Doomstacks, causes and non-FTL solutions

[TheDeadlyShoe]

My main concern with the new fleet cap is UI tools. Balance is always fixable. ( -40% fire rate per friendly fleet in the system?) But I hope we'll be able to set up rallies so we can have a corvette fleet, battleship fleet, etc.

 

[Airowird]

If you want more POIs, why not have planet bombardment/occupation hit faction happiness. Planet full of materialists getting assaulted without you stepping in? Entire faction gets -5% happiness, decreasing for the next X years. (rough idea) Suddenly defense matters more and being able to attack enemy planets with quick strikes becomes a viable strategy to crash their economy.

As for fleet & doomstack imbalance. I've always been in favour of a dynamic penalty on an overwhelming fleet. A good 3D formation will leaves plenty of space (heh) for you to miss a ship, but shooting 500 guns at a single ship at once is nearly impossible, if only because a hit would actually nudge the ship a bit, just out of the path of the next incoming round. And before you talk about splitting fleets, I'ld base it on the totals in the combat screen. (with modifiers for defense stations ofc)
This would also reduce the size of the doomstack efficiency, because more ships in a fleet might realisticly not have any effect, so might as well use those for a smaller strike team, which means more need for defense systems, which then concludes the rock-paper-scissor model. It's basicly saying that only the surface of your rock counts, so at some point the surface-to-weight ratio is no longer viable for you to haul that thing around.

 

[Hawklaser]

Changing the AI to not play like that, programming it to split up and spread out its fleets, will have no effect on doomstacking. The player will still doomstack. If a game is unbalanced such that one strategy outplays all others in all situations then the problem isn't that the AI uses that strategy, it is the strategy in the first place. AI changes are completely irrelevant to that.

You have contradicted your own argument here. If the problem is that Paper needs buffs, then buff it, changing the AI is irrelevant to that.

Except you are still missing how the AI always playing one way warps gameplay. Even if the scenario was that rock paper and scissors were perfectly balanced for human vs human, but the AI always uses rock because the others don't work for it will change how everyone plays in general. Because as soon as the AI is introduced(majority of opponents in Stellaris for most people) Scissors becomes unable to win because the AI always uses it. Now for paper, thanks to the AI being unable to deal with human players and typically getting cheats so it remains a challenge, most common being a resource and map advantage, and in a warfare game that diminishes the effectiveness of paper against the AI.

So if you then buff paper enough that it offsets the cheats that the AI gets, it then unbalances paper in the human vs human front. And you have to hope it didn't become too powerful that it also beats scissors now too or you are back at square one of no choice except to play one way in order to win.

Stellaris needs difficulty, and unlike table-top or pen and paper RPGs it is a grand strategy game. That means that difficulty, challenge and forcing you to play strategically plays a bigger impact than an RPG. Having the AI play sub-optimally, and in this case play really badly, is completely contradictory to the nature of the game, you end up with less strategy not more.

In addition to difficulty, you also need meaningful choices. Part of the reason we are lacking meaningful choices is the AI always uses the same strategy, which the only counter for currently is do the same basic strategy as them but on a larger scale. Each AI type should not act the same in war as all the rest, they declare war differently, but they all conduct it the same way, so only have to ever deal with one style of fighting.

 

[Riftwalker]

While the devs may not have found it fun for the AI splitting and destroying all your mining bases simultaneously fun, though I would be willing to wager it was due to it being imbalanced in scale at the time coupled with the habituated play of required doomstacking, and the reason it was unfun wasn't because of the change of tactics needed, but the mindless rebuilding afterwards on a large scale

no it was unfun because it was essentially the same as doomstacks except everywhere. when you had to keep up with a ton of AI fleets, you'd have to pause regularly and figure out how and where to move your counter fleets. It was boring tedium.

you're still just ramming fleets into each other. it's just more tedious.

I've said multiple times that doomstacks aren't an issue because they limit strategy, they're an issue because it's 1 battle and then it's pretty much over. (not in this thread)

 

[Tsu Chi]

Do you really REALLY want the AI to split it's fleet? Wiz has stated multiple times making the computer perfect micro fleets is trivial, and would be utterly unfun, you declare war against an empire 1/10th your size, this should be easy.
They destroy all your outposts and mining posts in your empire.

