Powerplay Musings on Aisling and Powerplay

For Reference:
My simple but elegant spreadsheet


Premise:
I created this document to help me, a new powerplayer to understand the situation with aisling at the moment and I have come to some very startling conclusions:

"Without fortification, Aisling Duval will have a deficit close to 900 CC"
"There are 20 control systems that provide less than 40 CC each"
"If Aisling dumps half of those systems we will have a surplus of over 400 CC"

Explanation:
These are my observations based on the data provided to us in game for cycle 45 which you can check by going to galactic powers.
Now I've had a look at the Aisling Duval trello we have here and while Conii is a relatively good system to go for, however

Point:
I feel that we would be better served if we started working towards scrapping the top 10 worst systems.
If not this week, then the start of the next week.

Now I'm well aware I'm new to powerplay and could be potentially wrong on a few things. If I am, please correct me! :) let's work together and share our ideas on this.

A little aside about overhead:
It seems that the average amount of overhead a power gets is determined by how many systems you have but it's not linear. There are different weight classes. I conclude we are too overburdened to be in the 62.1 average overhead weight class and should move down a class while removing the top 10 worst systems we hold.

A new point about Controlled Turmoil
data collected and calculated for reference

According to this, with the help or Carridin, if we were to focus our efforts on systems with upkeep more than 21 CC(or more than 32 ly from Cubeo)

"It would require 29 rank 5 fortifiers and 31 rank 5 underminers to enter into controlled turmoil."

Note: this is the best case scenario. It's a benchmark but a true one at that. You can check my data, and help me find faults in it. You can collect your data and check it yourself.
I have confidence the above statement is true.


From this, i would surmise that the effort required for Aisling Leadership to organise this would be of medium to high difficulty... why? I'm guessing. Comment below on how difficult getting 58 people to work together would be. I've not been on the slack long so I cannot give an estimate on how many active players we have whom might be interested in this.

To clarify, yes this isn't a physical plan yet, but I hope it is a step in the right direction. Creating and presenting plans and making it, bit by bit more concrete with evidence along the way.

Edit log
edit: corrected my data and modified the statements i made to be more truthful. Added conclusions from controlled turmoil logistics calculations.
edit: modified the number of rank 5 work done needed for fortifiers and underminers to take into account systems that are always fortified
 
Last edited:
Hello, I would recommend to check Aisling Duval Slack, where you can have answers to all your questions, see our plans and much more. :)

You can join by using this form: https://dinusty.typeform.com/to/ewQU8A . You should get invitation in about a day.

I'd love to! At the moment I am not allowed on the slack because I am a member of the black hand. I was to continue talking to Gswine about it but they has not yet contacted me about the potential to stay on. Here's a chat log if you're interested. don't worry it was a cordial affair. :) http://hastebin.com/ivovamihuk.vhdl

Edit: Note this isn't the topic i wish to talk about, what is it the spreadsheet i created. Is it correct? what is missing that could make it better? Where do we go from here if it is?
 
Last edited:
You are quite correct. Aisling really is in an unenviable position, in part due to simply being in a bad part of space with few profitable systems (largely being stuck 'below' the other imperial powers), in part due to having a serious grinder problem in the early stages of powerplay that made the initial situation worse than it needed to be (this was in part due to her bonuses if I understand correctly, as certain trade routes were very profitable if brought into Aisling space, so traders would work to make that happen regardless of whether the systems in question were worth it from a PP perspective), and in part due to the actions of an organized player group that apparently made it their mission to sabotage her preparations.

You are also correct in the remedy for it, and to my knowledge that's something the organized Aisling community (found Here) is already working toward - They've managed to shed some bad systems already, but due to the previously mentioned preparation sabotage, it hasn't been as effective as it could have been. The preparation sabotage is something FDev are actively looking at preventing though, so hopefully that will not be an issue for too much longer, at which point there's a lot of potential for Aisling's situation to improve :)
 
Last edited:
... (this was in part due to her bonuses if I understand correctly, as certain trade routes were very profitable if brought into Aisling space, so traders would work to make that happen regardless of whether the systems in question were worth it from a PP perspective), and in part due to the actions of an organized player group that apparently made it their mission to sabotage her preparations.

