Analysing the Thargoid Simulation

[Aleks Zuno]

Reading the earlier statements once again and checking the system positions, I think our plan to protect Ramandji by evicting Col 285 Sectors RX-Z b14-0 and RX-Z b14-8 would then disrupt the test you wanted!

If I understood, you need an evicted inhabited Control system to go through its Recovery process, but then also to remain a target desirable enough that it would become a new Alert at the earliest opportunity. That means its attackers need to remain present, and it seems to me that the two Ramandji attackers are also the only Yukait attackers. Those two systems are so close that Yukait itself could attack Ramandji had it been unable to target Tau Puppis.

Assuming no Recovery drive there, the time involved surely will be a full month due to Yukait having four ports. With all of that in mind, I could use very much some help deciding what to do with Col 285 Sectors RX-Z b14-0 and RX-Z b14-8:

• Keep them around as Control systems, assuming nobody else evicts them. This may yield a result after the Recovery month, but I presume this is not assured because much can change. Those attackers could end up expending their capacity at Ramandji, Prahlang, Aitvani or Bileku on the week that Yukait becomes available, disrupting the test anyway.
• Abort the test "opportunity", drop the Research team in and just effect their riddance. Their upkeep is high, any result will have an element of uncertainty, and they may not yield a result depending on actions of other Commanders.
• Something else which I may not have considered or not understood.

Thinking about the test requirements a bit, it seems to me that you would need a Control eviction to occur at a system with attackers which can not target anything else; this could be satisfied by sparse positioning, or by all systems within twenty light-years being Control, so that those within ten light-years have no distractions. Admittedly I have not checked how Recovery works now, though I imagine that you would want also a Recovery team to push it to completion in one cycle, making it eligible for another Alert as soon as possible.

I have a general preference for making M. Hadad smaller if we can; this does not preclude a test necessarily, given that said testing is an investment in your Maelstrom-reducing reports, so I thought to present the above for comment before doing anything untoward.

 

[Ian Doncaster]

Thanks for checking in, but please go ahead with your local plans - there is one inhabited control eviction that has already been re-attacked, so the data isn't worth disrupting existing defensive strategy over.

Thinking about the test requirements a bit, it seems to me that you would need a Control eviction to occur at a system with attackers which can not target anything else; this could be satisfied by sparse positioning, or by all systems within twenty light-years being Control, so that those within ten light-years have no distractions.

While that would be ideal, it'd also be sufficient for the eviction to be close enough to the Maelstrom and have a decent number of potential controls to target it that it was guaranteed a top-5 slot in an active week.

As far as Hadad goes, recapturing HIP 30158 or Fotlandjera would probably do that - but they're also close enough to make the actual recapture potentially tricky. There are likely better opportunities going to come up elsewhere.

 

[]M[]

After some time away from the war & all out exploring the past while, just wondering about the formation of the maelstroms and their expansion into nearby systems - do there seem to be any spatial constraints (an observable pattern even?), or have they simply spread out according to the kind of weekly rolled turn-by-turn rules that you have been observing / infering, @Ian?

 

[Ian Doncaster]

After some time away from the war & all out exploring the past while, just wondering about the formation of the maelstroms and their expansion into nearby systems - do there seem to be any spatial constraints (an observable pattern even?), or have they simply spread out according to the kind of weekly rolled turn-by-turn rules that you have been observing / infering, @Ian?

That's a difficult question to answer.

It appears to be the case that an iterative turn-by-turn model is sufficient to explain Thargoid behaviour. And that model does not contain any pattern or spatial constraints beyond those imposed by the 10 LY range limit, the local shape of the map, and the tendency of human forces to fight back. Similarly, statistical tests on the positions of Thargoid-held/contested systems don't show any particular directionality or pattern - at least, not the ones we've tried.

But it's always going to be difficult to tell if that's because the model is a successfully reverse-engineered description of the Thargoid strategy, or if it just meets it by coincidence in limited situations. If we can stick with the same model for multiple weeks and get a consistent 95%+ hit rate every time, with all misses explainable by incomplete data ... that increases the confidence that we're on the right lines, and also means that misses we can't explain, or lots of them at once, are then clearer indications of either new constraints or changes in Thargoid strategy.

 

[]M[]

That's a difficult question to answer.