In my opinion AI do not have to be perfect.
It has to:
· Simulate a human behavior – not many human can effectively control 10+ fleets (not talking about single ships that will join other fleets) making multiple strikes in different parts of the galaxy, but 3-6 fleets is totally manageable – this way you would not be able to counter it with only one doomstack.
· Make mistakes or different decisions as humans do – for each action AI should have some “probability to make other decision” and try to keep to that decision for some time. New decisions should take into account old decision.
· Wage goals – so that an order to destroy the home world of an enemy with almost no defenses left (just before ground invasion) was not countered with several frigates assault on some not important world (it is the most stupid thing AI does currently). It could ignore, build new ships, use other fleet that is close (and has some less important goal) or split a fleet and send several ships 2-3x stronger to destroy it.
· If planned action does not give effect (e.g. chasing enemy fleet) change orders.
· Don’t wait in a home system when facing a lot stronger enemy. It should harass enemy on borders, delaying strike as long as possible.

“If your enemy is secure at all points, be prepared for him. If he is in superior strength, evade him. If your opponent is temperamental, seek to irritate him. Pretend to be weak, that he may grow arrogant. If he is taking his ease, give him no rest. If his forces are united, separate them. If sovereign and subject are in accord, put division between them. Attack him where he is unprepared, appear where you are not expected.” Sun Tzu

When it comes to solving doomstack problem I invite you to read one of my posts.
https://forum.paradoxplaza.com/foru...lactic-terrain.1052958/page-229#post-23488791
​

 

[monsterfurby]

This reminds me of a debate I had about computer chess and complexity. In many cases, an equal trade is actually an advantageous move for the human player when matched against an AI (but not in human versus human play, making computer chess not a perfect environment for practice), simply because it reduces the complexity of the overall game - i.e. takes away part of the computer's home turf advantage.

So yeah, I do agree that having more complex fleets gives the AI an advantage. However, on the other hand, that's why making defensive play more viable is important. The AI may be able to implement perfect micro, but it is inherently inferior to a human at strategic thinking. A human can anticipate the AI's actions way better than the other way around, allowing them to properly build defenses to delay or even counter enemy offensives. Making battles last longer, thus giving humans more time to react, would also be quite beneficial, as would better AI for fleets on the tactical level (which is seemingly coming according to Wiz).

 

[ragehavoc]

One of the only things I can think of that would fix doomstacks would be to reduce fleets to single ships, that is you can only have one of each ship type, period, and then concentrate into research to make your ships more powerful.
The other option would be to add more tactics into battle, being able to change formations and the way ships attack.
I see no other option to fix mindless doomstacking, the way the game is setup is a pigeonhole into doomstacking, regardless of what FTL we use.

 

[Tsu Chi]

If you want more POIs

I have an idea about POI
why not make mining/science stations more interesting targets.
I always wondered how it is possible to only gather 2-4 rarely more resources from a planet/star compared to a medium colonized planet. Even whole system do not produce resources in such a scale as that one single planet
When the game starts each mineral counts. It is well balanced for 2-3 minerals a month at the start of the game. But later? It is so little. Why we can’t upgrade mining/science stations to 2nd 3rd or 5th level.
If it cast some money, took time (during upgrade you won’t get resources) but reward a player with additional amount of resources produced at each level. So a 1 lvl mining station produce 2 , then 4, 6, 8, 10 (as planet mining stations they would be researchable). Imagine level 1 that produce 6 minerals - then you would get 12, 18, 24, 30! Losing all resource stations in some uninhabited but mineral heavy system (which produce the same amount of resources as a fully developed planet) would hurt player a lot. This way each system becomes more important , and you have your POI in plenty
this of course would require economy rebalance and costs but is doable.

 

[Hawklaser]

no it was unfun because it was essentially the same as doomstacks except everywhere. when you had to keep up with a ton of AI fleets, you'd have to pause regularly and figure out how and where to move your counter fleets. It was boring tedium.

you're still just ramming fleets into each other. it's just more tedious.

