calculation I can predict that we will get retaliatory attacks a lot more often.
 

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
We've established the fact that a melee monster charging another melee monster would result in no retaliation attacks most of the time.

What about alternating attacks instead of modifiers? Even without calculation I can predict that we will get retaliatory attacks a lot more often.

It's a complex one to model, but yes you will get alternating attack more often becuase your only rolling 1 or 2 dice at a time before your opponent strikes with at least one dice.

If you can describe the scenario you want modeled e.g. exactly how you want attacks to alternate I can look into it.

I've got a bit less time to do modelling atm, so I've slowed up a bit, but I will look into a lot of the listed scenarios over time.
 
  • Like
Reactions: JawRippa

Kiro The Avenger

Gang Hero
Apr 4, 2018
1,224
1,588
128
Bristol, UK
For alternating attacks I'd like something like Death Maiden w/needle pistol and claw Vs stimmer with paired pulverisers.
Followed by the same death maiden v Stimmer with renderiser, resolved as if ADADADAAA.

That'll at least be a decent starter, contrasting a comparatively weaker but more attacks pulveriser Vs the hard hitting but fewer attacks renderiser. Plus toxin.
 

Al_Weeks

Gang Hero
Honored Tribesman
Dec 22, 2014
513
533
123
Bristol
For alternating attacks I'd like something like Death Maiden w/needle pistol and claw Vs stimmer with paired pulverisers.
Followed by the same death maiden v Stimmer with renderiser, resolved as if ADADADAAA.

That'll at least be a decent starter, contrasting a comparatively weaker but more attacks pulveriser Vs the hard hitting but fewer attacks renderiser. Plus toxin.

I'd argue for the inclusion of a stiletto sword two becuase I think parry is of value simulating in an alternating attacks world.