Has anyone made a list of systems traders would prefer in aisling space. that could be interesting as an agenda.

You are also correct in the remedy for it, and to my knowledge that's something the organized Aisling community (found Here) is already working toward - They've managed to shed some bad systems already, but due to the previously mentioned preparation sabotage, it hasn't been as effective as it could have been. The preparation sabotage is something FDev are actively looking at preventing though, so hopefully that will not be an issue for too much longer, at which point there's a lot of potential for Aisling's situation to improve :)

I haven't yet found any public posts about it there. Do you have any links on it?

How long would it take aisling to shed 10 bad systems, is there a way to know?
 
Last edited:
Has anyone made a list of systems traders would prefer in aisling space. that could be interesting as an agenda.

This is the route those people do:

Buy goods in nearby CS, head to closest paladium source.

Sell goods for tiny profit, pick up a free allotment of preparation/expansion goods worth 10,000 credits/ton. Buy palladium regardless of the price.

Head back to CS, drop off free allotments worth 10,000 credits/ton and sell the palladium at 10% higher base price, so typically the profit is 2,800 credits/ton instead of 1,400.

Good luck competing with that average.
 
I haven't yet found any public posts about it there. Do you have any links on it?
I'm not an Aisling pledge myself (I'm with Mahon), so I only have access to the stuff they talk about on their reddit. But they've made an effort to avoid fortifying bad systems so they have a chance to lose them if they enter turmoil, and in some cases they've sucessfully gotten rid of some real stinkers. There's apparently also a System Scrap List, but again, I can't actually see it.

In the weekly objectives posts they also make an effort to fortify only profitable systems, so that if they do get forced into turmoil, the systems that end up on the chopping block are the ones that are worth losing.

I'm afraid that's all I can find at the moment. I remember seeing more references to it some time back (as in, some months), but I can't seem to find them at the moment. And pretty much everything at Aisling is done behind closed doors these days, so there isn't really anything recent for me to find other than their weekly objectives. I don't really get all the secrecy for a power that large, but that's their decision, not mine. Myself though, I rather prefer having things in the open (that's one of the reasons I'm pledged as I am).

How long would it take aisling to shed 10 bad systems, is there a way to know?
In theory, not long at all.

1) Fortify all profitable systems, and only those
2) Get a friendly enemy to undermine the 10 systems you most want to lose, plus any extras needed to throw you into heavy turmoil despite fortifying the profitable systems.
3) Next cycle, you once again fortify only profitable systems, and once again you get someone to undermine the systems you want to lose, making sure they have the highest upkeep of your turmoiled systems.
4) Repeat as needed, as you might not lose all of them in one go.

That's the theory. But it assumes everyone cooperates. In reality you'll have random pledges who don't know any better fortifying systems you don't want fortified, enemies undermining systems you don't want undermined, and more. So controlling it is hard. ALD has managed it though, and Aisling to a lesser extent as well. So it IS doable.
 
Last edited:
This is the route those people do:

Buy goods in nearby CS, head to closest paladium source.

Sell goods for tiny profit, pick up a free allotment of preparation/expansion goods worth 10,000 credits/ton. Buy palladium regardless of the price.

Head back to CS, drop off free allotments worth 10,000 credits/ton and sell the palladium at 10% higher base price, so typically the profit is 2,800 credits/ton instead of 1,400.

Good luck competing with that average.
hmm, that's still not a list... however i do believe it's not impossible to get 2.8k or 3k per ton from trading. what with etn.io and eddb.io at our disposal.
Scott Dock, Bes to Shen Terminal, Guuguyni gets us 3341 CR per ton per loop or 2782 CR per ton. albeit a medium pad.. maybe not so great.
 