I've said multiple times that doomstacks aren't an issue because they limit strategy, they're an issue because it's 1 battle and then it's pretty much over. (not in this thread)

So because the game would require more pausing to sort out a situation or two during war, the game becomes tedious? Are we not playing grand strategy games to have to figure out solutions to a number of problems including what may pop up in war? If pausing a little more often to plan bothers you, you might be playing the wrong game, and would recommend a true RTS like Starcraft instead. The biggest reason for a game like Stellaris to proceed in real time is to keep multiple people engaged in a multiplayer game. The other fully turn based ones tend to have people loose interest thanks to one person doing massive amounts of micro taking 20+m a turn compared to others taking around 5m tops. Or have you done a play by email version, where the game dies because end up with weeks between turns due to waning interest? Stellaris could become full turn based right now and it would have negligible changes to the overall game. The real time with anyone able to pause at will is actually a clever solution to the player interest problem 4x/grand strategy games have by allowing major things to be dealt with strategically, and yet keep the game from getting bogged down and players loosing interest.

I have an idea about POI
why not make mining/science stations more interesting targets.
I always wondered how it is possible to only gather 2-4 rarely more resources from a planet/star compared to a medium colonized planet. Even whole system do not produce resources in such a scale as that one single planet
When the game starts each mineral counts. It is well balanced for 2-3 minerals a month at the start of the game. But later? It is so little. Why we can’t upgrade mining/science stations to 2nd 3rd or 5th level.
If it cast some money, took time (during upgrade you won’t get resources) but reward a player with additional amount of resources produced at each level. So a 1 lvl mining station produce 2 , then 4, 6, 8, 10 (as planet mining stations they would be researchable). Imagine level 1 that produce 6 minerals - then you would get 12, 18, 24, 30! Losing all resource stations in some uninhabited but mineral heavy system (which produce the same amount of resources as a fully developed planet) would hurt player a lot. This way each system becomes more important , and you have your POI in plenty
this of course would require economy rebalance and costs but is doable.

This is something was considering as well, but just using tech to get more per station doesn't change up the value much. I'd think making the mining/research nodes rarer but drastically upping their value would have a better impact. But with the boarder/claiming changes coming that creates another problem of claiming worthless systems. Maybe 90% with current values, and 10% with much higher? Since normal is 2-3 and occasionally up to 9 due to anomalies, the high ones could be say 30-45 on average, with the possibility of 90-100 if get an anomaly as well? I could definitely see nodes in that range being majorly contested if near people's boarders.

 

[Riftwalker]

So because the game would require more pausing to sort out a situation or two during war, the game becomes tedious?

yes

it's doomstack issues x20 or so.

 

[ragehavoc]

yes

it's doomstack issues x20 or so.

you seem to be arguing we shouldnt have fleets period, it would be interesting to see your ideas of how to fix the issue.

 

[Riftwalker]

you seem to be arguing we shouldnt have fleets period, it would be interesting to see your ideas of how to fix the issue.

i prefer doomstacks over having 20 or so fleets wandering around with no mechanical changes.

I prefer HoI4 naval combat to the current system as well, as well as air combat.

the issue with doomstacks is they're a single battle and that's that. With starbases and the staggered retreat that's been hinted at, I don't think we'll have these issues anymore. if the mechanics remained the same, it'd just be a bunch of smaller fleets doing exactly the same stuff as the current doomstack, and that's aggressively rubbing your fleet against their fleet until you feel safe enough to start taking planets.

 

[The Founder]

The FTL and Starbase is not about solving Doomstacks. People have corectly noticed that if anything, those two things will make Doomstacks more present.

We have word of Wiz that they are working on Doomstacks. Both on twitter and in Dev Diary 93: "That's all for today! Next week we'll continue talking about war, on the topic of space battles, command limits and doomstacks. See you then!"
Maybe there is some tie in with Bases (like Fleets needing a starbase to opearte from or a Supply System).
Or maybe those two Systems are totally unrelated.

More likely I think starbases and FTL rework are about solving an issue that was created by solving Doomstacks.

no it was unfun because it was essentially the same as doomstacks except everywhere. when you had to keep up with a ton of AI fleets, you'd have to pause regularly and figure out how and where to move your counter fleets. It was boring tedium.

I am convinced the new FTL and bases Systems are about solving issues inherent in teh "Doomstack Solution" as well.

The AI is quite frankly superior to managing multiple fronts/fleets. The process is inherently Multitaskable. And in anything like that, the AI has a unbeatable natural advantage over players. It odes not need to be smart, if it can just beat us in the Attention game.
But with FTL rework and bases, the thing changes again. Suddenly we have distinct "theathers of war" between two Fortess Systems. Those areas in wich fleets can roam freely are more limited then "the whole galaxy" or "both sides empires".
Attacking such Starbase System would risk a lot of damage to your fleets, so if you fail the enemy has a opening to counterattack. But in turn you have to take the Starbase if you want to go any deeper into enemy territory - both because of FTL blocker and as a Repair Hub.

