Campaign Refit, Resupply, Rest parameters

[PEB14]

@Ithikial_AU@Mr.X@WimO

(I summon you three guys, as you are the leading, active campaign designers on the forum…)

This weekend I performed some tests in order to understand the mysterious Refit, Resupply, Rest parameters of a campaign.

Using CMFI, I designed a 4-missions campaign featuring 1 company of Germans vs 1 company of Indians (without their light mortars), all core units. I recorded the headcount in each unit (down to squad level), both at the beginning and at the end of each mission. Not all units are featured in each scenario, except in the last one. I don't track ammo as I suppose that behaviour is similar to headcount's.

 

The first test I performed helped me understand a few things:

- parameters are applied at the BEGINNING of the corresponding mission. So parameters from Mission 1 will be applied before Mission 1 starts, parameters from Mission 2 will be applied before Mission 2 starts, etc. It seems obvious, but this is not what I thought… Anyway, as a corollary, beacause of that I see not reason why the parameters for the first mission of a campaign shall be different from 0…

- parameters are applied ONLY to the units taking part in the mission. Non participating, core units will NOT benefit from the parameters.

 

The quantitative results from this first test were unconclusive, so I performed a second test. Methodology is the same with the following differences:

- All units participate in Mission 1. They are all submitted to artillery bombardment in order to inflict significant casualties.

- Mission 2 involves 1 platoon from each side at start, another 1 platoon apearing as a reinforcement for each side. No combat action.

- Mission 3 is similar except that the starting and reinforcing platoon are different. No combat action either.

- Mission 4 involves all units.

 

The second campaign brings the following conclusion.

- In a given Mission, reinforcement units are treated the same as starting units.

- Refit is treated at the scale of the whole force of core units taking part in the mission, including reinforcement. This means that the whole force will get 20%, 50% or whatever Refit you've set. Percentage of headcount in platoon and squad differ very significantly, only the whole force percentage is close to the set one. Some units will get nil, other will get a lot. Hence it appears that Refit troops are distributed randomly into the sub-units.

- Percentage is based on the force size at the beginning of the mission. Let's imagine you're playing a company that started with 120 men and ended the first Mission with 20. If Refit parameter of the next mission the company is involved into is set to 20%, you'll get 4 Pixeltrüppen as Refit. If your company had less losses and ended the last mission with 45 men, you would get 9 Pixeltrüppen. Basically this means that if you suffered many losses in a campaign you'll get a double punishment as reinforcement will be lower that what you could expect…

- There is a saturation effect. If your Refit brings some of your units to full strength, you'll lose some reinforcements: as reinforcement are distributed somewhat randomly, it appears that some are lost if the random numbers exceed full-strength headcount.

It would be interested to test:

- what happens with units down to a very small headcount,

- what is the effect of a 100% Refit on forces whose headcount is below 50% of full strength: arithmetic says that a 100% Refit parameter should NOT bring the force back to full-strength in such a case…

 

If you have more information regarding the way these parameters work, or if you have ideas for additional tests, you are most welcome to share them here!

[Mr.X]

Hi PEB14,

first of all - you are commiting blasphemy - you have forgotten to mention @George MC  seriously, he is probably the most experienced and best designer of them all 

Personally, I´m afraid that I can hardly help you with your de facto scientific question. When I had made "Zawiya Uprising" for CMSF, I symply set all parameters in the script to very good or high - and there were never any problems...I think, the difficulty starts if you try to represent situations with low supply/refit possibilities...

For my pack of campaigns, I deliberately chose not to use these settings and instead chose simple linear campaign progressions. The reason is that I personally don´t like it when I ruin the whole campaign with one bad mission I play. And I almost never play any mission twice. 

So, I´m sorry for this weak answer  - I simply have no idea about the details of this system 

 

Best regards

Mr.X

[WimO]

@PEB14 Thank you for the research. Much appreciated. I will have to reread your results a few time to understand them better. A comprehensive example with each statement would be very helpful to make you points very clearly. I might do a similar experiment. The artillery bombardment test is a good idea and could be done without any other contact between opposing ground units.

I like to try to work with the 4Rs because but they are definitely not clear to work with, absent adequate manual instructions. Also the results need to be exaggerated upwards to compensate for the fact that the game inherently produces more casualties than in real life in many or most instances. That is to say, the troops don't know when to retire or keep their heads down properly.

For designers who do not like the 4Rs and wish to work without this mechanism, an easy alternative is just to work without a core units file or create one with units that will never be used in the campaign, and then create whatever units you want for each scenario individually, manipulating their strength percentage and fitness.

