Attack stats: Your number vs. Modern War's number

[JohnnyR]

One of the great mysteries here on the forums that I have seen time and again brought up, but never resolved-the attack score difference. Here is a possible reason nobody's numbers add up from their spreadsheet:

It could be that our country, building, or combination of boosts are applied AFTER our battle units are selected to fight. Here's an example of what might transpire:

Let's say you have your lowest set of units brought into battle are Super Hornets (air) and SCUDs (ground).

SH attack 36
SCUD attack 35

With max building boost of your ground units all the way up to 20% SCUDs should have an attack of 42. However, if units going into battle are decided on their UNboosted attack number, that extra seven attack points will be missing when the game chooses the SH over the SCUD. Times this by however many bottom units are affected and the "margin of error" can vary widely.

Anybody with a complete spreadsheet care to test this possibility out?

I've been away for a bit, so pardon me if this thing has been figured out. I saw this question again though in another thread and I figured a thread of it's own can enlighten all of us.

[Dreno33]

you're alive?

EDIT: and to the OP - that is a very interesting thought.

so you're saying maybe, even if after boosts incorporated and the SCUDs are stronger Att #, that the SH would be selected due to the fact of its raw attack? maybe, i guess..

so the profile att/def numbers...
do they reflect the CURRENT units taken to battle only? or only after raw att/def numbers are selected?

[JohnnyR]

Hahaha!

Very much so. :-)

By the way, if you still have the pacman up, I can't wait to check out your base. I'm playin' on me droid. :)

[albeezy]

Going to work on this a bit more tomorrow. I have 10 allies and take 40 units to battle. I have input my units into the toolkit but they are off. I don't think defense values for buildings are included in my stats? My attack is off by 8 and I suspect it is a rounding error. I will try to confirm this in morning. Is the guerilla striker +20% bonus active in the toolkit?

[Dr. Dengus]

You could be on to something JR. According to your theory, my attack score in my spreadsheet is only 10 points higher than my profile score. Before, when I would assume the boosts are taken into account before going to battle, my spreadsheet would be 120 points higher than the true attack value. So it's definitely a difference in the right direction.

[Poopenshire]

you're alive?

Seriously?? Get the f out of our forum if you cannot comment on topic. That should be a pm.

Now on topic

Johnny,

Buildings have an score that is added to atleast the defense score. This includes defense buildings and non-cash.
KOD attempted to make a spreadsheet to determine attack and defense but was always about 3 to 4 percent low for me.
The toolkit is also low but takes buildings into account. I do believe we discussed the nation boost determining top placement and then building boost to follow.

[Dreno33]

Hahaha!

Very much so. :-)

By the way, if you still have the pacman up, I can't wait to check out your base. I'm playin' on me droid. :)

of course it is!(: just shifted closer to the corner now. took about a day to do so, lmao.
im lvl 24 now. lil dreno is growing up.

how've you been man? looooong time. PM me your android number, ill add ya

also, first post edited

[albeezy]

One of the great mysteries here on the forums that I have seen time and again brought up, but never resolved-the attack score difference. Here is a possible reason nobody's numbers add up from their spreadsheet:

It could be that our country, building, or combination of boosts are applied AFTER our battle units are selected to fight. Here's an example of what might transpire:

Let's say you have your lowest set of units brought into battle are Super Hornets (air) and SCUDs (ground).

SH attack 36
SCUD attack 35

With max building boost of your ground units all the way up to 20% SCUDs should have an attack of 42. However, if units going into battle are decided on their UNboosted attack number, that extra seven attack points will be missing when the game chooses the SH over the SCUD. Times this by however many bottom units are affected and the "margin of error" can vary widely.

Anybody with a complete spreadsheet care to test this possibility out?

I've been away for a bit, so pardon me if this thing has been figured out. I saw this question again though in another thread and I figured a thread of it's own can enlighten all of us.
I will confirm this as right or wrong in the morning. I have my top 92 units in a spreadsheet and then decided to drop allies to have smaller data set. So I know all 40 units I take to battle and also the next 52. I can compare for anomalies

[Dreno33]

Seriously?? Get the f out of our forum if you cannot comment on topic. That should be a pm.

Now on topic

Johnny,

Buildings have an score that is added to atleast the defense score. This includes defense buildings and non-cash.
KOD attempted to make a spreadsheet to determine attack and defense but was always about 3 to 4 percent low for me.
The toolkit is also low but takes buildings into account. I do believe we discussed the nation boost determining top placement and then building boost to follow.

hey, d!ck,
watch your mouth. i edited it before you even posted. why are you becoming the new seq?

