Attacking - Gain in War and Offense to Send

[RiffArt]

OK, so I have two questions that the guide doesn't cover.

According to the guide, if the relative province networth is <56.7% then attack gains are zero. I've seen this happen. However in war, it seems gains might be capped (minimally) at 5 acres.

Question 1: How are attack gains different in war and out of war?

i.e. is it just that you always gain 5 acres during war no matter the nw ratio, or are the gains higher across the board for hitting out of range?

Random factor is +-3.5% for offense and defense. Off/def ratio needed to win is 0.97

Now I know that this random factor is out of date and if you send even slightly more than 100% the target's defense you win.

Question 2: If I send 97% of the target's defense, will I win as the guide suggests?

[Palem]

Question 1: How are attack gains different in war and out of war?

In war there's basically no GBP (it's capped at 20% so it's really not all that noticeable), there's also a minimum cap on gains, as you've noticed.

.e. is it just that you always gain 5 acres during war no matter the nw ratio, or are the gains higher across the board for hitting out of range?

Yes. Minimum cap is 5 acres.

Now I know that this random factor is out of date and if you send even slightly more than 100% the target's defense you win.

Question 2: If I send 97% of the target's defense, will I win as the guide suggests?

If you want a 100% guaranteed success, yes you should send at least 103.5% of your opponents defense, but technically you can break a target by sending slightly less (97%) of your opponents defense thanks to the random factor, but that is unlikely.

[Persain]

If you want a 100% guaranteed success, yes you should send at least 103.5% of your opponents defense, but technically you can break a target by sending slightly less (97%) of your opponents defense thanks to the random factor, but that is unlikely.

its 104.3% to break 100%.... i've bounced with 104.1 before...its rare but it can happen. That math is somewhere in the forums but im to lazy to find it. Other than that Palem answers question #2 nicely.

OK, so I have two questions that the guide doesn't cover.

According to the guide, if the relative province networth is <56.7% then attack gains are zero. I've seen this happen. However in war, it seems gains might be capped (minimally) at 5 acres.

Question 1: How are attack gains different in war and out of war?

i.e. is it just that you always gain 5 acres during war no matter the nw ratio, or are the gains higher across the board for hitting out of range?

The formula http://wiki.utopia-game.com/index.ph...etworth_Factor is wrong IN WAR...i played around with the math a little but was never able to get a working algorithm becuase people always use some % GS and i didnt want to survey before every hit. Needless to say though that gains formula isnt even remotely correct in war.

I've top fed into people 500% my size before and gotten like 5-6% land gains and to get 5 acres you need to be hitting a lot smaller than 50% your size. So i guess basic answer to question #1 is no one really knows :P

[2Utopian]

Yes. Minimum cap is 5 acres.

This is false, you can gain less then 5 on a hit in war. I saw gains of 4, 3, and 2 acres in a war this age.

[Palem]

Really? I'm pretty sure that shouldn't happen. Did they use -hours or something?

[2Utopian]

No, I have numerous hits from that war that produced sub 5 acres. All instances of this were spare hits on chained out provinces. I gained 4 acres on a few personally and - hours were not used on those for sure. The lowest gains were hits of 2 acres and that happened a few times over the course of the war.

[Palem]

The hell? Am I crazy or is there not supposed to be a min cap in war?

[pathetic sheep]

I have seen 3 acres many times.

http://wiki.utopia-game.com/index.ph..._%26_Defending

Mod Off = Mod Def * 1.040362694

its 104.3% to break 100%.... i've bounced with 104.1 before...its rare but it can happen.

They can receive soldiers and you can get errors in your intel. They can get online and cast spells. I often set up my attack page in one tab and then get a new SoT in another tab. I have seen it change even though my first SoT was only a few minutes old. I often get a bunch of SoTs and then pick a target. Then I get a SoM right before the attack so that the intel has useful information. Sometimes I find armies are out and I can hit with much less offense.