It appears to be the case that an iterative turn-by-turn model is sufficient to explain Thargoid behaviour. And that model does not contain any pattern or spatial constraints beyond those imposed by the 10 LY range limit, the local shape of the map, and the tendency of human forces to fight back. Similarly, statistical tests on the positions of Thargoid-held/contested systems don't show any particular directionality or pattern - at least, not the ones we've tried.

But it's always going to be difficult to tell if that's because the model is a successfully reverse-engineered description of the Thargoid strategy, or if it just meets it by coincidence in limited situations. If we can stick with the same model for multiple weeks and get a consistent 95%+ hit rate every time, with all misses explainable by incomplete data ... that increases the confidence that we're on the right lines, and also means that misses we can't explain, or lots of them at once, are then clearer indications of either new constraints or changes in Thargoid strategy.

Overall it still looks more like a feathered [fuzzy logic driven? :] slowly expanding wall …rather than a cage, or battering ram even?

 

[Tassier]

Other than a mystery 2% at Trianguli Sector GG-Y c18, nothing else of that sort seems to be happening.

That 2% was me last Friday, trying to get an idea of what progress a solo pilot can achieve against an inhabited control by delivering 51 tissue samples. This is about the max that I tend to get in a session with my sub-optimal setup (no carrier, no cutter).

There's no way I can scale up my solo harvesting to a high enough count to make a dent there so I shifted over to a neighbouring uninhabited control instead, Trianguli Sector BA-A d84, much easier. Even if I don't finish it this week I can definitely finish it next week.

Having seen the great progress made in this thread I'll move to more targeted systems in future

 

[Ian Doncaster]

Updated predictions due to Control recaptures.

In Taranis, the presumed recapture of Trianguli Sector BA-A d84 doesn't change the top 5 as JR-W b1-0 is still attackable from GG-Y c18, but it does adjust the reserve list a bit

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
Cao Tzu	21.52	4	4	Hyades Sector GW-W d1-87; Hyades Sector GW-W d1-86; Hyades Sector AV-O b6-4; etc	Yes	-0
Hyades Sector IM-L b8-5	25.77	2	2	HIP 25679; Nihal;	No	0.3
HIP 21991	25.79	2	2	Trianguli Sector JR-W b1-5; Hyades Sector GW-M b7-5;	Yes	0.4
HR 1812	26.62	1	1	Trianguli Sector KR-W b1-1;	Yes	0.8
Trianguli Sector JR-W b1-0	28.74	2	1	Trianguli Sector GG-Y c18;	No	0.9
Pentam	28.87	6	6	Hyades Sector NS-T c3-14; Haushu; Hyades Sector NS-T c3-12; etc	Yes	0.3
Hyades Sector EQ-O b6-4	29.07	3	3	Hyades Sector EQ-O b6-2; Hyades Sector DQ-O b6-2; Hyades Sector DQ-O b6-4;	No	-0.9
Hyades Sector XO-Q b5-4	29.26	3	3	Hyades Sector GW-W d1-82; Hyades Sector XO-Q b5-3; HIP 25388;	No	-0.5
Hyades Sector VT-Q b5-2	29.27	2	1.3	Hyades Sector XO-Q b5-3; HIP 25388;	No	-0.4
Nareg	29.54	1	1	HIP 25840;	Yes	0.8

The recapture in Oya is too far out to affect the predicted Alerts at all.

Hadad, of course, has seen more substantial changes:

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
Ahol	20.73	2	2	Col 285 Sector TS-Z b14-2; Col 285 Sector TS-Z b14-3;	Yes	0.9
Col 285 Sector WN-Z b14-6	28.93	2	2	HR 2138; Col 285 Sector MR-M c7-23;	No	-0.3
Col 285 Sector FQ-O c6-3	30.68	2	2	Col 285 Sector RH-B b14-1; Bagalya;	No	-0.1
Col 285 Sector FQ-O c6-4	31.51	1	0.5	Col 285 Sector RH-B b14-1;	No	0.2
Col 285 Sector KW-M c7-25	31.96	8	8	Col 285 Sector KW-M c7-24; Col 285 Sector XN-Z b14-1; Col 285 Sector KW-M c7-10; etc	No	-0.7
Col 285 Sector GQ-O c6-21	31.98	8	8	Col 285 Sector OB-D b13-0; Col 285 Sector OB-D b13-1; HIP 29217; etc	No	-0.7
Col 285 Sector RN-T d3-61	32.08	2	2	Col 285 Sector OR-B b14-0; Col 285 Sector OR-B b14-1;	No	0.3
Col 285 Sector XN-Z b14-5	32.13	8	7.3	Col 285 Sector VS-Z b14-1; Col 285 Sector KW-M c7-24; Col 285 Sector XN-Z b14-1; etc	No	-0.8
Col 285 Sector KW-M c7-20	32.23	8	6.4	Col 285 Sector VS-Z b14-1; Col 285 Sector KW-M c7-17; Col 285 Sector KW-M c7-24; etc	No	-0.7

