This topic is closed

Loss after collective defence

15 Replies
albert
18 November, 2016, 4:44 PM UTC

Hi everybody !


I'm slightly surprised by a recent battle result we had with a friend.


We made a collective defence, both of us having sent 6M of def. Before the attack, I use a +20% boost and parangon (lvl 9, +5% boost). As a result, he had 6M def and I had 7.5M def.

He lost 2.8% of his troops and I lost 3.45% of mine. I thus lost 25% more troops...


I would have imagined that the outcome of the battle is determined as follows:


Our def vs the attacker off and a fraction of loss computed as a function of (total def)/(total def+total off).


Then, we should have lost a percentage of our `apparent def', say 3%. This would have represented 3% of my friends units and (3/1.25)% of mine.


But this is actually the opposite which happened so I expect that what happened is the following: a fraction of loss is computed as a function of (total def)/(total def+total off). Then it is applied to our total army to determine the total number of each units that die. The ratio of losses among us is a proportional to our share of the def. Mine being bigger, I got a bigger share of the loss.


If that is indeed the case, then it is really a rubbish algorithm, since using boosts makes you lose more troops ! Plarium, please fix this !!!


Alb

UTC +1:00
2
Reaper
18 November, 2016, 5:07 PM UTC

its the second if you have more there, you lose more cos the mix of your troops is like that. For example take 5 white rabbits and put them in a group of 50 black rabbits and run trough the group you will hitting more black than white rabbits. The same with the defence if your def stat is higher you take the most dmg cos you are the main part (black rabbits) and the troops of your friend are the white rabbits. After you increased the stats it was like sending more def to defend (like putting +20 black rabbits more in the group).

For the enemy troops it was like you who runs trough the mixed rabbit group.
UTC +1:00
0
albert
18 November, 2016, 6:46 PM UTC
I understand, reaper, but I think it is just a poor way of deciding on the loss. This makes boosts violently counterproductive in collective defence and I do not think this was intended. Just poor programming.
UTC +1:00
2
Reaper
18 November, 2016, 6:52 PM UTC
i know what you mean you boost that you loose fewer and in total you lost more at the end. If it is a kind of poor programming i cant say to you im not the developer. But this is how i can explain that you lost more even with boosts
UTC +1:00
0
IronApex Turok
18 November, 2016, 9:08 PM UTC

albert said:


I understand, reaper, but I think it is just a poor way of deciding on the loss. This makes boosts violently counterproductive in collective defence and I do not think this was intended. Just poor programming.

nah it is working exactly as it was intended.   now  you have to buy more troops to get your defense back. 


people stopped defending in this game a year or two ago.  plarium changed the mechanics and its pointless to defend now. 
UTC +0:00
1
NIKITA METEOR
Administrator
21 November, 2016, 3:56 PM UTC

albert said:


Hi everybody !


I'm slightly surprised by a recent battle result we had with a friend.


We made a collective defence, both of us having sent 6M of def. Before the attack, I use a +20% boost and parangon (lvl 9, +5% boost). As a result, he had 6M def and I had 7.5M def.

He lost 2.8% of his troops and I lost 3.45% of mine. I thus lost 25% more troops...


I would have imagined that the outcome of the battle is determined as follows:


Our def vs the attacker off and a fraction of loss computed as a function of (total def)/(total def+total off).


Then, we should have lost a percentage of our `apparent def', say 3%. This would have represented 3% of my friends units and (3/1.25)% of mine.


But this is actually the opposite which happened so I expect that what happened is the following: a fraction of loss is computed as a function of (total def)/(total def+total off). Then it is applied to our total army to determine the total number of each units that die. The ratio of losses among us is a proportional to our share of the def. Mine being bigger, I got a bigger share of the loss.


If that is indeed the case, then it is really a rubbish algorithm, since using boosts makes you lose more troops ! Plarium, please fix this !!!


Alb

Hi! If you think that this tactic is disadvantageous, you may use another one. Unfortunately, we don't disclose the battle calculation formula.

Community manager
UTC +2:00
0
IronApex Turok
21 November, 2016, 9:02 PM UTC

NIKITA METEOR said:




Hi! If you think that this tactic is disadvantageous, you may use another one. Unfortunately, we don't disclose the battle calculation formula.

or will they?

here it is.

http://stormfall.playtamin.com/

UTC +0:00
2
BiohazarD
Moderator
22 November, 2016, 8:05 AM UTC
IronApex Turok said:

NIKITA METEOR said:




Hi! If you think that this tactic is disadvantageous, you may use another one. Unfortunately, we don't disclose the battle calculation formula.

or will they?

here it is.

http://stormfall.playtamin.com/

That formula/tool is not officially supported by plarium, and they do not guarantee that it is accurate. 
Any opinions expressed by me are my own and do not necessarily represent the opinions of or constitute official statements by Plarium.
UTC +0:00
0
albert
22 November, 2016, 3:46 PM UTC

NIKITA METEOR said:


albert said:


Hi! If you think that this tactic is disadvantageous, you may use another one. Unfortunately, we don't disclose the battle calculation formula.


