Analysis of RFCs by defenders set to Ignore Losses (Full Version)

All Forums >> [Current Games From Matrix.] >> [World War II] >> Norm Koger's The Operational Art Of War III



Message


Curtis Lemay -> Analysis of RFCs by defenders set to Ignore Losses (3/7/2012 9:44:02 PM)

I’ve finally had some time to do some testing on RFCs by Ignore Losses defenders. To summarize, what I’ve found is that there is a problem that warrants addressing. But, it’s not a problem that was introduced into 3.4. Rather, it’s a problem that has been with TOAW from the beginning. The changes in 3.4 just exposed it.

My tests prior to the 3.4 release had been done with defenders at Limit Losses. The units would retreat if losses exceeded a certain amount. These tests were conducted with defenders set to Ignore losses. Such units are oblivious to losses and only retreat if they fail a morale check.

My tests show that the 3.4 adjustments are working just as designed. Fortified units (as well as units in Fortified terrain) are about 6.3 times as hard to dislodge as mobile units in clear terrain – all else being equal. Entrenched units (as well as units in Dense Urban or Badlands) are about 2.9 times as hard to dislodge as mobile units in clear terrain. Units in Bocage, Mountain, Dunes, or Urban are 2 times as hard to dislodge as mobile units in clear terrain. Defending Units (as well as units in Hills, Forest, or Wadi are 1.35 times as hard to dislodge as mobile units in clear terrain. I still don’t think those figures are inappropriate. They could be adjusted a little, but I don’t think they are out of line.

The problem is that the only other factor that determines whether the unit retreats or not is the morale of the defender and attacker. (Remember that morale is a combination of the unit proficiency, readiness, and supply.) The higher the morale of the defender, the harder it is to dislodge. Attacker morale affects this in that if the attackers all drop out, the defender doesn’t have to make a RFC check. So, the lower the attacker morale, the harder it is for them to cause the defender to RFC. This makes it particularly hard for a low proficiency force to make progress against a high proficiency force.

The problem is that the combat odds make no difference. It must have been that way from the beginning of TOAW. We did make the odds affect RBC chances in 3.4, but not RFCs. This needs to be addressed. I’ll get to that later.

Before I show the test results, let me describe the test scenario. There are 13 terrains being tested and three deployments, for a total of 16 tests. The MRPB is 1, so no combat can last more than one round. Each test has 10 trials – for a total of 160 combats. Each combat consists of one attacking unit vs. one defending unit. The attacker is always 9000 infantry squads. The defender is initially 3000 infantry squads (x3 odds). Both sides are initially 70 proficiency and 100 readiness and supply. The scenario is then run 10 times for a total of 100 trials for each test. After the combats complete, I total how many defenders RFCd in each test.

In the next test, the defender has only 900 squads (x10 odds). In the next test, the defender has only 300 squads (x30 odds). In the next test, the defender has 3000 squads again, but has 1% supply and 33% readiness. In the final test, the defender has 3000 squads with full supply and readiness, but the attacker has 100 proficiency.




Curtis Lemay -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/7/2012 9:44:50 PM)

The attached table shows the results of the first test. To recap, the combat odds are 3:1, and the units all have 70 prof and 100 supply/readiness. The results show that the reductions in RFCs are about as designed (allowing some statistical variation). The somewhat lower results in this first test were probably due to the increased losses the 3000 defending squads could inflict – causing more attackers to go into reorganization before even making their morale check, thereby increasing the number of defenders that didn’t have to make that check.

[image]local://upfiles/14086/54B1DC8FC9C54FCD84699FAAA9A5880F.gif[/image]




Curtis Lemay -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/7/2012 9:45:28 PM)

The attached table shows the results of the second test. To recap, this changes the odds to 10:1. The results are not much different from the first test. This means that there was negligible benefit of the greater odds.

[image]local://upfiles/14086/710B7CA45ABA4788AA441E8E5741C15E.gif[/image]




Curtis Lemay -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/7/2012 9:46:05 PM)

The attached table shows the results of the third test. To recap, this changes the odds to 30:1. Again, there is very little benefit of the greater odds. Basically, the chance of RFC in clear terrain is about = (AP) x (1-DP) = 0.7 x 0.3 = 0.21. In other words, 100 trials will average about 21 RFCs in clear terrain – and that’s about what the results show (19+26+16=61; 61/3 = 20.33).