FQ-O c6-4 will be interesting, as if it is attacked it requires Bagalya was used for c6-3 ... whereas a miss might indicate that control systems are also used from the inside outwards (though that would need a lot more confirmation from other experiments)

 

[Aleks Zuno]

I fear I have meddled again—Col 285 Sector TS-Z b14-2 is already cleared, and we will be making an attempt at TS-Z b14-3 as well (they do not move quite as easily!). I had noticed and considered RH-B b14-1 earlier, though I imagine we will need all reasonable time for TS-Z b14-3.

There's no way I can scale up my solo harvesting to a high enough count to make a dent there so I shifted over to a neighbouring uninhabited control instead, Trianguli Sector BA-A d84

I was going to say that I would love to help there were it not for the our Col 285 Sector TS-Z b14-3 goal, but at 90% I think you will have it cleared regardless!

 

[Ian Doncaster]

I fear I have meddled again—Col 285 Sector TS-Z b14-2 is already cleared, and we will be making an attempt at TS-Z b14-3 as well (they do not move quite as easily!). I had noticed and considered RH-B b14-1 earlier, though I imagine we will need all reasonable time for TS-Z b14-3.

If you don't have any particular strategic need to keep them out of the FQ-O subsector, it'd be good to leave RH-B alone for this test, please.

Almar and Col 285 Sector HL-O c6-14 at Thor are another such testable pair that it would be helpful to have left alone this week if there's no local need to prioritise them.

I'll re-run the model tomorrow morning anyway for the final assessment of accuracy, so any further captures between now and then will be included.

 

[Ralph Vargr]

If you don't have any particular strategic need to keep them out of the FQ-O subsector, it'd be good to leave RH-B alone for this test, please.

Almar and Col 285 Sector HL-O c6-14 at Thor are another such testable pair that it would be helpful to have left alone this week if there's no local need to prioritise them.

I'll re-run the model tomorrow morning anyway for the final assessment of accuracy, so any further captures between now and then will be included.

All too soon, I will be watching the NHC hurricane model runs....which this reminds me of.

 

[Aleks Zuno]

FQ-O c6-4 will be interesting, as if it is attacked it requires Bagalya was used for c6-3 ... whereas a miss might indicate that control systems are also used from the inside outwards (though that would need a lot more confirmation from other experiments)

Exactly this appears to have occurred! Having cleared Col 285 Sectors TS-Z b14-2 and TS-Z b14-3 to strike off Ahol, the only miss in the top five was at FQ-O c6-4 exactly, and indeed the reserve RN-T d3-61 in sixth has taken its place.

Thank you ever so much for the Alert analysis; I find it most pleasing that a Maelstrom can be forced that way to spend its expansion capacity on easily-cleared empty systems!

 

[Ian Doncaster]

So, how did it go this week?

• Taranis: 100%
• Leigong: 80% - Pathamon has entered a fourth week of Recovery, so the first reserve was used
• Indra: 80% - HR 1403 has entered a fourth week of Recovery, so the first reserve was used
• Oya: 100%
• Cocijo: 80% - Hez Ur has entered a fourth week of Recovery, so the first reserve was used
• Thor: 100% (in potentially difficult circumstances, too)
• Raijin: 100%
• Hadad: 80% (after accounting for last minute pre-emptive defence of Ahol and re-running the model) - as considered likely, FQ-O c6-4 was omitted

Total accuracy 36/40 = 90% - a significant improvement on previous weeks, and in no case did the model need to go beyond the first reserve.