[JohnnyR]

you're alive?

EDIT: and to the OP - that is a very interesting thought.

so you're saying maybe, even if after boosts incorporated and the SCUDs are stronger Att #, that the SH would be selected due to the fact of its raw attack? maybe, i guess..

so the profile att/def numbers...
do they reflect the CURRENT units taken to battle only? or only after raw att/def numbers are selected?

Bingo.

Didn't realize bout KOD's sheet, hmm. I threw this out as a possibility in the past, seemed to fall flat then.

[Gambit12]

Seriously?? Get the f out of our forum if you cannot comment on topic. That should be a pm.



hey, d!ck,
watch your mouth. i edited it before you even posted. why are you becoming the new seq?

Hey Poopenshire you need to peace out on Dreno...remember you guys are in original 25 members who share your pics in this forum...the below thread can prove it...


http://www.funzio.com/forum/showthread.php?36013-Name-the-Players-Name-Come-Participate-or-watch

[Hivesy]

hey JohnnyR, seems like old times with you starting threads :)

You make a great point, it'll be interesting to see if anyone can prove it one way or the other. My personal feeling is that boost IS taken in to account before units are picked for battle just because my top 45 units that are shown are in the correct order. i.e. An air unit with a lower attack will be placed higher than a ground unit with a higher attack as the boost gives it a higher total attack. This makes me think the game does look at boost stats for deciding units but obviously that's just a theory.

The other issue is one of pure maths. How are boosts calculated by the game? Results can vary hugely.

Super Hornet = 36 attack raw.

With Russia nation boost (10%) and a level 5 air building boost (10%) and an imaginary Guerilla Striker boost (20%) your attack stat could be:

36 + 10% + 10% +20% = 52.27

It could also be:

36 + 40% = 50.4

Big difference. Maybe this is the very simple maths question we need to be asking CCM? How do they calculate the boosts?

[Mcdoc]

. . .You make a great point, it'll be interesting to see if anyone can prove it one way or the other. . . Maybe this is the very simple maths question we need to be asking CCM? How do they calculate the boosts?One thing I can predict - CCM will not divulge any part of any formula :)

[Poopenshire]

The other issue is one of pure maths. How are boosts calculated by the game? Results can vary hugely.

Super Hornet = 36 attack raw.

With Russia nation boost (10%) and a level 5 air building boost (10%) and an imaginary Guerilla Striker boost (20%) your attack stat could be:

36 + 10% + 10% +20% = 52.27

It could also be:

36 + 40% = 50.4

Big difference. Maybe this is the very simple maths question we need to be asking CCM? How do they calculate the boosts?

I have seen in the past Option 2 being the correct one. The boosts were always on the original stat and not compounded. I even believe i remember seeing this from CCM, I will look today and see if i can find the thread and link it.

[Hivesy]

I have seen in the past Option 2 being the correct one. The boosts were always on the original stat and not compounded. I even believe i remember seeing this from CCM, I will look today and see if i can find the thread and link it.

Thanks poop that would be great if you can link to the info as I'm really intrigued. All my figures and spreadsheets are worked out on option 1 so this could be the answer to why I'm out.

[Thief]

hey JohnnyR, seems like old times with you starting threads :)

You make a great point, it'll be interesting to see if anyone can prove it one way or the other. My personal feeling is that boost IS taken in to account before units are picked for battle just because my top 45 units that are shown are in the correct order. i.e. An air unit with a lower attack will be placed higher than a ground unit with a higher attack as the boost gives it a higher total attack. This makes me think the game does look at boost stats for deciding units but obviously that's just a theory.

The other issue is one of pure maths. How are boosts calculated by the game? Results can vary hugely.

Super Hornet = 36 attack raw.

With Russia nation boost (10%) and a level 5 air building boost (10%) and an imaginary Guerilla Striker boost (20%) your attack stat could be:

36 + 10% + 10% +20% = 52.27

It could also be:

36 + 40% = 50.4

Big difference. Maybe this is the very simple maths question we need to be asking CCM? How do they calculate the boosts?

I like this Theory! JR you might be on to something as well. I will keep lurking in the hopes that i can add some useful information later down the road. Great to see Forums that are informative going up!