[image]local://upfiles/14086/B7F3AE4463054E148AB27D95329511FB.gif[/image]




Curtis Lemay -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/7/2012 9:46:41 PM)

The attached table shows the results of the fourth test. In this test, the first test was changed so that the defenders have 1% supply and 33% readiness. This knocks the odds up a bit – but we’ve already shown that odds don’t really make any difference. The results show that RFC chances about doubled. So supply and readiness do affect RFC chances.

[image]local://upfiles/14086/FEB5EB9CF05F4DFD8EEA11724A2B7353.gif[/image]




Curtis Lemay -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/7/2012 9:47:59 PM)

The attached table shows the results of the fifth test. In this test, the first test was changed so that the attackers all have 100 proficiencies. This also knocks the odds up to about 4:1, but that is still irrelevant. But note that the results show that the RFC chances are up about 30% - the amount of the increase in attacker prof. That seems to show that the attackers have to pass their check before the defenders are required to pass theirs.

[image]local://upfiles/14086/D2A53B79690145A4AC87C8CE6B5EC07F.gif[/image]




Curtis Lemay -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/7/2012 9:48:38 PM)

The final table is a sum of all the other test results. This gives a total of 500 trials of each combat. That reduces the statistical variation enough that it can be clearly seen that the reduction levels closely track the desired target levels (pay particular attention to the “Reduction” vs. “Target” numbers). This correlation is why I’m sure the 3.4 adjustments are working as designed. Note that the “Factor” is just the inversion of the “Target”.

[image]local://upfiles/14086/C5E190B8E67F4CB3AE9DCDBC81392B05.gif[/image]




Curtis Lemay -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/7/2012 9:49:10 PM)

Note that these results suggest tactics to use to maximize RFC chances, as things now stand:

1. Make sure that at least one attacker passes its check. That is best done by increasing attacker loss tolerance, increasing the number of individual units assigned to the assault, and/or using high morale units.
2. Try to burn the defender’s supply and readiness.
3. Try to make sure that all 10 combat rounds are executed against the defense. Each one gives a chance to RFC, even if they are all in the same combat phase.
4. Don’t forget RBCs. Odds matter in those, so don’t overlook a chance to do so. Keep as powerful an overrunning unit available as possible. Remember to use it first after the combat completes.

But odds should matter in RFCs. The chances to RFC should be modulated by the combat odds. That’s going to take a code change. I’d suggest that 3:1 odds be “neutral” and would produce the results shown above. But if the odds were 6:1, then the chances should double from above. If they were 1.5:1 then they should halve, etc. Then if a Fortified defender was attacked at 19:1 odds, it should retreat as often as a defender in clear terrain attacked at 3:1 odds. Once this is done, RFCs will be working much better than before 3.4: Terrain will matter and so will odds.

I want to add that there may be other factors involved – it’s just too complicated and too hard to test to know for sure. Based on my experience, I’m pretty sure that units made of passive equipment are easy to RFC, for example. And it seems that a stack of defenders can often be easily cleared of all but the last unit. But I’m sure that the tests show that combat odds don’t seem to make any difference, and that needs to be addressed.




Oberst_Klink -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/7/2012 11:03:20 PM)

So... there was/is an issue with the 'ignore losses' thingie after all. However, minor adjustments to the MRPB and AD can 'ease the pain' of it; as you suggested. Thanks though for the thorough report. I am sure a few of the TOAW community thought ye'ole Oberst and the 'Alter Schwede' aka Rob were paranoid when we brought the issue up.

Thanks again Bob!

Klink, Oberst




r6kunz -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/8/2012 8:23:58 AM)

Gen Lamay
Quite a bit of research here. Nice work. And a very practical conclusion to all of this. Of course, the real result will be in the next update.
Thanks again for all of you hard work!
rak




Sensei.Tokugawa -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/8/2012 8:54:58 AM)

The way I am making it out is that it would contribute to my understanding of why it was so hard for my troops to make progress in my Falotti's WWI scenario and is for me currently to HG Sud any deeper into the Ukrainian soil back in bad ol' '41 in "The Waters of the Dnieper' 41" in particular; I felt most of the time that there is something badly wrong about the relation between the planned and final results of my combat operations there, shifting heavily the balance even more into the defending force's favour.