 

[Airowird]

I have an idea about POI
why not make mining/science stations more interesting targets.
I always wondered how it is possible to only gather 2-4 rarely more resources from a planet/star compared to a medium colonized planet. Even whole system do not produce resources in such a scale as that one single planet
When the game starts each mineral counts. It is well balanced for 2-3 minerals a month at the start of the game. But later? It is so little. Why we can’t upgrade mining/science stations to 2nd 3rd or 5th level.
If it cast some money, took time (during upgrade you won’t get resources) but reward a player with additional amount of resources produced at each level. So a 1 lvl mining station produce 2 , then 4, 6, 8, 10 (as planet mining stations they would be researchable). Imagine level 1 that produce 6 minerals - then you would get 12, 18, 24, 30! Losing all resource stations in some uninhabited but mineral heavy system (which produce the same amount of resources as a fully developed planet) would hurt player a lot. This way each system becomes more important , and you have your POI in plenty
this of course would require economy rebalance and costs but is doable.

While I'ld go slightly less on the multipliers (150-300% rather than 200-500%) so as to make a more tall vs wide thing and be a LONG investment, but it could have some uses.but I would make the last tier require a planet in the system just to make an interesting balance choice of terraforming & colonising 8-tile planets vs lower tier stations. Those tiers would ofc be linked to the relevant building tier, but it would certainly make certain systems more wanted, especially if the new star ports have some sort of resource gathering buff. The research station one would be linked to the lowest tier, or a new neutral one.


Unrelated note: As for the science stations goes, why don't we have a science building doing 1/1.5/2/2.5/3 in all rather than the X/1/1 system we have now? Heck, Why not just make those buildings be X+2 and have the neutral/combo one be a 4th choice? (neutral would ofc be a society research type because not only is it the least useful end-game, but it also serves as the cross-field research thing)

 

[Riftwalker]

as a side this is how i view the current change to warfare, and why I think it will be more enjoyable.

I think for the most point large wars will now consist of multiple war 'chapters' in that there will be the fleet fights in the 'no man's land' between the respective empire's choke points and starbases. After a side has obtained superiority in no-man's land, they then will rally and try to push through the choke point, with the defender rallying at the station to do the same. If the aggressor is beat back no man's land is up for grabs again, if the aggressor wins then the no man's land is pushed back and fighting over new deeper systems begin. The Previous choke point going to the aggressor and acting as their new choke point. They have to be careful not to let the enemy through other starbases they still own to recapture their rear however.

that's how i see the whole system coming together currently, with multiple fleets runing around no-man's land capturing points and engaging the enemy. multiple no-man's-lands might appear on a particularly large front as well.

with staggered retreats, players and AI will be able to put everything they can into battles and with forces split up fleets will be able to take multiple points in no-man's-land.

this allows a defender to rally for a defense and an aggressor to keep up the pace if they manage to win a critical choke point. IT all coming to a stop when the current loser can get the enemy to 100% war exhaustion.

 

[methegrate]

]

If you want more POIs, why not have planet bombardment/occupation hit faction happiness. Planet full of materialists getting assaulted without you stepping in? Entire faction gets -5% happiness, decreasing for the next X years. (rough idea) Suddenly defense matters more and being able to attack enemy planets with quick strikes becomes a viable strategy to crash their economy.

Not really, for the same reason that economic raids don't matter much: neither makes any difference to the outcome of the war. I can suffer through unhappiness penalties and mineral/energy losses in exchange for a better short-term military strategy.

When you think about how much someone has in terms of income and stored wealth, someone could take half my systems without impacting my ability to fight. Over a long time, sure, it would be a problem. But while you're hoping to starve me out, I'm taking your systems one planet at a time. Same with unhappiness.

In order to be effective, POI-bsaed warfare (imho) needs one fundamental statement to be true: By destroying something in System A, I affect his fleet's ability to fight in System B. In other words, targets which degrade the effectiveness of a fleet even if that fleet didn't participate in the battle.