Hi !


I of course will not use it again but my comment was mostly that having a BOOST option for defence that INCREASES YOUR LOSS is broken, not "disadvantageous". It would be much smarter to estimate the loss as a percentage of the def used (rather than as a percentage of troops used), and then apply it taking the boosts into account. With the same overall losses, the splitting between the defenders would be much more fair.


Alb




UTC +1:00
1
Reaper
22 November, 2016, 3:53 PM UTC
Here you are far away from fair. A lot things dont work how they should or could and that you see day by day now.
UTC +1:00
0
Aaron KT3
22 November, 2016, 10:14 PM UTC
There are other reasons too.  Relative Lost Arts & other bonuses from Hero & Dragon, etc.
Aaron
UTC +0:00
0
Sifr
23 November, 2016, 12:33 AM UTC
You lost more units but you also kill more units and in turn more pvp points for you.
Fighter Sifr of Fellowship
UTC +0:00
0
toggit
23 November, 2016, 9:00 PM UTC

NIKITA METEOR said:


Unfortunately, we don't disclose the battle calculation formula.

Really?

So what is this supposed to be? Just another plarium lie?

http://prnt.sc/bgrn44

"unmistakingly predict the outcome of an important battle" or complete horse-feathers?


Every victorious player will suffer EXTRA casualties, up to 9% extra
It is a stealth tax on victory

Maybe you should all read this :-

http://plarium.com/forum/en/stormfall-age-of-war/-news-and-announcements/29976_ask-the-developer-4/?post=186946


Maybe then somebody can explain why offensive losses were higher than defensive losses here  http://prnt.sc/d8xkl5

and don't give us any garbage about boosts and castle walls, they are already included in total off / (total def+total off)

UTC +0:00
2
albert
24 November, 2016, 4:55 PM UTC

Hi Toggit,


I confirm that this formula is wrong. I have done tests with friends, putting units in castles whose bonuses we knew and checking the relationship between the fractions of off or def over total forces on the one hand and the outcome on the other hand, and it is clearly not the formula found at the link you gave.


Best,


Alb
UTC +1:00
0
Ryu
24 November, 2016, 8:33 PM UTC

Well, maybe you can find me the point where you entered castle defense in that calculator (the playtamin one). Because I do not see that option, and it matters, much.


Second, bigger army takes bigger share of losses and pvp points. Try putting the same effective defense in that your friend has without booster. Compare losses on a unit basis.


Also, given the total ugly loss ratio an attacker can suffer compared to the defenders in a castle with high defensive bonus, nothing on the pvp points side of things can repay for the loss in troops. Do the math for 5m with 300% def bonus vs. 20 million off. Then find me how much offense you need so that the losses (after infirmary) are equal. After that, figure out what happens if a league mate puts 5m defense next to yours. Finally, calculate rebuild times with four chains running. Is really eye-opening. CREDIT CARD DOES NOT COUNT FOR FAIR COMPARISON. IS PAY-TO-WIN GAME.
UTC +2:00
1
albert
26 November, 2016, 2:55 PM UTC

Ryu said:


1/ Second, bigger army takes bigger share of losses and pvp points. Try putting the same effective defense in that your friend has without booster. Compare losses on a unit basis.


2/ Also, given the total ugly loss ratio an attacker can suffer compared to the defenders in a castle with high defensive bonus, nothing on the pvp points side of things can repay for the loss in troops. Do the math for 5m with 300% def bonus vs. 20 million off. Then find me how much offense you need so that the losses (after infirmary) are equal. After that, figure out what happens if a league mate puts 5m defense next to yours. Finally, calculate rebuild times with four chains running. Is really eye-opening. CREDIT CARD DOES NOT COUNT FOR FAIR COMPARISON. IS PAY-TO-WIN GAME.

Hi,


I agree with both your points but I just think that there is a smarter way to implement how boosts work for 1/.


Regarding 2/, attacking a castle with an off only 4 times the def, especially if there is a defensive bonus is suicide indeed in terms of rentability. But no one forces you to do that :-) With an offensive army around 15M, you can make 2k of pvp points with barely any losses. And this is clearly worth it. PvP needs time and preparation if you want to make a profit. If you just want to have fun, then suiciding some of your units is always worth it, of course ^^


Alb



UTC +1:00
0
1779696 users registered; 48304 topics; 286923 posts; our newest member:Castle 1350829