[PEB14]

@WimO

I can share the Excel file showing the results above, as well as the files I used to build the campaign (so that other people can reproduce the tests easily). I'll share a dropbox link later today.

4 hours ago, WimO said:

I like to try to work with the 4Rs because but they are definitely not clear to work with, absent adequate manual instructions.

Indeed, that's why I felt the need for this test...

 

4 hours ago, WimO said:

For designers who do not like the 4Rs and wish to work without this mechanism, an easy alternative is just to work without a core units file or create one with units that will never be used in the campaign, and then create whatever units you want for each scenario individually, manipulating their strength percentage and fitness.

The problem with this approach is, you lose one of the main dramatic resource of the campaign structure... Not using the 4R will deprive the campaign author of a way to reward or punish players depending of their performance...

[PEB14]

Here you can find the files related to the test above: Missions btt files, core file, campaign script and Excel result and analysis file:

https://www.dropbox.com/sh/a9ao9osxt0e4xel/AAAcsaZHt9zQEfPDj3AfRQppa?dl=0

[PEB14]

19 hours ago, Mr.X said:

Hi PEB14,

first of all - you are commiting blasphemy - you have forgotten to mention @George MC  seriously, he is probably the most experienced and best designer of them all 

You are correct. My deepest apologies to @George MC. May he forgive me for my failure…

Edited November 13, 2023 by PEB14

[kohlenklau]

Merci mon ami. Excellent work Pierre! Thank you for doing this experiment with report and discussion.

[WimO]

Stimulated by PEB14's work I also ran a mini campaign to try to determine exactly against which parameter the R values are calculated. I have attached a pdf summary of my exel table. In conclusion the following is a near exact result.

Replacement Value = (Core unit value - End of scenario value) X R%

Example One:

Number of soldiers in CORE Unit "A" = 100

End of scenario "Good Order" soldiers (i.e. not KIA, WIA or MIA) = 80

R value for "Refit" = 50

Calculation thus: (100-80)0.50 = 10

Again, the above is a close enough approximation. The real formula is very slightly more variable.

Example Two: If example number one seems too straight forward, what about when we are at about scenario four and the units have gone through a number of strength reducing combat?

Number of soldiers in CORE Unit "A" = 100

Number of soldiers still alive at start of scenario #4 = 65 good condition + 3 wounded = 68

End of scenario "Good Order" soldiers (i.e. not KIA, WIA or MIA) = 42

R value for "Refit" = 50

Calculation thus: (100-42)0.50 = 29

As a result, in the next scenario the unit could start off a wee bit stronger than it did in the current scenario, i.e., 29 replacements + 42 good order surivors + an undertermined number of wounded = >71 vs previous scenario's 68.

 

====

As has been noted previously by PEB14, myself and others, the 10 reinforcements are randomly distributed across the unit. Thus not all sections will receive exactly 50% replacements; some more, some less.

=========

Other observations:

1. Units will never be topped up in excess of their original Core unit value

2. Sections that are wiped out completely may reappear

3. Some of the casualties noted as WIA at the end of a scenario will die and become KIA between scenarios without you being notified.

4. Some (or all?) of the casualties noted as WIA at the end of a scenario that survive, appear in the next scenario as wounded (less than perfect "Fitness"?) in the next scenario.

------------

Unresolved: I have not been able to determine if the WIA solders who become KIA are included in the "R" calculation. It does not really matter to me as a scenario designer because what i have learned from the formula above is adquate for my design needs.

Hope that's reasonbly clear.

Good enough for my needs.

 

 

[PEB14]

Hi @WimO,

I must say I'm not sure that I understand your results.

8 minutes ago, WimO said:

Replacement Value = (Core unit value - End of scenario value) X R%

Example One:

Number of soldiers in CORE Unit "A" = 100

End of scenario "Good Order" soldiers (i.e. not KIA, WIA or MIA) = 80

R value for "Refit" = 50

Calculation thus: (100-80)0.50 = 10

Again, the above is a close enough approximation. The real formula is very slightly more variable.

You suggest that Replacement value (that is the heacount brought back by the Refit parameter) is proportional to the theoretical headcount of the unit?

I didn't check it yesterday, but I had the idea while you performed your own tests. And I disagree with your statement. The better approximation, according to my own data, is:

Replacement Value of Mission A = (Number of men in Core units at the beginning of Mission A) x R%

Example:

Number of soldiers in all CORE Units at the beginning of mission A = 100

R value for "Refit" in Mission A data = 50

Calculation thus: 100x0.50 = 50

 

8 minutes ago, WimO said:

4. Some (or all?) of the casualties noted as WIA at the end of a scenario that survive, appear in the next scenario as wounded (less than perfect "Fitness"?) in the next scenario.

