GA statistical musings

[rustysi]

Also I assume defenders have it exactly the same as attackers with respect to HQ bonuses at least.

WRT preparation points, I would say yes.

[Yaab]

Interesting.

Why did you do your LCU/HQs tests with LCUs having adequate or surplus support squads? In the game, almost all LCU exist and fight with inadequate support in their TOE (the proportion of non-support devices to support devices is roughly 100:70). Can you do a test with LCUs at 70% support and 0 prep, and accompanying HQ with 0 prep, with enough support to put the LCU in GREEN-support category?

[GetAssista]

ORIGINAL: Yaab
Interesting.

Why did you do your LCU/HQs tests with LCUs having adequate or surplus support squads? In the game, almost all LCU exist and fight with inadequate support in their TOE (the proportion of non-support devices to support devices is roughly 100:70). Can you do a test with LCUs at 70% support and 0 prep, and accompanying HQ with 0 prep, with enough support to put the LCU in GREEN-support category?

Well, potential combinations of factors are nigh infinite, so one needs some limiting assumptions.
I run all combat tests under the assumption that support does not matter for combat. Earlier I observed that LCU with zero support had the same adjusted AV on average as a fully supported LCU.

What I'm planning to do is what I mentioned in the thread - augmenting my recovery speed tests to look for the support (white/green) thresholds effects on the recovery from disablements.

[rustysi]

augmenting my recovery speed tests to look for the support (white/green) thresholds effects on the recovery from disablements.

Guys, I think you are chasing your 'tails' again. All the white/green really mean is that the white unit has enough inherent support and the green is getting some from an HQ because it is deficient in support devices. Any base with surplus support will help any unit in the base recover its disabled devices.

Again we search for definitive numbers in an abstracted element. There are a number of factors that effect this, and I'm sure many more that GG has in there as well. Trying to get absolute results is an exercise in futility IMHO. To each their own though, I guess.[:)]

[Lokasenna]

He is testing to see if extra support is even more beneficial.

For the purposes of AV and combat, I don't think there was ever any question that having more than enough support does nothing. It is in unit recoveries that we have hints that having more than enough support in the hex (i.e., "green" even though that is a unit-level measurement instead of hex-level) is better than having just enough.

[Yaab]

Well, as the Allies I don't pay much attention to undisabling. My units in CONUSA, Australia or India undisable nicely. However, the Chinese in the game are so wretched that the whole process of undisabling warrants further scrutiny. Many Chinese units start half disabled. Couple that with their white-support quirk and they are in a sort of stasis of wretchedness. The manual is vague, the forum myths abound, the devs are dispersed to the four winds. My GREEN-supported Chinese units undisable at acceptable rate, though still slower than their CONUSA counteraparts.

The Allies cracked the Enigma code, we, the players, must crack the WITP:AE code.

[rustysi]

It is in unit recoveries that we have hints that having more than enough support in the hex (i.e., "green" even though that is a unit-level measurement instead of hex-level) is better than having just enough.

Oh no, this is absolutely true. The more the merrier. Add to that, a large base, tons of supply, as far as you can be from the enemy, other surrounding bases, etc.

However, the Chinese in the game are so wretched that the whole process of undisabling warrants further scrutiny.

You only need to look at their situation and see why. They lack everything they need to do so effectively.

must crack the WITP:AE code.

Pay no attention to that man behind the curtain.[:D]

[Lokasenna]

I don't think distance from the enemy, in itself, (aside from just not having any enemy in-hex) means anything for unit recovery. Why would it?

[Macclan5]

Very nice analysis - and thank you.

Cracking the code is an interesting facet for the fans of this game and the contribution is appreciated

So the "laymans" analysis is ??? [8D]

1a) Having Units supported with full TOE and supplies is best.

1) Having units prep'ed is better

2) Having HQ in hex is better

3) Having HQ in range is better but not as good as above

4) Having HQ prep'ed is better but not as good as above.

[GetAssista]

So, I ran the tests aimed at determining whever distribution of support (between native to LCU and outside from HQs) helping LCU to recover matters for the speed of recovery.