I'm not an Aisling pledge myself (I'm with Mahon), so I only have access to the stuff they talk about on their reddit. But they've made an effort to avoid fortifying bad systems so they have a chance to lose them if they enter turmoil, and in some cases they've sucessfully gotten rid of some real stinkers. There's apparently also a System Scrap List, but again, I can't actually see it.

In the weekly objectives posts they also make an effort to fortify only profitable systems, so that if they do get forced into turmoil, the systems that end up on the chopping block are the ones that are worth losing.

I'm afraid that's all I can find at the moment. I remember seeing more references to it some time back (as in, some months), but I can't seem to find them at the moment. And pretty much everything at Aisling is done behind closed doors these days, so there isn't really anything recent for me to find other than their weekly objectives. I don't really get all the secrecy for a power that large, but that's their decision, not mine. Myself though, I rather prefer having things in the open (that's one of the reasons I'm pledged as I am).

Well i feel the same way. I think a lot more discussion could happen in the open. It's partially why i created this. I don't think I've seen much discussion in the open. i don't think we have a system scrap list, at least according to our trello.


In theory, not long at all.

1) Fortify all profitable systems, and only those
2) Get a friendly enemy to undermine the 10 systems you most want to lose, plus any extras needed to throw you into heavy turmoil despite fortifying the profitable systems.
3) Next cycle, you once again fortify only profitable systems, and once again you get someone to undermine the systems you want to lose, making sure they have the highest upkeep of your turmoiled systems.
4) Repeat as needed, as you might not lose all of them in one go.

That's the theory. But it assumes everyone cooperates. In reality you'll have random pledges who don't know any better fortifying systems you don't want fortified, enemies undermining systems you don't want undermined, and more. So controlling it is hard. ALD has managed it though, and Aisling to a lesser extent as well. So it IS doable.

Can we spreadsheet that? calculate as best we can in hard numbers how many pilots we need? I like the theory but i'd love it if strategies like this can be described in real world figures... hmm i'll see what i can do myself with the spreadsheet and the data i have.
 
Can we spreadsheet that? calculate as best we can in hard numbers how many pilots we need? I like the theory but i'd love it if strategies like this can be described in real world figures... hmm i'll see what i can do myself with the spreadsheet and the data i have.
Absolutely.

For a control system to be break-even or profitable in the long term, which are the ones worth keeping, it needs a profit which is greater than 62CC (the per-system overhead for powers with 55+ control systems). So start by tallying up the fortification triggers for those systems. That's the total fortification work you need to do.

Next, divide that total number it by 5000 to find out how many Rating 5 commanders are needed to complete all the fortification, assuming they each only do enough to maintain Rating 5. That's a fair estimate of the number of pledged players you need - some will do more, some will do less, but it'll get you into the ballpark.

Then pick ten systems you want to lose, and which are actually possible to lose. Some systems are so close to HQ and have such a low income that they're hard to lose even if undermined, so ignore those. You want systems which are lossmakers but which are also not right next door to your HQ, as upkeep increases with distance, thus making them easier to get into turmoil. Also, systems close to HQ are more likely to be fortified by randoms, which you don't want to happen to your picks.

Tally up the undermining triggers for those 10 systems, and see if that's enough to throw you into turmoil by at least 1,000CC. If not, add more systems to the undermining list, but try to make those systems that are also lossmakers. Once you have enough systems on your undermining list to make sure you hit -1,000CC even with all profitable systems fortified, that's the undermining number you're looking at. Then you can divide that number by 30 to get the total number of kills your enemies need to score to reach that total, given perfect efficiency. And then finally you can divide the number of kills by 166 (5000/30) to find the number of enemy Rating 5 pilots you'd need to enlist, again given perfect efficiency and assuming they only kill enough to maintain their rating. Or you could use collusion undermining, but FDev seem none too pleased with that method so it is likely to get nerfed at some point.