Oberst_Klink -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/8/2012 9:34:04 AM)


quote:

ORIGINAL: burroughs

The way I am making it out is that it would contribute to my understanding of why it was so hard for my troops to make progress in my Falotti's WWI scenario and is for me currently to HG Sud any deeper into the Ukrainian soil back in bad ol' '41 in "The Waters of the Dnieper' 41" in particular; I felt most of the time that there is something badly wrong about the relation between the planned and final results of my combat operations there, shifting heavily the balance even more into the defending force's favour.

Witam Kamerad Polski,

tweaking/checking the AD and MRPB settings, especially at pre-3.4 scenarios might 'ease' the issue a bit. I did it with LSiA, adjusted values make the defender, if he/she really uses 'ignore losses' all the time, pay. They units literally will fight until death and suffer. 2-3 combat rounds per turn with minimum losses will wear the defenders eventually out and they evaporate... So, it has a penalty, especially if 100% of the equipment gets lost and doesn't show up as replacement.

Klink, Oberst




shunwick -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/8/2012 2:21:26 PM)

...




governato -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/11/2012 12:14:52 AM)

Really useful stuff!




Sensei.Tokugawa -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/11/2012 10:30:48 AM)


quote:

ORIGINAL: Oberst_Klink


[Witam Kamerad Polski,

tweaking/checking the AD and MRPB settings, especially at pre-3.4 scenarios might 'ease' the issue a bit. I did it with LSiA, adjusted values make the defender, if he/she really uses 'ignore losses' all the time, pay. They units literally will fight until death and suffer. 2-3 combat rounds per turn with minimum losses will wear the defenders eventually out and they evaporate... So, it has a penalty, especially if 100% of the equipment gets lost and doesn't show up as replacement.

Klink, Oberst


Guten Tag Herr Oberst

Could it be done while the PBEM game has already been in progres for quite some time? I am desperate as the schedule has already been badly wrecked, it's the second week of July '41, I haven't even taken Lvov yet nor any other important hub, can't get the panzers groups any more traction, it's still pretty much the border battle despite crossing the river on the Romanian border and a jab at Odessa where the Soviets are massing for a counterstrike anyway after a hasty withdrawal. There has to be something wrong with the scenario working under the 3.4 patch and the new advanced rules applied.Hilfe!




Oberst_Klink -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/11/2012 12:51:30 PM)


quote:

ORIGINAL: burroughs


quote:

ORIGINAL: Oberst_Klink


[Witam Kamerad Polski,

tweaking/checking the AD and MRPB settings, especially at pre-3.4 scenarios might 'ease' the issue a bit. I did it with LSiA, adjusted values make the defender, if he/she really uses 'ignore losses' all the time, pay. They units literally will fight until death and suffer. 2-3 combat rounds per turn with minimum losses will wear the defenders eventually out and they evaporate... So, it has a penalty, especially if 100% of the equipment gets lost and doesn't show up as replacement.

Klink, Oberst


Guten Tag Herr Oberst

Could it be done while the PBEM game has already been in progres for quite some time? I am desperate as the schedule has already been badly wrecked, it's the second week of July '41, I haven't even taken Lvov yet nor any other important hub, can't get the panzers groups any more traction, it's still pretty much the border battle despite crossing the river on the Romanian border and a jab at Odessa where the Soviets are massing for a counterstrike anyway after a hasty withdrawal. There has to be something wrong with the scenario working under the 3.4 patch and the new advanced rules applied.Hilfe!

You can use the 'do not send after turn'.SAL file in the editor and adjust the settings as mentioned. I would ask your opponent first or alternatively I can do it as 3rd party. No guarantee that it might work. Send me the .SAL file anyway I will have a look.

Klink, Oberst




secadegas -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/11/2012 4:44:26 PM)


quote:

ORIGINAL: Oberst_Klink

You can use the 'do not send after turn'.SAL file in the editor and adjust the settings as mentioned. Klink, Oberst


[:-]

Thank god anti-cheating system doesn't allow. Imagine otherwise....








Sensei.Tokugawa -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/11/2012 6:06:21 PM)

Nobody meant cheating " ... because my peachy war plan has just went badly wrong ... "It's all for the sake of improving what we already have. Myself I am a Blitzkrieg Wargaming Club member and a hard core realism freak: I don't cheat, peak into the other side deployment or dispositions before the game gets launched, I don't even play mirrored games of the same scenarios nor swap sides right after a scenario got concluded for the sake of "illegitimate intel" on the enemy shop which used to be mine. I talked to my opponent about that and he agreed. We have still faulty scenarios with no regards to their designers who did their best; 2nd Indo re-creations in which Viet Cong has 10 % theater reconnaissance level ( nuts!) or with a map scale which compounded with the unit size allows for creation of an impenetrable complete front line all the way from the DMZ to the Mekong Delta and beyond. Does that make sense? What's bad in trying to get a grip on that? I have been defeated on numerous occassions and turned historical disasters into sound victories, too.Had no problem with that unless the gameplay kept realistic and there was no cheating on either side.