A scenario without enemy activity wtasoever.
3 types of IDs resting in 3 identically built, supplied and climatized bases took part, each type in a particular base.
All of the IDs had 450 infantry squads disabled at start, custom commanders with 60 in each stat, replacements are turned off for everything.
Types of IDs had varying number of healthy support squads: 451, 449 and 1. Means type 1,2 will always have enough native support for healthy squads (i.e will always be WHITE), and type 3 will almost always have outside support only (will always be GREEN). Difference between 451 and 449 is there to check if ratio of support to total non-support squads matters. No other squads or devices in the IDs.
Each base further had a HQ with 200+ support squads tailored to have identical number of total support squads in each base, with custom commanders with 60 in each stat.

Result: there are no statistical differences in the speed of healing between different types of divisions.

Result 2: confirmation of previous research. A/B/C fragments heal even faster than whole divisions in absolute terms (that is # of squads healed). This means that total healing speed will be 3x+ faster if you divide the division into A/B/C for R&R.

[rustysi]

This means that total healing speed will be 3x+ faster if you divide the division into A/B/C for R&R.

Know this works for replacements, but was unsure/unaware in the recovery mode. Will have to use it.

[Yaab]

ORIGINAL: GetAssista

So, I ran the tests aimed at determining whever distribution of support (between native to LCU and outside from HQs) helping LCU to recover matters for the speed of recovery.

A scenario without enemy activity wtasoever.
3 types of IDs resting in 3 identically built, supplied and climatized bases took part, each type in a particular base.
All of the IDs had 450 infantry squads disabled at start, custom commanders with 60 in each stat, replacements are turned off for everything.
Types of IDs had varying number of healthy support squads: 451, 449 and 1. Means type 1,2 will always have enough native support for healthy squads (i.e will always be WHITE), and type 3 will almost always have outside support only (will always be GREEN). Difference between 451 and 449 is there to check if ratio of support to total non-support squads matters. No other squads or devices in the IDs.
Each base further had a HQ with 200+ support squads tailored to have identical number of total support squads in each base, with custom commanders with 60 in each stat.

Result: there are no statistical differences in the speed of healing between different types of divisions.

Result 2: confirmation of previous research. A/B/C fragments heal even faster than whole divisions in absolute terms (that is # of squads healed). This means that total healing speed will be 3x+ faster if you divide the division into A/B/C for R&R.

Thanks for testing.

I am baffled. My WHITE 74th Chinese Corps in scen 100 is frozen into some kind of a disablement stasis. Exp 35/morale 35, undisables 1-3 squads each turn max. The 10th Chinese Corps in the same city is GREEN, and undisables 6-8 devices per turn. Dividing into ABC of course works in general, however I find it tedious to browse through divided units icons in big bases like Chungking or Tacoma, so I try to keep all div/corps LCUs undivided.

[Yaab]

ORIGINAL: GetAssista

ORIGINAL: Lokasenna
@GA - did the units have enough Support devices without the presence of the HQs in the hex? That is the one indirect thing that an HQ in the hex could do outside of what you appear to have just tested. If they didn't have enough Support without the HQ units, I'm hesitant to chalk this difference up to the presence of HQs alone.

I've done some tests earlier on that urban legend that support helps in adjusted AV. Had a tank regiment attack completely stripped of support. Adjusted AV was roughly equal to base AV even without any support squads in the hex.

Also:

ORIGINAL: GetAssista
TOE of ID had 450 infantry squads and 450 support

GetAssista, do you still have the data from this test? How was the enemy AV in them? If support devices are any good in combat, then the fully supported tank regiment should disable more enemy devices in the firing phase of the attack, thus lowering enemy's AV on average in assault phase. If enemy AV was comparabale when facing both kinds of tank regiment, then maybe your analysis your right, and the support squads only help in undisabling devices.

[GetAssista]

ORIGINAL: Yaab
GetAssista, do you still have the data from this test? How was the enemy AV in them? If support devices are any good in combat, then the fully supported tank regiment should disable more enemy devices in the firing phase of the attack, thus lowering enemy's AV on average in assault phase. If enemy AV was comparabale when facing both kinds of tank regiment, then maybe your analysis your right, and the support squads only help in undisabling devices.