[Edit] I would assume Aisling has a spreadsheet something like This somewhere (warning, might take a while to load). If so, that's an ideal place to grab all the numbers instead of having to type them in by hand. Which is painful. I know from experience...
 
Last edited:
Isn't one ton of prep materials worth 100 CR? And also, for preparation, we have to take them FROM CS to a system, and for expansion from the expansion system to a CS.
 
I thought it was 1000 credits per hauled merit personally, unless Martin was referring to the opportunity cost vs buying them.
 
Isn't one ton of prep materials worth 100 CR? And also, for preparation, we have to take them FROM CS to a system, and for expansion from the expansion system to a CS.

100 if what you get for delivering the merit.

10,000 is the fast track cost per merit.

In essence, fast tracking costs 9,900 credits per ton.
 
Last edited:
Absolutely.

For a control system to be break-even or profitable in the long term, which are the ones worth keeping, it needs a profit which is greater than 62CC (the per-system overhead for powers with 55+ control systems). So start by tallying up the fortification triggers for those systems. That's the total fortification work you need to do.

Next, divide that total number it by 5000 to find out how many Rating 5 commanders are needed to complete all the fortification, assuming they each only do enough to maintain Rating 5. That's a fair estimate of the number of pledged players you need - some will do more, some will do less, but it'll get you into the ballpark.

Then pick ten systems you want to lose, and which are actually possible to lose. Some systems are so close to HQ and have such a low income that they're hard to lose even if undermined, so ignore those. You want systems which are lossmakers but which are also not right next door to your HQ, as upkeep increases with distance, thus making them easier to get into turmoil. Also, systems close to HQ are more likely to be fortified by randoms, which you don't want to happen to your picks.

Tally up the undermining triggers for those 10 systems, and see if that's enough to throw you into turmoil by at least 1,000CC. If not, add more systems to the undermining list, but try to make those systems that are also lossmakers. Once you have enough systems on your undermining list to make sure you hit -1,000CC even with all profitable systems fortified, that's the undermining number you're looking at. Then you can divide that number by 30 to get the total number of kills your enemies need to score to reach that total, given perfect efficiency. And then finally you can divide the number of kills by 166 (5000/30) to find the number of enemy Rating 5 pilots you'd need to enlist, again given perfect efficiency and assuming they only kill enough to maintain their rating. Or you could use collusion undermining, but FDev seem none too pleased with that method so it is likely to get nerfed at some point.

[Edit] I would assume Aisling has a spreadsheet something like This somewhere (warning, might take a while to load). If so, that's an ideal place to grab all the numbers instead of having to type them in by hand. Which is painful. I know from experience...

I've done it. it's in the spreadsheet now. 23 fortifiers, 46 underminders.

What is collusion undermining?

We do have a spreadsheet here: https://onedrive.live.com/redir?resid=AC5B7079459436FC!118&authkey=!AORYK1TSXMscs4E&ithint=file,xlsx
However the information how much fortification and undermining points you need is absent so i had to do it by hand.

I've chosen the top 10 systems but a fair amount of them are 21 upkeep systems which as you say are hard to get rid of.

Is there no way to get rid of any of them?
 
I've since modified what to undermine and have forgone systems that have 21 upkeep and now it's at 23 fortifiers and 46 underminders. unfortunately it bleeds out of the top 10 worst systems.
 
What is collusion undermining?
There's a thread about it Here. Summarizing, it involves having pledges dump fortification cargo which is then picked up by a collaborator in a different power from the same major faction (Empire, in your case, so either ALD, Patreus or Torval), who in turn hands it in at one of their control system. It is essentially exploiting the piracy undermining mechanic by deliberately dumping cargo, which allows for very fast and very precisely targeted undermining of specific systems.

I've chosen the top 10 systems but a fair amount of them are 21 upkeep systems which as you say are hard to get rid of.