I was to ask Herr Oberst to do that for me in order not to see too much in the editor.Thanks for the heads-up anyway.




secadegas -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/11/2012 7:17:46 PM)

quote:

ORIGINAL: burroughs

Nobody meant cheating " ...


I wasn't suggesting that either.

Now just imagine... playing PBEM, save sal. file as sce. file, open on the editor and then... anything goes [:@]

It was done in the past, more often that you might think. Fortunately that's over now.







Oberst_Klink -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/11/2012 8:55:19 PM)

You have to restart the modified .sal as .sce anyway; It's just like a new scenario file where both forces pick up where they were; though sometimes the .sal gets corrupted when opened in the editor. I shall conduct some tests and report back on the matter later on.

Klink, Oberst




sPzAbt653 -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/11/2012 10:44:00 PM)

You can use the sal to sce feature to 'test' if an adjustment has the desired effect, but not to continue playing an 'in progress' game. The Event Engine does not know how to reset a game from turn 1 to a different start point, therefore things can get whacky.




Shazman -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/14/2012 12:43:54 AM)

I hate to say it but the current version of TOAW is slowly devolving into an unplayable mess. AA that doesn't work right and now ground combat that doesn't work right. I certainly hope the next patch has been fast tracked.




Oberst_Klink -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/14/2012 10:00:02 AM)

Shazman,

there are some 'issues' and certainly challenges; nothing that can't be fixed. Regarding the AAA, I had some revealing experiences *see http://www.matrixgames.com/forums/tm.asp?m=3053630&mpage=1�*

Interdiction as well as the integrated AA defences seem to work; my Soviet gunners were able to down more than a few Lw planes while on the move...

I agree though... the combat model regarding the RFC should be adjusted and I am sure Bob will use his weight to canvass it. That's why the community IS important - us.

Klink, Oberst




Shazman -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/14/2012 9:28:20 PM)

This:


quote:

ORIGINAL: Curtis Lemay

We were really hoping to get 3.5 out before this got noticed. Yes, sadly, this bug was introduced in about the middle of 3.4 development and never spotted till it was too late. I had a good chance to spot it when I checked out the AAA scale factor feature in the editor, but, of course, used AAA icons on the test units - and they worked. We didn't spot it till testing Naval AAA in 3.5 and found no ships ever did any AAA (and Telumar's work around won't work for that).

It is fixed in 3.5, though.

By the way, I'm posting this from the NE corner of Kenya.


And this:

quote:

ORIGINAL: Curtis Lemay
The problem is that the combat odds make no difference. It must have been that way from the beginning of TOAW. We did make the odds affect RBC chances in 3.4, but not RFCs. This needs to be addressed. I’ll get to that later.


These are not trifles nor small issues. However, I too know that these will be fixed. In fact, Bob has already mentioned in the quote above that the AAA issue is resolved. And the lack of combat odds being used will be resolved as well. But as 3.4 stands right now you are hard pressed to get an accurate resolution to any scenario EXCEPT for the fact that both contestants are equally hog-tied. [:D]




Oberst_Klink -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/15/2012 6:37:13 AM)

Shaz,

there are always 'workarounds' and a good example how even a pre-3.4 scenario(s) can be adjusted is Rob's RTM series... Have a look and play-test RTM - Crossings with the updated AD and MRPB. I am currently working *thanks to Joao Lima's permission' on a complete overhaul of Kharkov '43; alas I shall name the post-beta Kharkov '43 - Operation Star & Gallop; not to confuse it with his and the other version. For testing purposes I recommend to use my Tutorial '41 as template. You can use, 'abuse', change, modify. etc. to your liking. Just mention the Gefechtsstand, ja?