Error sources:

• Three errors were caused by defended inhabited systems entering an "unexpected" fourth week of Recovery and so being unavailable as targets. I will try to highlight these specifically on future predictions so that I can track where this error is getting in to the source data for the model a bit better.
• One error in Hadad where a system at 0.5 confidence was skipped over.

The behaviour in Hadad's FQ-O subsector, and Thor's Almar region, would be consistent with a rule where the Alert targets are considered from the inside outwards, and where multiple controls could be used, the Control nearest the Maelstrom carries out the attack. However, there are other potential explanations so I won't be incorporating this theory into the model just yet: at this stage it's better to list the possibilities with the low confidence levels, rather than potentially omit an actual target.

 

[Ian Doncaster]

Narwhal Nose report generated 20 April

If a system is currently in its third week of recovery, I've marked its confidence level with a '*' - EDIT: all such systems re-checked in-game and they all say "recovery complete, all ports active in 6D". So let's see how many of them actually do that...

-------------------------------------------

Calculating Taranis
Read 123 control systems and 870 targets.

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
HIP 23716	18.43	7	6.5	Trianguli Sector BA-A d85; Trianguli Sector CA-A c22; Trianguli Sector FL-Y b4; etc	Yes	0.9
63 Eridani	19.56	1	1*	Hyades Sector PN-T c3-9;	Yes	-0.3
Senocidi	20.26	4	4*	Hyades Sector GW-W d1-87; Hyades Sector PN-T c3-8; HIP 24076; etc	Yes	-0.4
Awara	22.00	4	4*	Hyades Sector EB-N b7-0; HIP 23593; Hyades Sector GW-W d1-86; etc	Yes	0.4
HIP 26274	23.64	3	3	Hyades Sector GW-W d1-81; Hyades Sector NS-T c3-14; Hyades Sector ZZ-O b6-1;	Yes	0.1
Vukurbeh	24.25	1	1	Matshiru;	Yes	0.8
Paeni	26.73	4	3.7*	29 e Orionis; Hyades Sector GW-W d1-81; Trianguli Sector ZJ-A b1; etc	Yes	0.3
Lovaroju	26.93	3	3*	Hyades Sector PN-T c3-7; Hyades Sector QN-T c3-11; Hyades Sector PN-T c3-5;	Yes	-1

A lot of these are *d so this may need reissuing once I've had chance to check. Notice how the defense strategy in use at Taranis is being responded to by the Thargoids picking a full set of low-range (i.e. high-difficulty) inhabited systems, including as reserves.

-------------------------------------------

Calculating Leigong
Read 160 control systems and 835 targets.

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
Pathamon	22.06	6	6	HIP 7277; Arietis Sector JM-W d1-56; Arietis Sector MX-U c2-17; etc	Yes	0.5
Cabarci	24.87	1	1	Arietis Sector AQ-P b5-3;	Yes	0.6
Hyades Sector LN-K b8-4	25.71	1	1	Hyades Sector LN-K b8-3;	No	0.2
Arietis Sector YU-P b5-1	25.81	1	1	Arietis Sector AQ-P b5-2;	No	0.4
Hyades Sector LN-K b8-2	26.38	1	1	HIP 10056;	No	0.4
Wolf 121	28.92	1	1	Arietis Sector CQ-P b5-2;	Yes	0.5
Arietis Sector YJ-R b4-3	35.37	2	2	HIP 10616; Arietis Sector YJ-R b4-4;	No	0.6
Medzist	35.48	1	0.5	Arietis Sector YJ-R b4-4;	Yes	0.8
Arietis Sector VO-R b4-4	36.03	3	3	Arietis Sector VO-R b4-2; Arietis Sector VO-R b4-1; Arietis Sector VO-R b4-0;	No	0.7
Ceti Sector ZU-Y b3	36.45	3	3	Ceti Sector ZU-Y b1; Ceti Sector ZU-Y b2; HIP 6679;	No	0.8
Dao Tzu	36.64	1	0.5	HIP 10616;	Yes	0.8
Arietis Sector DW-N b6-4	37.59	2	1	Arietis Sector JM-W d1-83; HIP 6380;	No	0.5
Arietis Sector ZU-P b5-0	38.33	3	2.7	Arietis Sector ZU-P b5-4; Arietis Sector VO-R b4-0; Arietis Sector ZU-P b5-2;	No	0.8

There are a lot of low-confidence targets here picked only by a single control, but equally I'd expect most of these to be fine as those controls almost certainly weren't used last week. If there was an attempt to cut off controls and move down the list, note that Medzist and Dao Tzu can't both be targeted on the reserves (the draft theory would say they hit Medzist).