To have happiness or resources do that, you'd need a complete economic rewrite such that unhappiness penalties or resource losses actually become relevant to fighting capacity. As-is a player who losses a bunch of systems can still easily deficit finance their war. You'd need to make it such that, at a reasonable number of losses, they will have to legitimately worry about whether they can still afford their fleet. Alternatively you could make it based on a supply or capacity mechanic as has been suggested elsewhere, so that if you don't keep certain stations alive, your fleet just can't fight.

Or perhaps a whole other option altogether. The central point being, though, this only works if your fleet's fighting ability is directly tied to assets in systems it doesn't currently occupy. Right now there's no direct link, and players are too rich for the economic links to matter.

 

[ElectricEel]

I think Gateways may become very significant points of strategic interest in Cherryh. They enable instant transportation between long distances, but you can't gate to systems that are controlled by an enemy, so having one of your Gateway systems seized may drastically lengthen your lines of reinforcement or even cut them entirely between two parts of your empire, while the opponent benefits from the converse if they or their allies have another Gateway in their own territory. In such a circumstance, you have to be prepared to commit some defenses to your Gateway or risk letting the enemy seize a really important system easily, and you won't be able to afford to have your doomstack sit on a single point of strategic interest the whole war in most cases. This recommends putting a fortress in such a system, but naturally, this has other strategic tradeoffs, such as having one less starbase to fortify your borders with (unless that's also where the Gateway is, in which case you almost certainly want to put that fortress on it).

Of course, this assumes that lines of reinforcement are actually relevant in the first place, rather than wars being decided by one side's doomstack wiping out the other's entire navy in a single battle (or in a series of battles where it crushes a series of smaller forces while suffering minimal attrition by the virtue of being a doomstack). Though if an enemy doomstack is busy elsewhere, a particularly swift offensive might be able to seize the Gateway before the doomstack can react, potentially trapping it in a suboptimal position.

I think fortresses, besides more production- and economy-focused starbases, will be strategically important as well, even beyond their role as blockers at chokepoints. It's reasonable to assume that given sides with equal resources, a fortress will be beaten by a main battle fleet[1] (as fortifying chokepoints would become a largely unbeatable strategy prior to the invention of jump drives otherwise), but a fortress supported by a moderate-sized fleet will beat a main battle fleet (because why even invest in expensive high-level fortresses if they don't provide an actual combat advantage).

This means that built-up fortresses will become safe harbors where fleets can shelter without fear of being destroyed easily, enabling those fleets to threaten surrounding systems and function as a fleet-in-being even if they're not strong enough to challenge an enemy main battle fleet by themselves. The opposition has to commit forces to containing such a fleet if they want to keep their territorial gains in the surrounding systems while their main force, which can't directly assault a properly-supported fortress without there being serious risks involved, moves on.

The difficulty in attacking such a fortress directly means that grinding down an opponent's economy becomes key if you wish to obtain total military victory, as you need to ensure that you outmuscle them by a sufficient margin that even their fleets cannot protect their fortresses. Combined with the new war exhaustion system, this is likely to lead to less decisive wars, but you always have the opportunity to try to seize territory from the enemy to bolster your margin of economic advantage so you can hope to have better luck in the next war.

[1] Which I will define as a fleet with a composition close to the maximum practical concentration of force given the game parameters. All doomstacks are also main battle fleets, but assuming the devs succeed in making a good dent in the doomstacks issue, not all future main battle fleets will be doomstacks.

 

[zizard]

The title is misleading because FTL is not necessarily a solution either. Look at a decent game like TW series. Higher travel time, higher upkeep, battle micro. First 2 naturally keep single army sizes down as empires expand and give attrition effects and purpose of static defence. Last one is much better than trying to build tactics into campaign map, which ends up being RTS micro with all tedium and no fun. Somehow people think it is good to have to play 20 fleet whack a mole against untiring AI, but will absolutely reject suggestion of tactical control in battles. Real time aspect is wasted in Stellaris since the whole point is to have economics interacting with battle timescales. The Stellaris economic timescale is at least 10x as long as the battle timescale, so there is no reinforcement nor even change in the fleet power in the time between a war starting and ending (except the AI completely fails to push wars quickly and gives unnecessary free time, but this is basically a bug).