Wow! You mean, not the lightly wounded casualties?

How were you able to track them?

I never considered it possible! I was pretty sure that both KIA and WIA were out of the game as casualties.

It seems to me (but I didn't check it thouroughly) that the lightly wounded Pixeltrüppen soldier on, keeping their "lightly wounded" status. One exception (that I check today): all lightly wounded Pixeltrüppen are replaced by soldiers in perfect condition when the Refit parameter is set at 100%.

 

More tests on the way!

[WimO]

I forgot to attach my pdf last note. Here it is.4Rs Test Case.pdf

The red numbers in the pdf chart are wounded soldiers carried forward. Nine such appeared at the start of scenario #2 eventhough at the end of scenario #1 the number of WIA reported was 18. This means that 9 died between scenarios. In some scenarios the number of wounded that appear at the start of a scenario sometimes exceed the number reported in the previous scenario. This means that some wounded troopers are carried foward through multiple scenarios. This suggests that there is much more computing going on behind the scenes than we can easily determine.

My scenario #1 uses the whole of the Core Units file.

And yes, I was suggesting that replacement percentages are being calculated by multiplying losses "to date relative to the original Core Units File value". That is what I understand my data as suggesting.

I am receptive to changing my understanding ...

Edited November 13, 2023 by WimO
additions

[PEB14]

Things get more complicated test after test…

17 minutes ago, WimO said:

The red numbers in the pdf chart are wounded soldiers carried forward. Nine such appeared at the start of scenario #2 eventhough at the end of scenario #1 the number of WIA reported was 18. This means that 9 died between scenarios.

I'm under the impression that you are confusing WIA soldiers, that are treated the same as KIA in CM terms (the only difference being that they're shown in the WIA tally at the end of the game), and the lightly wounded ones, that soldier on with a small red cross attached to their rifle icon. These lightly wounded soldier are NOT shown as WIA at the final roster of the game.

At least, this is the way that I have understood things.

Regarding your testing procedure: Are all headcount figures given at mission start or at mission end?

 

I made one additional, unconclusive test.

Red Force

Mission 1

• All core forces used: 93 men
• Final headcount: 25 men (including lightly wounded)
• Final headcount for units involved in Mission 2: 17 men (including 5 lightly wounded)
• Losses: 68 men
• Losses for for units involved in Mission 2: 49 men

Mission 2. Refit parameter set to 25

• 2/3 of core force used
• Initial headcount: 24
• Replacement headcount: 7 men

7 men out of 49 theoretically missing (your computation): 14%

7 men out of 17 end of Mission 1 headcount (my computation): 41%

Neither fit the 25% set parameter, yours is closer but misses the target…

The 5 lightly wounded men are still here in their original squads.

Blue Force

Mission 1

• All core forces used: 99 men
• Final headcount: 24 men (including lightly wounded)
• Final headcount for units involved in Mission 2: 20 men (including 11 lightly wounded)
• Losses: 75 men
• Losses for for units involved in Mission 2: 46 men

Mission 2. Refit parameter set to 75

• 2/3 of core force used
• Initial headcount: 56
• Replacement headcount: 36 men

36 men out of 66 theoretically missing (your computation): 78%

36 men out of 20 end of Mission 1 headcount (my computation): 180%

Your computation clearly fits the 75% Refit parameter.

Additionally 6 out of the 8 lightly wounded men (75%, exactly the Refit parameter) have been replaced.

 

In my previous test with Refit parameters in the 20 to 60% ranges my computation was a better fit...

So what should one conclude?!

 

Modding in Combat Mission is really a matter of masochism… Why the hell these parameters are not properly documented in the Game Manuel !?

[WimO]

@PEB14

Yes, I have assumed that lightly wounded soldiers are the same as more seriously wounded soldiers. I am not aware of anything in the manual to indicate that the WIA troops at the end of the scenario are exclusively serious wounds. They could well be as you say.

Yes, list of troops in each of my scenairos are the 'at start' numbers that were the result of each scenario after the first refitting at the 50% rate.

And yes, I agree with your final expression regarding lack of 'comprehensive' documentation to assist map creators and scenario and campaign authors.

[Vacillator]

1 hour ago, PEB14 said:

Modding in Combat Mission is really a matter of masochism… Why the hell these parameters are not properly documented in the Game Manuel !?

Pierre, never mind all of that annoying stuff.  Continue with your work on the campaign my friend .  Just make sure it's WW2 (so stick to CMFI in this case).  I'll test it if it helps?