The attack page updates when the cursor is outside of the box. It can be slow to update sometimes. Sometimes computer connections are slow too. That is another reason to set up the attack page and then get new intel. If the server is really really slow you will not get the intel.

Edit:
If you are super lucky someone could mystic vortex a fanaticism spell and cause an attack to fail. Aggression too.

[darkl1ght]

I've had 1 acre attacks done on me in war, when i was at around 80 acres. So no, its not capped at 5 in war.

[Karazy]

Well if the minimum you could get was 5 acres then it would be possible to kill a province in war without pez killing them. Pretty sure it goes all the way down to 0 acre gains. I think the min gains thing is still % based so like everything else in uto it gets rounded down.

[octobrev]

It goes down to zero. When you hit out of range in war there's a minimum gain on 3(.5?)%. That's what allows calves to pull in 12+% gains on their 4x bottomfeeds.

[Persain]

It goes down to zero. When you hit out of range in war there's a minimum gain on 3(.5?)%. That's what allows calves to pull in 12+% gains on their 4x bottomfeeds.

3.6%

[RiffArt]

Thanks Persain and octo for the answer to Q1. 3.6% min gains in war and nw factor gains from guide is wrong in war.

Thanks people for the confirmation that random factor does actually exist. Saw a few posts say it didn't, tried it out myself a few times and must have just got lucky.

Now I did a bit more digging around and found this old post (sorry I can't link yet not enough posts):

What the devs said: (offense*(1+random(3.5%)-random(3.5%)))/(defense*(1+random(3.5%)-random(3.5%))) needs to be bigger than 0.97 to succeed with the attack.

Now this is very different to what I'd assumed: Offense*x/Defense*y > 0.97 where x,y are randomly (uniformly) chosen between [-3.5,+3.5]

Question 3: Is the random factor to offense/defense uniform or is it non-uniform (like in the quote)? (I realise this may not be known, but there might be other mathmos like me out there who've done the work)

[pathetic sheep]

Now I did a bit more digging around and found this old post (sorry I can't link yet not enough posts):

Now this is very different to what I'd assumed: Offense*x/Defense*y > 0.97 where x,y are randomly (uniformly) chosen between [-3.5,+3.5]

No. I think it is exactly like what you assumed. In spoken English: "Offense plus or minus three point five percent divided by defense plus or minus three point five percent must be greater than or equal to zero point nine seven". The quote is bad algebra but using characters +/- in a typed sentence confuses people because they assume the "/" is a part of the equation and means "divide" instead of an "or".

Question 3: Is the random factor to offense/defense uniform or is it non-uniform (like in the quote)? (I realise this may not be known, but there might be other mathmos like me out there who've done the work)

The developers are programers so the "random" is probably a "rand function". But I am just guessing. Technically that should be "96.5 plus a random rational number normally distributed between 0 and 7." It could, however, be "95.5% plus a random integer between 1 and 8". That would make sense if you had a table top game.

I doubt anyone is collecting good data. A lot of bad or debatable data shows up in the forums. No one who would care would then also believe your data. See the debate about Nature's Blessing spell.

The developers could switch from uniform distribution to normal distribution and not tell us. They could then switch it back next age or right in the middle of an age and not tell us that either.

In order to get the data on the distribution of the random factor you would need to have an age of failing a lot. Is it worth ages of failure in order to have a minor detail worked out? The NB debate can happen because people can cast it a lot and still keep attacking and thieving. People who calculate numbers rarely fail attacks and when they do fail they do not like posting the details in the forums. Failing NB 15 times is easy to track because your mana drops from 54% to 9%.

Did you fail the attack because of the random factor? A new SoT requires crunching numbers to subtract loses and add back gb protection. Adjusting for lost building requires before and after survey. A province can also change after the attack but before a new SoT.

[neanderthal]

Pretty sure that bishop has stated that the Magic number is 104.04 on a hit not to fail.