[albeezy]

Okay here are some preliminary resluts.

My Stats:
Nation - UK
Attack - 10173 (as shown in my profile)
Allies 10
Units brought to battle 40
My boost buildings are as follows:
Air - 4
Sea - 7
Ground - 4
Infantry - 4
Composites - 8
Infirmary - 2

List of battle units and raw attack
1. desert rat...............................1100
2. dragon viper...........................387
3. elite heavy..............................370
4. Deadly Tiger............................360
5. elite base defense..................336
6. heavy carrier...........................287*
7. black dragon fighter................311
8. Guerilla striker.........................292
9.-23. transport raider(x15)........283
24. blue bomber..........................259
25. Fortuen Talon........................180
25. storm breaker........................155
26. stealth command...................118*
27. veteran captain.....................121
28. rapid thunderbolt...................85
30. reinforced defense.................79
31. stealth sniper.........................72
32. advanced recon .....................70
33. quick globe traveler................70
34. stealth sea shadow................59*
35. Skull Aggressor.......................63
36. Deadly Jet...............................62
37. Hidden Operative....................62
38. Swift Osprey............................60
39. Stealth Runner........................54*
40. Crystallized Artillery.................59

Using just these raw attack stats my total raw attack is 9316
Using just the nation boost my total attack is 9384 (68 additional attack points from nation boost)
Just applying the building boost my total attack is 10105 (789 additional attack from boost buildings)

My total attack in profile was 10173 which appears to be raw attack + attack points from nation boost + attack points from building boost or 9316+68+789=10173

Yes, I was able to calculate my profile attack to the exact number! I did this by finding my raw attack score, and then adding each boost separately

Just appllying the Guerilla Striker 20% attack boost to my raw attack, my total attack should be 10671
This means my profile if the Guerilla Striker boost was showing would be 11528 (i.e., 10173+1355)

This unequivocally proves that the Guerilla Striker + 20% is not showing in the stats.

A cash hit on the boss is doing 8757 damage which is about 94% of my raw attack score.
A free hit on the boss is doing 5868 damage which is about 63% of my raw attack score.
A gold hit (yes I spent 20 gold just to test this out, hook me back up CCM) is doing 35028 damage which is about 376% of my raw attack score.

This clearly and undeniably demonstrates that the Guerilla Striker Boost is not working on the Boss events.

* It appears that at least nation boost is consdiered when choosing the order in which units are brought to battle.

The steal runner has attack of 54 but is listed before the crystalized artillery with attack of 59. With my nation boost they each have an attack of 59 and the stealth runner may be listed first because of a higher defensive value.

Also, look at the stealth sea shadow with attack 59 being listed above the skull aggressor attack 63 and after the quick globe traveler attack 70. With nation boost the stealth sea shadow has attack of 65 placing it ahead of the raw attack of the skull aggressor's 63 but less than the 70 raw attack of the quick globe trotter. If just boost buildings were considered the stealth sea shadow would have attack of 67 while the skull aggressor would have an attack of 68. Thus, if only boost building boosts were considered the skull aggressor would be chosen ahead of the stealth sea shadow.

There are similar results for the Heavy Carrier at spot 6. In each of these cases, we see a unit with a lower raw attack score being chosen for battle over another unit with a higher raw attack score due to at least nation boost. It is also clear that the units chosen for battle are not determined based on only the boost from buildings. For example, the Heavy Carrier has a raw attack of 287 while the Black Dragon Fighter's raw attack is 311, yet the Heavy Carrier is listed before it. With just boost from buildings the Heavy Carrier would have an attack of 327 while the BDF would have an attack of 336. This proves that Building Boost is not the only thing considered when choosing units for battle. What I cannot clearly discern is whether the units brought into battle are chosen on nation boost alone, or both nation boost and boost buildings. I would need to upgrade boost building other than the naval research center until I could get that data. I will make that my next upgrade.