Klink, Oberst

[image]local://upfiles/28259/1E663691DE574063ACB4BACCD4C3DA13.jpg[/image]




Curtis Lemay -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/15/2012 12:40:29 PM)

quote:

ORIGINAL: Shazman

This:


quote:

ORIGINAL: Curtis Lemay

We were really hoping to get 3.5 out before this got noticed. Yes, sadly, this bug was introduced in about the middle of 3.4 development and never spotted till it was too late. I had a good chance to spot it when I checked out the AAA scale factor feature in the editor, but, of course, used AAA icons on the test units - and they worked. We didn't spot it till testing Naval AAA in 3.5 and found no ships ever did any AAA (and Telumar's work around won't work for that).

It is fixed in 3.5, though.

By the way, I'm posting this from the NE corner of Kenya.


And this:

quote:

ORIGINAL: Curtis Lemay
The problem is that the combat odds make no difference. It must have been that way from the beginning of TOAW. We did make the odds affect RBC chances in 3.4, but not RFCs. This needs to be addressed. I’ll get to that later.


These are not trifles nor small issues. However, I too know that these will be fixed. In fact, Bob has already mentioned in the quote above that the AAA issue is resolved. And the lack of combat odds being used will be resolved as well. But as 3.4 stands right now you are hard pressed to get an accurate resolution to any scenario EXCEPT for the fact that both contestants are equally hog-tied. [:D]


There's no question that the AAA bug was a major screw up. But, that's the risk that comes with doing updates. Note that the beta was out there for quite a while without anyone spotting this.

And RFCs are not right yet either. But they were hardly right prior to 3.4. Prior to 3.4 neither combat odds nor terrain made any difference in RFC chances. 3.4 fixed the terrain part of that. So, we cured its cold, but that revealed that it has the flu as well. That doesn't mean we shouldn't have cured its cold.

By the way, I'm posting this from Jakarta.




secadegas -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/15/2012 7:41:07 PM)

quote:

ORIGINAL: Oberst_Klink

That's why the community IS important - us.


I'd like to support "Oberst" here. Even if i have strong reserves about current version combat resolution, actual TOAW gaming platform is much better than what we had before in earlier versions.

And all the steps forward were only due to very dedicated people from the community.








jmlima -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/17/2012 11:33:59 AM)


quote:

ORIGINAL: Curtis Lemay
... So, the lower the attacker morale, the harder it is for them to cause the defender to RFC. This makes it particularly hard for a low proficiency force to make progress against a high proficiency force. ...


I may be missing something here but, is the above noted as being a problem?...

(I'm completely baffled if so, because the above makes absolute sense in terms of what history as been showing us. If that's the way the game is working, then it's working fine.)




Oberst_Klink -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/17/2012 11:53:42 AM)


quote:

ORIGINAL: jmlima


quote:

ORIGINAL: Curtis Lemay
... So, the lower the attacker morale, the harder it is for them to cause the defender to RFC. This makes it particularly hard for a low proficiency force to make progress against a high proficiency force. ...


I may be missing something here but, is the above noted as being a problem?...

(I'm completely baffled if so, because the above makes absolute sense in terms of what history as been showing us. If that's the way the game is working, then it's working fine.)


Joao,

The problem is, as I stated in one of my threads, the 'ignore losses' issue that prevents just doing what Bob found out. You use a crap unit, no supply, no readiness and they hold out nearly infinitely against any attacker. Hence Bob did a comprehensive test that highlighted the RFC problem and confirmed what I and the Swedish Major *he really is one* found out battling various PBEM's only to realize that even a mobile campaign bogs down to a WW1 stalemate.

Klink, Oberst

P.S. off-topic: Still tweaking Kharkov '43 *which I shall rename to Kharkov'43 - Operation Star & Gallop in order not to confuse it with the others...* 'n thanks for giving me the permission.




Curtis Lemay -> RE: Analysis of RFCs by defenders set to Ignore Losses (3/17/2012 12:44:33 PM)


quote:

ORIGINAL: jmlima


quote:

ORIGINAL: Curtis Lemay
... So, the lower the attacker morale, the harder it is for them to cause the defender to RFC. This makes it particularly hard for a low proficiency force to make progress against a high proficiency force. ...


I may be missing something here but, is the above noted as being a problem?...

(I'm completely baffled if so, because the above makes absolute sense in terms of what history as been showing us. If that's the way the game is working, then it's working fine.)



It is a problem if combat odds don't matter. That means that a really huge low prof force can't make much progress against a really small high prof force. And that's not historical. Remember the saying: "Quantity has a quality all on its own".




Page: [1] 2 3   next >   >>

Valid CSS!




Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI
1.46875