Is there no way to get rid of any of them?
In a perfect world, it could perhaps be done. But I doubt any of the rival powers (or random grinders, for that matter) would let you get away with setting up the conditions to make it happen. Then again, we're kinda busy with the Federation at the moment and they're VERY busy with us, so who knows, and you might get lucky on the grinders. But I wouldn't count on it. Better to go for more realistic targets further from HQ, in my opinion.
 
There's a thread about it Here. Summarizing, it involves having pledges dump fortification cargo which is then picked up by a collaborator in a different power from the same major faction (Empire, in your case, so either ALD, Patreus or Torval), who in turn hands it in at one of their control system. It is essentially exploiting the piracy undermining mechanic by deliberately dumping cargo, which allows for very fast and very precisely targeted undermining of specific systems.


In a perfect world, it could perhaps be done. But I doubt any of the rival powers (or random grinders, for that matter) would let you get away with setting up the conditions to make it happen. Then again, we're kinda busy with the Federation at the moment and they're VERY busy with us, so who knows, and you might get lucky on the grinders. But I wouldn't count on it. Better to go for more realistic targets further from HQ, in my opinion.

This might be something to talk about with members of the other imperials. thanks!
 
Well i feel the same way. I think a lot more discussion could happen in the open. It's partially why i created this. I don't think I've seen much discussion in the open. i don't think we have a system scrap list, at least according to our trello.
This might be something to talk about with members of the other imperials. thanks!
There are very good reasons for such a scrap list not being publicly displayed, here is some of them:
1. Our best weapon against grinders and 5C's preps is our ability to tailor our end cc level. It's not much, but that's all FDev has given us to work with.
So let's say we plan on preparing a 100 prep cost system (just for the sake of using a simple number): then we need to end the cycle as close to that 100 cc as possible, to ensure no other system sneaks into our prep list (so we have a less than 50 cc window to work with, based on the cost of some of those bad systems). The issue with a public scrap list is that you're giving the go-ahead to lower our cc whenever anyone feels like it, which more often than not will make us drop below and miss that 100 cc target we have in the worst case, or forces our commanders to rush forts in the last few hours to salvage it in the best case.
A public scrap list might be good for small powers with a lot of cc to spare and no big prep wars to fight or powers with a lot of extra cc they need to burn each cycle anyway. We're not in either of those cases.

2. Any AD system undermined early will receive extra attention from the grinders. By posting a scrap list of systems we want to lose we will effectively end up with our uninformed player base fortifying even more of them, if not all of them. As a result the number of bad systems we can lose would actually be lowered and not increased by having such a public scrap list for AD.
I'm not arguing that a public scrap list is bad for every power; that would be a stupid thing to say as it was proven effective on many occasions. Simply that in the current AD's situation it wouldn't help us.


Also while I agree with you that a lot more things could happen in open, discussion of turmoil plans is not one of those. Having other powers (or simply rogue groups, such as but not limited to, our 5C) know what our intentions are (for example reaching 300cc / 200cc / 100cc at the end of the cycle or even starting a turmoil) will make it incredibly easy for them to mess up with our plans.

PS: I am well aware that 100cc prep cost systems are bad systems, I'm not saying we're actually trying to reach that exact number :p
 
Last edited:
There are very good reasons for such a scrap list not being publicly displayed, here is some of them:
1. Our best weapon against grinders and 5C's preps is our ability to tailor our end cc level.

It's not much, but that's all FDev has given us to work with.

example:
So let's say we plan on preparing a 100 prep cost system (just for the sake of using a simple number): then we need to end the cycle as close to that 100 cc as possible, to ensure no other system sneaks into our prep list (so we have a less than 50 cc window to work with, based on the cost of some of those bad systems). The issue with a public scrap list is that you're giving the go-ahead to lower our cc whenever anyone feels like it, which more often than not will make us drop below and miss that 100 cc target we have in the worst case, or forces our commanders to rush forts in the last few hours to salvage it in the best case.

conclusion:
A public scrap list might be good for small powers with a lot of cc to spare and no big prep wars to fight or powers with a lot of extra cc they need to burn each cycle anyway. We're not in either of those cases.