The last unit I take into battle is the Crystalized Artillery 59/14 (Ground). With the boost from the Electronics Factory its stats are 59*1.08 = 63.72/ 14*1.08 = 15.12, (/)
When I had more than 10 allies the next 5 units brought to battle were:
Ambush Missle Frigate 53/48 (Sea),
Rapid Strike Spetsnaz 57/27 (Infantry),
Attack Turbo 56/34 (Air), then the
Black Ops Infiltrator 55/42 (Infantry),
Reinforced SSGN 50/31 (Sea)

[Poopenshire]

Here is a nice attempt by KOD to work on this:
http://www.funzio.com/forum/showthread.php?36987-Spreadsheets-and-more.&highlight=boosted+attack+scores

[albeezy]

I have also been able to confirm that the boosts are applied individually. In other words to determine your total attack you take your raw attack for each unit and ADD the rounded totals for each boost to the raw attack separately. For example, My profile attack as stated above is 10173. I calculated the attack points contributed by boost buildings by rounding the raw attack*the boost from each boost building and calculated the attack points contributed by nation boost by rounding the raw attack*nation boost for my sea units. I then added each of these together and I arrived at 10173.

When I was trying to calculate it by adding the boosts together I was always off a few points. Take the Storm Breaker with an attack of 155 for example. WIth the 14% boost from my naval research center and the 10% nation boost for being UK compare the folowing 2 formulas:
Rounded Raw Attack * total boost (nation + building boost)
round(155*1.24,0) = 192

vs.

Raw Attack + Rounded total attack points from boost building + Rounded total attack points from nation boost
155+round(155*.14,0)+round(155*.1,0) = 193

Using the second calculation on all my units I arrive at the correct attack value as shown in my profile. I think this confirms Hivesy's Question

If you have multiple of the same units, such as my transport raiders it is as follows:
Raw Attack*# of units + Rounded ((# of units * raw attack)*boost from building) + Rounded Rounded ((# of units * raw attack)*nation boost )
so for my 15 transport raiders it would look like this
15*283+round((283*15)*.08,0)+round((283*15)*0.0,0) = 4585

[JohnnyR]

Here is a nice attempt by KOD to work on this:
http://www.funzio.com/forum/showthread.php?36987-Spreadsheets-and-more.&highlight=boosted+attack+scores

Thanks poop, you're always good for a link, lol. Haven't had time to take a thorough look but it appears the nut wasn't cracked eve. By KoD. :-/ Seems Albeezy is on the path tho.

[JohnnyR]

I have also been able to confirm that the boosts are applied individually. In other words to determine your total attack you take your raw attack for each unit and ADD the rounded totals for each boost to the raw attack separately. For example, My profile attack as stated above is 10173. I calculated the attack points contributed by boost buildings by rounding the raw attack*the boost from each boost building and calculated the attack points contributed by nation boost by rounding the raw attack*nation boost for my sea units. I then added each of these together and I arrived at 10173.

When I was trying to calculate it by adding the boosts together I was always off a few points. Take the Storm Breaker with an attack of 155 for example. WIth the 14% boost from my naval research center and the 10% nation boost for being UK compare the folowing 2 formulas:
Rounded Raw Attack * total boost (nation + building boost)
round(155*1.24,0) = 192

vs.

Raw Attack + Rounded total attack points from boost building + Rounded total attack points from nation boost
155+round(155*.14,0)+round(155*.1,0) = 193

Using the second calculation on all my units I arrive at the correct attack value as shown in my profile. I think this confirms Hivesy's Question

If you have multiple of the same units, such as my transport raiders it is as follows:
Raw Attack*# of units + Rounded ((# of units * raw attack)*boost from building) + Rounded Rounded ((# of units * raw attack)*nation boost )
so for my 15 transport raiders it would look like this
15*283+round((283*15)*.08,0)+round((283*15)*0.0,0) = 4585

w00t! Have been wondering this for ages! Awesome work dude, pretty dang thorough I must say!

[albeezy]

Thanks poop, you're always good for a link, lol. Haven't had time to take a thorough look but it appears the nut wasn't cracked eve. By KoD. :-/ Seems Albeezy is on the path tho.
The path for determining your stats is done if you know ALL units you take to battle. The only question left is whether the units brought to battle are determined based on nation boost only or both nation and building boosts.

Unfortuneatly I will have to upgrade my airship hangar 4 times to determine this...

[Dr. Dengus]

That was pretty thorough albeezy, I'll compare it with my army later. I like how someone rated this thread 1 star too. "Duurrr math hard I rate 1 star."

[albeezy]

If anyone talks to Kimshoo have him compare this to his app. When I plug the units listed above into his app it is off by 11 (probably a rounding error?)

