Analyzing and Correcting the Ease of Locking Land in 3.0

[taekwondokid42]

I'm going to start this post by saying that I don't mean to offend anyone. Everybody has been trying to make 3.0 the best game possible, and there are a lot of good things that happened. But people don't care about the things that work, they care about the things that don't work, and most of my posts have been about that. So I just wanted to say that I think a lot of interesting and good things have happened, but I intend to focus on the weak stuff with the hopes of making it stronger.

Having said that, there's a lot of discussion about weaker empires vs. stronger empires, and the difficulty with which you can move up the ladder. There's been a lot of attempted fixes too, such as diminishing returns on things like the amount that you can store in the bank. But I think that these complicated approaches to the problem are the wrong way of going about it, and that it's almost always to have simpler equations and simpler game mechanics. And I'd like to say that simpler game mechanics don't mean a simple game, simpler game mechanics means it's easier to figure out how each piece works. If there are enough pieces that fit together in enough different ways, it won't matter how simple each piece is, together they will have enormous potential for complexity.

Ease of locking land is closely related to the land flow in the game. If every person usually acquires 90% of the land in the game every run, then the chances of someone being able to hang onto 95% when they finally make their big move is much smaller; everybody [should] have lots of resources to take them down early, even if they have to make a gimped run to do it. Additionally, the person who usually clings tightest to their land is also the person who usually has the easiest time locking land; people learn that attacking them is usually a larger effort and prefer to wait for someone else to do it for them. Why waste 100 turns and half your marketed troops when someone can do it for you?

So in general you want a situation where land is moving freely. One thing that's different about 3.0 is the lack of an 'increase attack opportunity' button. Because of this, someone who has just run (and is unclanned) can only lose about 50-60% of their land, something that really decreases land flow. It also makes unclanned emps substantially harder to break, because even if you get through once they'll STILL have just as much land as you, and they'll be maxed. Getting enough resources to do a take down attack on an emp usually (and should) takes multiple runs, and so the likelihood of doing 2 in a row successfully is very small.

The other thing is that an emp used to have to defend against attacks from 3 major fronts and 2 minor fronts: rats, stoats, and leaders for the major fronts and skiffs and weasels for the minor fronts. In 3.0 an emp hardly has to worry about leaders: sevs was on 30% leaders and DogCat was on 150+% leaders, yet DogCat still couldn't pull off an espy. Furthermore, an emp cannot lose land to leader attacks, though health sabotages and murders are still useful. But now with towers, the emps position becomes even easier to hold! The 4 troop gaps begin to merge into a single more easily defended fortress. Instead of needing 2x the rats of your opponents you can just keep more skiffs or weasels around and reduce that number to needing about the same number of rats as your opponents. This reduces the stress you'll need to bear to deal with the food costs of having a large rat army.

Another thing that makes locking land easy is aid: you can aid all of your troops to an ally, make your run on a massive amount of land with little costs, and then have all of your troops aided back to you.

So I have a bunch of mostly simple ways to make locking land more difficult, and make land flow throughout the game better overall.

1. Reduce the number of defence points a stoat has. Right now a stoat is 7 offence 5 defence. Reduce it to 4 defence and suddenly it's 25% easier to break with stoats. This means that the emp will need more stoats.

2. Increase the number of attack points a stoat has. 8 offence vs. 5 offence will still make a large difference, forcing the emp to get more stoats to withhold attacks against opponents.

3. Make towers more expensive to own. IE reduce the number of workers that move onto land with towers.

4. Make towers less effective at having troops defend each other. Instead of maxing out at 50% boost to each troop type, have it max out at 25% and grow slower. This forces someone trying to emp to have more rats and stoats, but it especially forces them to have more rats. Rats eat a lot of food, and if the emp needs to hold twice as many food-eating-monsters as you do,  the emp is going to have a harder time keeping his spot at the top.

5. Make rats eat even more food. Make them painfully expensive to hold onto. This will work best if used in tandem with forcing emps to keep more rats around.

6. Weaken aid. I don't know if this is how it currently happens, but make it so that the turns are taken before the aid is sent. Force whoever is aiding their troops away to deal with their expenses for 2 turns before being able to lighten the burden.

7. Lower the amount of aid that can be sent per package. Instead of 50% cash and 20% everything else, set the limits at 25% cash and 15% everything else. This will keep emps from unloading their goods as fast, keep them from unloading as many goods, and force them to deal with the expenses of the goods for longer. Unclanned warlords will still be able to aid away more than 60% of their goods in 12 turns.