Did not check enemy AV at that time. I don't really think support is affecting the enemy in any way when attacking. My running hypothesis so far is that support is used in R&R + provides some default defensive AV like any other non-combat squad and that's it. Might do some tests in the future

[BBfanboy]

ORIGINAL: Yaab


I am baffled. My WHITE 74th Chinese Corps in scen 100 is frozen into some kind of a disablement stasis. Exp 35/morale 35, undisables 1-3 squads each turn max. The 10th Chinese Corps in the same city is GREEN, and undisables 6-8 devices per turn. Dividing into ABC of course works in general, however I find it tedious to browse through divided units icons in big bases like Chungking or Tacoma, so I try to keep all div/corps LCUs undivided.

The game engine does replacements/support operations on units in index number order. If you 74th Corps has a much higher index number than other Chinese units present, the latter would get support points and replacement devices first (assuming all have "replacements allowed". There is also some influence from the Admin. Skill of the unit leaders, but I am not clear on just how much it affects things. I don't know if the Corps/Area HQ assignment alignment has a big bearing either, but I believe it helps somewhat with Admin. tasks.

[Zorch]

ORIGINAL: BBfanboy

ORIGINAL: Yaab


I am baffled. My WHITE 74th Chinese Corps in scen 100 is frozen into some kind of a disablement stasis. Exp 35/morale 35, undisables 1-3 squads each turn max. The 10th Chinese Corps in the same city is GREEN, and undisables 6-8 devices per turn. Dividing into ABC of course works in general, however I find it tedious to browse through divided units icons in big bases like Chungking or Tacoma, so I try to keep all div/corps LCUs undivided.

The game engine does replacements/support operations on units in index number order. If you 74th Corps has a much higher index number than other Chinese units present, the latter would get support points and replacement devices first (assuming all have "replacements allowed". There is also some influence from the Admin. Skill of the unit leaders, but I am not clear on just how much it affects things. I don't know if the Corps/Area HQ assignment alignment has a big bearing either, but I believe it helps somewhat with Admin. tasks.

I hate this kind of 'feature'.
Either the player should be allowed to specify priority, or there should be a rational basis (like admin ability) for ordering. Even if that rational basis is semi-random.
Ditto for TF execution order...

[BBfanboy]

ORIGINAL: Zorch

ORIGINAL: BBfanboy

ORIGINAL: Yaab


I am baffled. My WHITE 74th Chinese Corps in scen 100 is frozen into some kind of a disablement stasis. Exp 35/morale 35, undisables 1-3 squads each turn max. The 10th Chinese Corps in the same city is GREEN, and undisables 6-8 devices per turn. Dividing into ABC of course works in general, however I find it tedious to browse through divided units icons in big bases like Chungking or Tacoma, so I try to keep all div/corps LCUs undivided.

The game engine does replacements/support operations on units in index number order. If you 74th Corps has a much higher index number than other Chinese units present, the latter would get support points and replacement devices first (assuming all have "replacements allowed". There is also some influence from the Admin. Skill of the unit leaders, but I am not clear on just how much it affects things. I don't know if the Corps/Area HQ assignment alignment has a big bearing either, but I believe it helps somewhat with Admin. tasks.

I hate this kind of 'feature'.
Either the player should be allowed to specify priority, or there should be a rational basis (like admin ability) for ordering. Even if that rational basis is semi-random.
Ditto for TF execution order...

Your control requires you to shut off replacements for other units, but I am not sure if the allocation of support to reduction of disablements is affected by this.
I put this in the same category with the fact that you cannot decide the priority for repairs to your base's forts, port, airfield services or runways. There are times when you have no intention of bringing in aircraft yet but you still get the runways and airfield services fixed first ... [:(]

[Lokasenna]

But if you have plenty of devices in the pools (or more than enough support), it wouldn't matter what order the units received them - it would just be up to the individual units' rolls.

[BBfanboy]

ORIGINAL: Lokasenna

But if you have plenty of devices in the pools (or more than enough support), it wouldn't matter what order the units received them - it would just be up to the individual units' rolls.

True, but there seem to be some units that just won't fill out at any kind of acceptable rate. I had that problem with 6th Marines - oodles of devices in the pools, oodles of supply, great leader with high admin skills, unit in rest mode, WCUSA HQ in range. I got maybe one marine inf. squad and one support squad per turn. I even moved it to San Francisco where other units were filling out quickly - no change. The clincher was that other Marine regiments that arrived after the 6th filled out and went on their way while the 6th just would not take replacements at any kind of decent rate.

[Lokasenna]

I'd have rebuilt it into the Division and then split into 3rd