[Vballmadam]

You are awesome Beezy! Thank you for all the time and effort it has taken you. Now if only I knew what I take into battle! Lol

[MadManDK]

Old news: http://www.funzio.com/forum/showthread.php?36987-Spreadsheets-and-more./page3

[albeezy]

Old news: http://www.funzio.com/forum/showthread.php?36987-Spreadsheets-and-more./page3
Is the OLD NEWS that sort order is nation boost only?

[MadManDK]

I didn't see any formulas on your sheet MMDK? I'm confused

Oh and while the fact that the guerilla striker doesn't work is old news, it is now a fact! LOL

Plenty of formulas on the different sheets/pages/tabs/..., but the main point back then was to try to prove once and for all how your army is sorted and selected. And my conclusion was that it did it based raw stats plus the nation boost.

Not sure about my second conclusion, I'll have a look at that tomorrow again, time to sleep.

[albeezy]

Plenty of formulas on the different sheets/pages/tabs/..., but the main point back then was to try to prove once and for all how your army is sorted and selected. And my conclusion was that it did it based raw stats plus the nation boost.

Yeah, I saw your posts on that, but the one scenario I didnt see (may have missed it) was where the total boosted stats of one unit that gets a nation boost is less than another unit that has a higher total boost off of just boost buildings? All of my data indicates that the order includes at least some accounting for nation boost, but my data just so happens to also only include units that receive a nation boost that also have a higher total boost than the next lower unit. If I up my electronics factory to level 8 I can create that scenario between the stealth runner with an attack of 54 (67 with both nation and building boos) and the crystallized artillery having an attack of 59 (64 with building boost). If I take my electronics factory to level 8 the CA will have an attack of 68 with total boost, which will be 1 point higher than the SR. If it switches then total boost is used. If it does not switch then only nation boost is used.

I couldn't tell from your link whether you created this scenario? Save me from wasting my next four upgrades on the electronics factory! LOL

[MadManDK]

I'll try your formula in my sheet tomorrow and see if the results are better than what I have now.

[albeezy]

I'll try your formula in my sheet tomorrow and see if the results are better than what I have now.

Can you PM me a link to your spreadsheet so I can check it out as well? Thanks!

[Hivesy]

I bow at your feet albeezy and madman. Awesome work, this is almost more than a game isn't it :)

[Poopenshire]

my only request is that you find a way to export my army in an easy simple to use method.

[Mcdoc]

We're all nerds :)

[albeezy]

my only request is that you find a way to export my army in an easy simple to use method.

I gave up programming when I went to law school. Need to get back into it and make myself a UP-TO-DATE (AKA MODERN) WAR game that works and trust me, admirAL BEEZY would provide excellent customer service and 50% off palm trees while supplies last! Oh and speaking of palm trees, why is it that when I had over 600 palm trees I could still build them on certain days of the week but now that I only have 150 or so it keeps telling me I can only build 200? WTF?

[John Snow]

Is there a defined order in which units with the same raw attack score get brought to battle? I thought I read something about this but can't find a link to it - something like "air before sea before ground before infantry".

[MadManDK]

Can you PM me a link to your spreadsheet so I can check it out as well? Thanks!

I've been thinking about this and there are a tons of ways Gree could calculate the stats, so I tried to capture all these in one sheet and put in my army to see what the different calculations give. But as I've realized with my original sheet, something is off, as none of them fit even close (i used to be a few points of out of over 50K, but now it's like 2K off). This seems to be since a couple of weeks ago (2-5 weeks), so I'm wondering if Gree has messed up the calculations in order to fix something.

Anyhow, I'm out of time for today, so I'll share the sheet and you can put in your own numbers and see what gives. Not a lot of explanation with the sheet, so on the two first sheets fill in the grey areas.

Link: http://www.filedropper.com/modernwartest

[John Snow]

The path for determining your stats is done if you know ALL units you take to battle. The only question left is whether the units brought to battle are determined based on nation boost only or both nation and building boosts.

Unfortuneatly I will have to upgrade my airship hangar 4 times to determine this...


I can partially answer this question. I've been tracking my stats consistently since I started playing, however, being US, I've never had to factor in the nation boost. Only two of my boost buildings are at the same level so it's been relatively easy to see which units are going to battle when I have multiple units with the same raw attack stat.

Raw attack stat not your building boosted stat determines which unit is brought to battle. So for example, I have a bunch of B-52s (raw att 30) and a handful of Hardened Marines (raw att 31). My air boost is 6% and my infantry boost is 2%. Boosted, the stats are: B-52 - 31.8, Hardened Marine - 31.62. Even though I have more than enough B-52s to replace the Hardened Marines, the HMs always go to battle.