 

[PK_AZ]

Obviously, in space, you don't really have fronts like a land war. But that doesn't remove the problem of defending your border. Now ships auto-enter combat if they get close enough, and cannot disengage except manually with emergency FTL. So a scouting fleet just hanging out in a border system is easy prey. Small fleets cannot do anything to large fleets.

You do realize fronts in land war are XIX-century invention? And they might actually be already song of the past? (with rest of quote I agree)

There seem to be three main reasons that have been identified so far as to why Doomstacks happen currently. 1 The AI doomstacks consistently so therefore the player must also doomstack. 2. Combat is essentially all or nothing (they look to be adding changes to this aspect) 3. No significant POIs besides enemy fleet, planets, and occasional chokepoints (primarily hyperlane related).

Seven. https://forum.paradoxplaza.com/foru...om-thread-redoox.991261/page-46#post-22815629

Well we're several games into Paradox developed games and stacks have pretty much been a persistent element in nearly all of them.

Actually no, EU4 (which I believe should be used as 'your standard this-generation Paradox-style GSG) is not so much about doomstacks, and MUCH more about manpower. Enough to say that after three succesful campaings against netherland rebels trying to take terriotry thatwere and justly should be mine (i.e. Netherlands) I was more or less mentally able to surrender.

And I suppose that's even mostly true in real-world warfare, if one doesn't account for supply and small-scale tactics.

Half-true. I believe Paradox-style GSG lack two important features of real world warfare: more than one attack vector and army ability to hold her ground against stronger enemy.
Lets imagine simple strategy game, where there are two bases (yours and enemy) and three paths between them. You and enemy have both 20 regiments you can commit to any path. Your goal is to get into enemy base before he gets into yours.
In that game you want to create local advantage at one path, and you want that local advantage to be higher than enemy local advantage at any other path. (for example 14;8;8 win against 10;10;10, because 14/10 = 1.4, 10/8 = 1.25 and 10/8 = 1.25, so your local advantage is higher than enemy). Your breakthrough army will be able to destroy one enemy before he will be able to destroy ANY of your armies, so you can wage havoc on his rears, destroying his ability to wage war. Thats more-or-less why concentration-of-force works in real life.
And in Paradox-style GSG? Well, it works exactly the same, but you have only one path.

So because the game would require more pausing to sort out a situation or two during war, the game becomes tedious? Are we not playing grand strategy games to have to figure out solutions to a number of problems including what may pop up in war?

I play GRAND strategy games to think about GREAT picture. I want to designate theaters of war and then move warships between them. Will all its suckiness, doomstack warfare is still closer to that than microing 20 fleets can be.

 

[Hawklaser]

The title is misleading because FTL is not necessarily a solution either

How is Non-FTL solutions misleading? The goal was to focus on solutions that did not involve FTL changes,(hence Non-FTL or Not FTL if non- does't make sense to you)

Seven. https://forum.paradoxplaza.com/foru...om-thread-redoox.991261/page-46#post-22815629

I read that, that is reasons doomstacks are the go to strategy, not why doomstacks are used. Those 7 also overlap a bit, and basically boil down to two of the three I mentioned, no significant other POIs, and all or nothing combat. The AI is a different issue involved in the thing due to how unvarying AI warps players gameplay.

I play GRAND strategy games to think about GREAT picture. I want to designate theaters of war and then move warships between them. Will all its suckiness, doomstack warfare is still closer to that than microing 20 fleets can be.

You want to deal with multiple war theaters, but not micro-ing a number of fleets? Isn't that a contradiction? You don't have to be needing to micro 20 fleets, I would see it more like 2-5 per front at least or so on average, depending on what the enemy is doing(this ties into why the AI is a problem). Having the AI perfectly split into 20+ fleets to destroy mining stations and such isn't tuning that correctly, for an analogy the AI will always be able to go between Lion(raiding) and Voltron(doomstack) mode more efficiently than a player will. The problem is in the balancing of that, and what % of its forces the AI throws against you at the time.Yes, the AI fleets should group up once an engagement looks to be happening, but the AI should also not always be putting the entirety of its forces in 1 fleet either. On offense, with multiple entities, rarely should it be 100% of your forces moving out but that is how the game is playing out, and no other enemy is taking advantage of the fact that all the forces are away either. The AI should also have more dynamic reasons for declaring war on you than just its force being x% bigger than yours+defensive pacts(it looks like this could be changing a bit in the patch)