8. Bring back 'increase attack opportunity,' to allow everybody to run more often on greater volumes of land.

9. Up the attack limit maximums. Bring it to 25 instead of 21. Or maybe even 27.

10. Increase the amount of land that gets lost in an attack. I think it's currently around 3% (which means over 21 attacks you would take on average 48% of their land), so raise it to 4.5% (which means that you would take on average 62% of their land)

But to me the biggest ones are the ones that make the land more fluid as a whole. Land will be harder to lock down if it in general moves around with greater ease. That means changing stoats so they have less defence points and are harder to defend with, and changing attacks so that an assailant can take 60-70% of someones land instead of 50-60%. Either do this by raising the attack cap to 25, or by bringing back 'increase attack opportunity', or by increasing the amount of land you take each attack.

[Firetooth]

Without people running leader strats, nobody has enough leaders to stop anybody taking the land, going 30% huts and making enough troops. It is extremely expensive to produce enough leaders on the remaining land to do damage, so people don't bother.

Shadow did mention adding a buff which kills leaders, though, which sounds pretty cool. He doesn't have any time to code atm though.

[Shoot]

Though I have to disagree with you on a few points, this was a very good post. My thoughts on some of your points:

#1-2 - Taking defense points away from any unit at all (or adding offense points) would (theoretically) force lockers to favor that unit more (taking away from being able to defend other types), or risk being too weak in that type. It wouldn't necessarily have to be limited to stoats, but even the slightest adjustment could prove helpful here.

#3 - I like the idea, but with current game mechanics, I really don't see this having much effect. People can quickly demolish a good chunk of their production buildings at the end of their run and build towers. Likewise, at the beginning of their run they can demolish the towers from the previous run and stick them back into production buildings for their run, so penalizing people for owning them won't really do much as they only actually have towers for no more than 15-20% of their run.

There would need to be something in place to make it harder to demolish buildings as fast, and a change like that would have other ramifications that would need to be considered before implementing.

#5 - This would most likely have the opposite effect (that is, making it easier to lock). While being struck, the defender isn't having to pay upkeep on their rats, while the attacker has to keep a considerable amount of rats on hand for a great deal of turns. The locker would most likely be ditching their rats as soon as possible when running, so they won't be feeling the effects of a consumption increase too much while someone seeking to break the lock with rats is going to have a hard time coping with the extra food loss.

#7 - With the proposed numbers, what would take 8 shipments to aid out would take 12, which I don't think would have enough of an effect (around 65% or so increased upkeep to aid back out, then free to run with low troops/upkeep after).

#8 & 9 - Agree completely. I'd even say increase the rate in which the hit limit is lifted (3 per hour at least as opposed to the current 2 per hour). With the fast pace of the game, it is all too common that the vast majority of viable land targets have reached their hit limit as a result of people who ran shortly before you maxing them all out resulting in very little for the taking (or waiting a few hours to run which is quite inconvenient).

#10 - Personally, I think the land average is fine as it is, especially if some hits get added to the hit limit. The variation in land taken, however, I feel is ridiculous. One hit you could get 3000 acres, and the next you can end up with 12 acres. A bad streak of hits on a big land target can easily turn what could have been a good run sour. Personally I would suggest the minimum gain to be 33% of the current maximum, and the maximum be 67% of the current maximum which would result in the same average but less variation.

[Shadow]

this took like an hour to type...
1-2: troops were balanced so average totalmop and dp output per barracks per turn would be independent of training settings. chaging one number mesns changing them all. I could redo them with dp eeighted above op, but I'd like to exhaust simpler routs first.

3.people dont run with towers, so workers held and cost don't matter. we could make them take longer to build, but that seems kludgy.

4. towers are easy totweak, but single buggy round isn't enough to convince me it's necessary. also recall that lockers don't usually run with much army, so upkeep costs are less of a factor then you seem to think.

5. shoot nailed it, also see 4. again, troops were balanced so thaat one barracks-turnworthh ofoutputt consumes he same cash-value of resources (assuming $15 food) regardless of training settings. changing one number means changing all of them.

6-7:2 extra upkeep turns wont chanhe anything. I don't see aid as much of a problem tbh. we coild bring cash aid in line with the rest, though. as long as we don't have anonymous market, people can get around aid (more on markets later when3n I can code, i have ideas).

8. I hate that spell, its terrible and completely destroys he possibility of balance between leader and nonleader play. I will introduce an attack buff that kills leaders and towers, though, to make unlockinf easier.