When different units have the same raw att stat, there's an order in which they are taken to battle that I haven't taken the time to figure out. Sometimes the units with the 2% boost go to battle and sometimes it's the units with the 6% boost. Whenever my spreadsheet has been off by 3 or 4 points, this is usually where I find the discrepancy.

[albeezy]

I can partially answer this question. I've been tracking my stats consistently since I started playing, however, being US, I've never had to factor in the nation boost. Only two of my boost buildings are at the same level so it's been relatively easy to see which units are going to battle when I have multiple units with the same raw attack stat.

Raw attack stat not your building boosted stat determines which unit is brought to battle. So for example, I have a bunch of B-52s (raw att 30) and a handful of Hardened Marines (raw att 31). My air boost is 6% and my infantry boost is 2%. Boosted, the stats are: B-52 - 31.8, Hardened Marine - 31.62. Even though I have more than enough B-52s to replace the Hardened Marines, the HMs always go to battle.

When different units have the same raw att stat, there's an order in which they are taken to battle that I haven't taken the time to figure out. Sometimes the units with the 2% boost go to battle and sometimes it's the units with the 6% boost. Whenever my spreadsheet has been off by 3 or 4 points, this is usually where I find the discrepancy.

I don not think that is conclusive with regard to not including building boosts when choosing units that go to battle although it does seem to support the notion that defense stats break ties with units having the same attack. With regard to the 31.8 vs. 31.62, they would both end up boosted at 32 because it rounds the values, so relatively speaking they have the same boosted attack stats based on building boost. The HMs might make it into battle over the B52s not because of the higher raw attack, but because of the higher defense stats of 35 vs. the B52's 17. (not necessarily simply because of the raw attack). To eliminate the use of boost buldings you need the scenario I mentioned in post 19. Also, if you upgraded your airship hangar twice so that the B52s had an attack of 33 after building boost while the HMs stayed at 32 and yet the HMs made it into battle this would also prove building boosts are not considered

[John Snow]

I don not think that is conclusive with regard to not including building boosts when choosing units that go to battle although it does seem to support the notion that defense stats break ties with units having the same attack. With regard to the 31.8 vs. 31.62, they would both end up boosted at 32 because it rounds the values, so relatively speaking they have the same boosted attack stats based on building boost. The HMs might make it into battle over the B52s not because of the higher raw attack, but because of the higher defense stats of 35 vs. the B52's 17. (not necessarily simply because of the raw attack). To eliminate the use of boost buldings you need the scenario I mentioned in post 19. Also, if you upgraded your airship hangar twice so that the B52s had an attack of 33 after building boost while the HMs stayed at 32 and yet the HMs made it into battle this would also prove building boosts are not considered

One observation on the rounding issue is that I don't round until the att stats from all units are first added together. So the six Hardened Marines are factored into my total att stat at 9.72 and my 106 B-52's are factored in at 3370.80. I calculate the sum of the units first before rounding. This consistently gets to my exact profile att stat. My spreadsheet isn't off by a point or two, so there's no "fudge factor" for the rounding.

I'll give my airship hangar a couple of bumps to get it to 10% to see if that changes the mix. It'll probably take a week since I just started a 72 hour upgrade this AM.

[MadManDK]

OK, I've had some time to look at this again and Gree has changed things around.

I read somewhere that they fixed the raw stats, so that all players saw the same raw stats independent of country. This seems to be true, as their fix was to remove the country bonus form the raw stats. So the raw stats for me are now the sum of my units taken to battle without country boost (and without building boost, which is always has been).

However they seem to have fixed the 3% boost from my Warlord Siege Tank (+3% Alliance Attack) which now shows up in my raw and boosted stats.

So raw stats for me are the sum of unites doing battle with boost from the Warlord Siege Tank (WST). However the sort order when doing battle is still based on country boost (probably different code and not changed).

So with this my old conclusion that the units doing battle where dependent on my country boost, seems void, which my calculations support. So I'm looking at identifying my units into battle based on raw stats purely and I seem to have a difficult time, as I'm missing the criteria to select units with equal attack or defense score. No matter how I do it (look at the other score or based on something like ground before sea), my boosted stats are always few points off on either attack or defense, while the other is 0 (out of 70-80k). So any suggestions here?