I just want to be clear, this spreadsheet i made isn't a public scrap list. it's just a calculator to show proof of concept and calculate expected cost.

Now that that's said, given that all this data is publically accessible to every power player in the galaxy map. There is no point keeping discussions about scrapping hidden.
When you have the same data, you'll come to the same conclusions. Because of that, hiding spreadsheets like this calculator is a mute point because if other powers were really interested, they'd have their own spreadsheets.

2. Any AD system undermined early will receive extra attention from the grinders.

Explanation:
By posting a scrap list of systems we want to lose we will effectively end up with our uninformed player base fortifying even more of them, if not all of them.
As a result the number of bad systems we can lose would actually be lowered and not increased by having such a public scrap list for AD.

Conclusion:
I'm not arguing that a public scrap list is bad for every power; that would be a stupid thing to say as it was proven effective on many occasions. Simply that in the current AD's situation it wouldn't help us.

I think I agree with part of this. while we have an optimal number of rfmd(rank five, work done), we will definitely need more than that number to deal with those who don't follow the plan. The more information about how best to calculate this the better i'll understand what you mean.

I don't think it's untenable to wrestle with the uninformed player based. as you've said, scraps have occurred with AD. In fact AA has done scraps before. Perhaps I don't understand why it wouldn't help us.

More than 70% of our systems produce less CC than their upkeep + average overhead costs.


We're spending a whole lot of effort propping up our systems with fortification efforts just to stay afloat. You're saying this is easier?

Addendum:
Also while I agree with you that a lot more things could happen in open, discussion of turmoil plans is not one of those. Having other powers (or simply rogue groups, such as but not limited to, our 5C) know what our intentions are (for example reaching 300cc / 200cc / 100cc at the end of the cycle or even starting a turmoil) will make it incredibly easy for them to mess up with our plans.

PS: I am well aware that 100cc prep cost systems are bad systems, I'm not saying we're actually trying to reach that exact number :p

oh i agree we should have secure comms as well, i'm not disputing that. I'd love to be a part of that process, but I shan't say more until I have more information on whether i can be a part of that process. For now, the community and the various community platforms are the best place i can share my concerns and ask questions.
 
Last edited:
Thanks for your thoughtful reply :)
I am not trying to say that we should not organize a scrap effort at some point, I - and anyone that has a good understanding of our situation - agree that it needs to be done.
Still, it has to be done right. For AD, that means taking the grinders into account.

Here is how the grinder fortifications happen, and this is true for every cycle:
- They will fortify the bad systems close to Cubeo each cycle, no matter what we do: this is at least Lambda-1 Tucanae, Karsuki Ti and Kuki An that will end up over-fortified no matter what. Chnumar could also be added to that list although it could just be some 5C elements that are behind it.
- As soon as any amount of undermining happens in any of our bad systems they will fortify said system. It does not matter how vocal anyone is about the fact those should not be fortified, using trello/reddit/forum or even in-game chat, they will most likely be fortified. That's just the way it is and we have to take that into consideration.

That second part is why we need to not have any of the systems we are trying lose undermined. A public scrap list would actually be asking friendly commanders and powers to undermine us, hence why we don't have one at the moment.

For any scrap effort to be successful, it will probably need to be done less than a day before the end of the cycle and preferably only hours before if we want it to be effective. Any undermining done to our bad systems before that point will merely result in preventing them from being turmoil candidates.
Which is why I - and many others - believe the undermining of our bad systems to be something that must be done mostly (if not only) by AD commanders that can work closely together and communicate with one another, so we can have a complete control over which system end up undermined and most importantly when it happens. That is not to say we do not want the help of our allies; we will need to have as many receivers as possible when we do decide to go ahead with such a plan.
 
Last edited:
Back
Top Bottom