[Centurian52]

I'm pretty certain that the WIA listed at the end of a scenario do not include lightly wounded pixeltruppen. The number of men listed as ok has always been equal to my unscathed plus lightly wounded troops. Some variable percentage of seriously wounded will be listed as KIA at the end of the scenario if they are not buddy aided, which is why you will count more light red bases than the end screen lists as WIA. The end screen will list fewer KIA and more WIA if you buddy aid all of your seriously wounded. From a game mechanics perspective there is no difference between seriously wounded and KIA apart from which category they are counted in on the end screen. A casualty is a casualty in the eyes of the scoring system and replenishment parameters. But I always try to buddy aid as many of my seriously wounded pixeltruppen as practicable anyway since, from a roleplay perspective, I want as many of them to survive as possible.

[WimO]

Final Scenario casualty data does not match GUI feedback!

At the end of one test scenario the final report indicates KIA = 21 and WIA = 9.

But the GUI shows the following when I click on the German sections: Casualty text with accompanying wound dot (or cross) = 24 (3 more than KIA reported) and Light wound dot (or cross) only = 18 (9 more than reported).

That means a complete disconnect between the overall report and the GUI feedback. That means that 3 of the "casualties" somehow got better as did 9 of the light wounds.

Anyway, that's the end of it for me.

 

[PEB14]

EUREKA !

@WimO

La nuit porte conseil, my friend!

I'm pretty sure I found the answer. BTW, the answer, although somewhat encrypted , is written in the manual (page 116):

Quote

The next section in the Battle Entry is the percentage chance an individual unit has of being completely replaced if lost, repaired if damaged (vehicle only), topped off with full ammo, and brought back to a fully rested state.

We both started from the misconception that the Refit percentage is referring to some headcount: either the theoretical headcount of the core force, or the number of missing men (your hypothesis), or the current number of men in the core force (my hypothesis).

But this is wrong.

As written in the manual quote above, the Refit parameters (and probably the 4 other as well) are "percentage chance", that is a PROBABILITY, for each "individual unit", to be replaced. I'm pretty sure that "individual unit" means just that, each INDIVIDUAL - either a Pixet soldier or a vehicle. Not a squad or a platoon or a core force.

So basically, at the beginning of each mission, the game rolls a die (1-100) for each and every missing soldier. If the result is below the Refit percentage, the soldier is replaced. Otherwise it is not. And it's probably the same for vehicle repair and ammo (clips) as well.

Statistically, it means that @WimO approximation is correct. The MEAN replacement value is:

MEAN Replacement Value = (Core unit value - End of scenario value) X Refit%

The higher the number of missing soldiers, the closer to the Mean Replacement Value you'll get.

On the other hand, if the missing heacount is too low, statistical aberrations will become more sensitive (hence, probably, my weird result shown in first post):

• Example 1: if you miss 100 soldiers at the beginning of a mission, whose Refit parameter is 60%, you're likely to get 60 men back. But there is a very significant probability that you get only 58 or, if you're lucky, 63. In all cases, the ACTUAL refit you get will be either 58% or 63%, which is still close to 60%
• Example 2: if you miss 5 soldiers at the beginning of a mission, whose Refit parameter is 60%, you're likely to get 3 men back. But there is a non negligible probability that you get only 2 or, if you're lucky, 4. In the first case, the ACTUAL refit you get is only 40%, while it's 80% in the latter case.

And I'm pretty sure that the lightly wounded soldiers are treated along the same lines (no pun intented...).

While as a designer I'd really prefer to know how many men I give back to the player as Refit, from a programming point of view the probability approach is clearly much more simple to handle, if only because you don't have to manage the distribution of Refit troops among the different sub-units: the computer only checks wether a soldier is replaced or not. As simple as that!

Well, that's it

And as @WimO says, that's the end of it for me!

 

I hope that this post will be of use to some people at least...

 

[PEB14]

8 hours ago, Vacillator said:

Pierre, never mind all of that annoying stuff.  Continue with your work on the campaign my friend .  Just make sure it's WW2 (so stick to CMFI in this case).  I'll test it if it helps?

Thank you.

Don't worry, I'll soldier on. And thank you for the playtest offer, I file the proposal for future use!

But you must be patient, I don't expect to be able to offer something for playtest before several monthes...

And FYI, this campaign will be CMBN, not CMFI. I have ideas for CMFI (in particular, a red vs. red campaign...), but this is for an even more distant future...

[WimO]

Great spotting that manual entry Piere. Very helpful!