Calculating my boosted values I've used a number of methods and the two that provide the best result (still have the issue just mentioned) are the following (they are mathematically the same):

1. Take each unit into battle and calculate its attack or defense score and the multiply (do not add) bonuses on one at a time. For instance 3 x HM's attack = 3*31*110%(country boost )*120%(boost building)*103%(WST). Then sum up the numbers and round.
2. Or sum up all units into the four categories, then apply boosts per category, sum up the results and round..

So the 2 questions i have are:

A. Anyone else have any ideas on how equal units are selected? Please provide me ideas (PM is fine) and I'll try them out.
B. Is the above only valid for me or is this how it works for everyone?

Here [1] is the sheet I'm using with all my data, if you feel like looking at it to solve the question of how units are selected for battle, or if you want to add your own units and see the result.

[1] http://www.filedropper.com/modernwartest_1

[Lawyer]

I will let all you figure out the math... My head hurts.lol

[Crime City Mark]

snip

That's some really in-depth testing.

[albeezy]

That's some really in-depth testing.
Thanks! So what do you think about my results?

[John Snow]

I don not think that is conclusive with regard to not including building boosts when choosing units that go to battle although it does seem to support the notion that defense stats break ties with units having the same attack. With regard to the 31.8 vs. 31.62, they would both end up boosted at 32 because it rounds the values, so relatively speaking they have the same boosted attack stats based on building boost. The HMs might make it into battle over the B52s not because of the higher raw attack, but because of the higher defense stats of 35 vs. the B52's 17. (not necessarily simply because of the raw attack). To eliminate the use of boost buldings you need the scenario I mentioned in post 19. Also, if you upgraded your airship hangar twice so that the B52s had an attack of 33 after building boost while the HMs stayed at 32 and yet the HMs made it into battle this would also prove building boosts are not considered

I can now safely say that the building boost is NOT a factor when units are chosen to go to battle.

Hardened Marines and B-52s are the units with the lowest attack stat that I bring to battle. The B-52 has a raw attack stat of 30 and the Hardened Marine has a raw att stat of 31. I just completed my Airship Hangar upgrade to L5. The boosted attack stat of my B-52 is 33 (30 raw + 3 boost). My Genetics Lab is still L1. The boosted attack stat of my Hardened Marine is 31.62 (31 raw + 0.62 boost). Despite the B-52 having a significantly higher boosted att stat, my total attack stat shows that all 6 of my hardened marines are brought to battle before B-52s. I "over bought" B-52s to double-check that surplus B-52s wouldn't replace Hardened Marines and they didn't.

I'm consistently able to match my spreadsheet with my profile att and def stats. PM me if you want a copy. Note that it's not a polished product.

[albeezy]

I can now safely say that the building boost is NOT a factor when units are chosen to go to battle.

Hardened Marines and B-52s are the units with the lowest attack stat that I bring to battle. The B-52 has a raw attack stat of 30 and the Hardened Marine has a raw att stat of 31. I just completed my Airship Hangar upgrade to L5. The boosted attack stat of my B-52 is 33 (30 raw + 3 boost). My Genetics Lab is still L1. The boosted attack stat of my Hardened Marine is 31.62 (31 raw + 0.62 boost). Despite the B-52 having a significantly higher boosted att stat, my total attack stat shows that all 6 of my hardened marines are brought to battle before B-52s. I "over bought" B-52s to double-check that surplus B-52s wouldn't replace Hardened Marines and they didn't.

I'm consistently able to match my spreadsheet with my profile att and def stats. PM me if you want a copy. Note that it's not a polished product.


That's some really in-depth testing.

What he said. Sounds good!

[MadManDK]

So one more person to verify the boost building, nice work. But still missing all the other details on units are selected.

[albeezy]

So one more person to verify the boost building, nice work. But still missing all the other details on units are selected.

You referring to when the raw attack and nation boost attack are the same?

[MadManDK]

You referring to when the raw attack and nation boost attack are the same?

Yes, I'm currently in a "state" where I cannot find a sorting that makes both my attack and defense numbers line up. If I do the sorting with country boost, then it's impossible, but if i do it without country boost and use 2 different methods for attack and defense (for equal attack i use defense numbers as second sort; for defense i use infantry before ground before air before sea), then I can. That seems to imply that the sorting is based on some order of units I can't find (not even in the plists). My mix is slowly changing, so the problem might soon be "gone", but I captured the data in the spreadsheet linked in my earlier post.