9-10: we tried high attacks limits, it was awful. I encourage you to read the 3.0 themed posts in turbo forum to get an idea of failed experiments like this over the last 2 years or so. high land floe actually makes locking easier, other things being equal, because it makes it easier to quickly land starve opponents. that being said, I am all for faster unmaxxing, since this solves the maxxed problem shoot mentions without increasing land flow on a run by run basis.

[taekwondokid42]

I will admit, 5 was a bad idea. Perhaps instead you reduce the amount of food needed to hold onto rats. But right the problem is really obtaining enough rats more than holding onto them, so I don't think that lowering the amount of food they eat will help. Plus I just plain like the mechanic that rats eat enormous amounts of food. No idea why but it's always felt like 'the right thing.'

Right now, assuming you are trying to defend against attacks while locking land, you'll get about 61op for every 57dp (there's a lot of math behind this, which involves protecting yourself equally from all 4 types of attacks and assuming that your opponent knows exactly where your weakness lies). This means things are weighted towards offence by about 7%, except that because of towers all races get something like a 20% boost from the other races, and so things are actually weighted towards defence by a substantial margin - as much as 15%. That means that as an omniscient attacker who is presumably running on about 20% the land of your opponent, you are going to need to gain troops at somewhere around 30% the rate of your attacker, and you are going to have to hope between runs that the team locking land doesn't realize what troop you've chosen and starts to beef it up more relative to the others.

I seem to remember some tradition of always keeping the total number of op and dp the same regardless of unit comp, but now I'm questioning why it's needed. If an offencive troop comp gets you more total points than a defencive troop comp, doesn't that mean that land is going to flow more freely and emps are going to be easier to break? I think that this is the ultimate weapon against people who lock land, especially if you are very reluctant to remove/underpower the buff you gave to towers.

But if you are determined to also keep total war points even regardless of troop composition, then I can help make the numbers turn out correctly. Here are some potential alternatives that make defence harder and offence easier:

rats are base, ( for total )
weasels are 2x rats,
stoats are 4x rats,
skiffs are 6x rats.

so if rats are 3 points you get 6, 12, and 18 points respectively.
if rats are 7 points, you get 14, 28, and 42 respectively.

Potential Alternatives:
Unit:op:dp
R:2:1
W:2:4 (1:5)
S: 8:4 (eep)
K: 8:10 (9:9)

R: 5:2
W: 4:10 (5:9)
S: 17:10
K: 21:21 (20:22) (19:23)

of course then you'd have to change anything else that relies on op and dp (I think that's just the towers base dp but you would know better) to be 2.33 times more, because under this setting we've dramatically increased the total number of points in the game.

But I do think that tweaking the offence and defence points of the different types of troops is the best way to stop people from locking land.

[Shadow]

it wasnt always balanced like that, andnit isn't on reg... rats there cost twice as much as stoats and skiffs per barracks turn, which is why nobody uses them solo there. it only makes sense to do it that way imo. but we could tweak thigs to assign higher value to defensive troops so as to bring the edge between o and d closer to zero

[taekwondokid42]

The point was that there should be a gap. If you don't want people locking land, you'd have to change the troops so that defence is inherently weaker. I was thinking like a 15% or 20% gap in favour of offence instead of only a 7% gap in favour of offence.

edit: completely forgot the second half of my post:

I was thinking that we should actually reduce the crazy defensiveness of weasels. Nobody in the history of forever has ever broken someone locking land by massing weasels. Maybe it has happened once but it's never really a consideration. What if you reduced weasels to a 2-3 instead of a 2-4? I still don't think people would ever consider using weasels but it would change the math up and make locking land at least theoretically harder.

[Shadow]

the point of weasels is not to break, they are to provide cost effective defense against standards and kills at he like.

also, I never said locking land was bad. we just need to make it hard enough.

[Shadow]

please don't take argumentation to mean nothing will gt done btw. people around here rarely stick around long enough to defend th3ir ideas and take argumentation as an attack. This sort of debate is how we separate good ideas from the rest. if you think its good, argue with me so we can close the theory loopholes before spending code time.

[taekwondokid42]

No I understand. I am putting my thoughts out so people will be reading them and thinking about them. Everything is a rough draft.

I'm sure you'll hear more about it as the round continues, but right now nothing terribly interesting is happening (at least not from the scores page, heh heh). I have a post I <intend> to make soon but I just moved and my new house doesn't have internet, so I'm stuck posting from the library for now.