Combat Bonus tidbit

[SurlyBen]

I haven't ever seen any estimates on what exactly the double lighting bolt means, exactly, so I decided to try and figure it out. So far my results are far from conclusive, but I have enough info for a few ball park estimates (for regular troops, at least). (And, since that was really all I was hoping for, I may stop here, but I'd be happy to email my test map to anyone who asks)

Anyway, without further ado, my very preliminary estimate of what a combat bonus does to regular troops is this: A plus 1 combat bonus gives a 10-15% increase in firepower, and a plus 2 gives somewhere between that and a 50%(!) increase.

The test I used to arrive at these numbers was this: I took a single regular axis rifle 44 squad and had it shoot at a single low ammo regular british rifle squad at a range of 100 meters for one minute. The brit squad was in the open, the axis squad got a foxhole to minimize the effect of return fire from the brit. The british squad was under command of a plus 2 morale HQ. The axis squad was under command of an HQ with no combat bonus, plus 1 combat bonus, or plus 2 combat bonus. Other than the combat bonuses for axis and morale for british, the HQs were normal. I ran 50 tests for each combat bonus.

I am assuming a linear relationship between firepower and ability to cause casualties (this may not actually be true. If anyone knows, by all means let me know)

Unfortunately after I ran the test I noticed that for no combat bonus and plus one combat bonus, I was getting more zero kill results than I would have liked, which I'm afraid was skewing my results a bit (hence the (high in my opniion) 50% number for a plus 2 bonus...) Ideally the test should be run at a range where you don't get very many no kill results and you also don't get many kill the entire squad results. The reason I ran it at such a long range was to prevent the brits from panicking and running during the test (man, that sounds evil), but I figure I could get away with running it at 60-80 meters and get better results.

Anyway, the numbers were:

No combat bonus average # killed 1.38

+1 combat bonus average # killed 1.54

+2 combat bonus average # killed 2.1

Surlyben

[russellmz]

you'll be hanged for this at the warcrimes tribunal!

hmmmmm...maybe if you try elite snipers and FOs with 0 ammo.

did u do this with 50 squads lined up separated by cliffs so that they can't kill anyone other than "their" target squad?

50% does sound high, but how much better does an elite or crack unit do than a reg? i heard a bonus raises you to the next level of experience for that trait (combat, stealth, etc). so the reg was at elite level maybe.

[SurlyBen]

Yeah, I have a bunch of rows separated by cliffs. The command bonus gives units delays and command radiuses that are the same as units of higher experience. So a plus one command bonus on a regular HQ gives a regular squad the same command delay as a vet unit. I'm hoping that something similar happens with combat bonuses, since then I'll be able to use higher experience squads to check my numbers.

I should be able to run a similar test for the morale modifier, assuming I get everything worked out.

One thing I noticed is that infantry ROF seems to be fairly constant regardless of command bonus (5-6 shots per minute).

[Clubfoot]

If you really want your targets to stay still, place them on a piece of open ground in the 3D editor. Then return to the 2D, top-down map editor and place a water tile where your target/soldiers are.

They'll be forced to stand there, knee-deep in water, unable to move, and sponge up the punishment you dish out.

Place your firing units in thick woods, w/ foxholes, behind a wall to minimize reciprocal damage.

[SurlyBen]

Hmm. I may try that tomorrow. I just ran my test again. This time at 80 meters. Unfortunately I don't seem to be getting the kind of results I would call consistent.

For axis rifle '44 against american rifle 44 for one minute at 80m (I tried 90 times)

+2 combat bonus: Avg kills=4.26 StDev=2.0

+1 combat bonus: Avg kills=4.09 StDev=1.9

+0 combat bonus: Avg kills=4.16 StDev=1.9

As you can see from the standard deviation, there is a lot of variation in how much gets killed by any given squad. Over the 90 tests I did, the no combat bonus squads actually outperformed the +1 combat bonus squads. I also ran the test with no bonus vets and crack troops with similar results. (the vets did the worst of anyone)

Now this could be a result of my having a test which isn't very focused. A few things I could do to tighten it up have already been suggested, and I can think of a few more. But I doubt that tightening up the parameters of the test would have that much of an impact on the results. My guess is that to get meaningful results the tests would have to be run a lot more times (on the order of 1000 at least) and that is really more effort than I am willing to put into it. If anyone else has a lot of spare time, I've got some maps all set up though...

[edit: I suppose another possibility could be that my assumptions are just plain wrong... ]

[ 10-08-2001: Message edited by: SurlyBen ]

[Vanir Ausf B]

Interesting.

Vet troops receive a FP bonus that can be seen with the target or LOS command. Troops under the command of a hq with a combat bonus do not. It has been suggested that the combat bonus causes troops to fire more frequently, but some quick tests I have done do not support that. So, I really don't know what it does exactly or how it does it.

[SurlyBen]

One thing that might produce clearer results would be trying the test with conscripts. The command bonus differences are the most pronounced on conscript HQs... Rof differences (if any) would be more pronounced there too. The difficulty would be in minimizing the results of return fire (what with them being conscripts and all)

[Galatine]

Interesting.

Perhaps doing a multi-variate analysis adding additional ranges (and so varying FP's) to check for a non-linear function of firepower to ratio of casualties inflicted?

It does seem that FP's of 100 or less tend to supress rather than cause causualties. It's possible that two seperate but related results are at work; suppression and caualities.

How about setting conditons for three ranges and taking data related to: 1)number of caualties inflicted per minutes 2)Time until suppression break points reached.

Then you could run a analysis on a statistical program like minitab or SPSS to test for relationships.

It would also be interesting to see if the suppress and FP are confounding variables. I'd think that as supression rises cover would be taken and casualties would drop. Then when higher moral breakpoints are reached panic sets in and cover would be abandoned in favour of retreat and causalites would rise. A bell curve.

If you set-up the test conditions and posted them on the board, anyone who wanted to particpate could use the exact same set up and experimental controls. We could then pool our results and do a meta-analysis on the collected data and shrink experimental error and variablity.

But I've only taken 2-3 courses in Statisics for Psychology, I'm really out of my depths. Any Math wizards's out there?

[WineCape]

Very interesting study, Ben.

These combat bonusses are only cryptically noted in the CMBO handbook. Apart from a general "it-must-be-good-to-have-HQ-units-with +2 bonusses" feeling, nobody has really sat down and tried to work out some sort of %'s (a very rough estimate will do IMO!)

I would also like to see the particular effects of the zap (combat) bonus on AT-guns in particular. Keep 'em case studies rolling.

Kind regards and thanks for the Chateaux maps!

Regards,

Charl Theron

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

"During one of my treks through Afghanistan, we lost our corkscrew. We were

compelled to live on food and water for several days."

-- Cuthbert J. Twillie (W.C.Fields) in "My Little Chickadee" (1940)

[Wreck]

You might try using HMGs as sources of fire, with SMG infantry as targets. That would solve the return fire problem. Use the kind of HMG which gets massive ammo -- vickers. Put the targets at 250m or so, and there will be no return fire. Actually I suspect that beyond 150m vanilla squads of any kind will not return fire (assuming the source has good cover, of course.) You want to use the largest targets possible, so as to use the most sources (see below). So probably the best units to use for targets (with appropriate tinkering) would be the 13-man assault squads the Germans have.

BTW I had never thought to use elevation as a compartmentalizer for testing... I always use hedgerows.

Regarding the test itself: I think it would be better to put the target into good cover and then vent a lot of ammo, rather than using less ammo with worse cover. In fact I would suggest using two or more HMGs per target with the target in stone buildings. You want all firing guns to run down to zero ammo without killing the last man in the target.