-------------------------------------------

Calculating Indra
Read 147 control systems and 844 targets.

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
HIP 20890	21.89	16	16*	HIP 20577; HR 1354; HIP 20679; etc	Yes	0.9
HIP 20527	22.54	21	20.8	Hyadum I; 71 Tauri; HIP 20916; etc	Yes	0.9
HR 1403	24.66	10	9.6	HR 1354; HIP 20712; Arietis Sector JR-V b2-4; etc	Yes	0.9
Hyades Sector PI-S b4-0	26.75	2	2	Hyades Sector PI-S b4-3; Hyades Sector PI-S b4-1;	No	-0.9
Hyades Sector WK-L a9-2	26.79	2	2	Hyades Sector AR-J a10-0; Hyades Sector BR-J a10-0;	No	-0.3
HIP 20850	26.99	13	12.4	77 Theta-1 Tauri; 70 Tauri; HIP 20815; etc	Yes	0.9
57 h Tauri	27.03	12	10.5	77 Theta-1 Tauri; 70 Tauri; HIP 20815; etc	No	1
Hyades Sector EX-H a11-1	27.04	6	5.8	Lei Hsini; HIP 22350; Arietis Sector TY-P a6-0; etc	No	0.6

All pretty high confidence, though HIP 20890 will need its recovery progress checking.

-------------------------------------------

Calculating Oya
Read 83 control systems and 911 targets.

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
Muncheim	23.56	5	4.5	HIP 7338; Cephei Sector FB-X b1-1; Bumbo; etc	Yes	0.7
HIP 2422	24.01	8	7.2	HIP 8525; HIP 4041; HIP 7338; etc	Yes	0.5
Jaoi	25.46	3	3	Cephei Sector ZE-A c8; Cephei Sector ZE-A c10; Cephei Sector ZE-A d69;	Yes	-0.4
Gliese 3050	25.58	5	3.4	HIP 8525; HIP 7338; Cephei Sector DQ-Y b2; etc	No	0.6
Cephei Sector EL-Y c23	26.55	1	1	MCC 105;	No	1
HIP 38225	26.96	2	2	Lyncis Sector YF-O b6-1; Ikpen;	Yes	-0.3
Orong	27.13	3	2.5	HR 1107; Ikpen; Cephei Sector YJ-A c8;	Yes	-0.2
Benanekpeno	27.19	3	1.5	HR 1107; Ikpen; Cephei Sector YJ-A c8;	Yes	0
Elboongzi	27.39	1	1	Lyncis Sector VU-P b5-5;	Yes	-0.9

Also good confidence in the top-5 here. Cephei is the only dubious one and it seems very unlikely that its attacker was used last week. Oya's target list has the potential to settle into a fixed 3-week cycle unless they also start retaking systems within the radius.

-------------------------------------------

Calculating Cocijo
Read 194 control systems and 800 targets.

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
Khwal	21.53	9	9	Mapon; Col 285 Sector ZE-P c6-15; Col 285 Sector ZE-P c6-16; etc	Yes	1
Hez Ur	22.00	4	3.9	Col 285 Sector WY-F b12-3; Mapon; Col 285 Sector WY-F b12-5; etc	Yes	0.7
Sukurbago	24.15	1	1	Col 285 Sector YT-F b12-7;	Yes	0.4
Col 285 Sector VN-H b11-2	25.80	1	1	Col 285 Sector WN-H b11-5;	No	-0.8
Col 285 Sector UD-G b12-2	25.95	3	3	Kurumanit; Col 285 Sector WY-F b12-1; 19 Puppis;	Yes	0.4
Col 285 Sector BA-P c6-10	27.85	2	2	Col 285 Sector OS-T d3-106; Col 285 Sector BA-P c6-4;	No	0.4
Asletae	28.54	3	2.8	Col 285 Sector WY-F b12-0; Picenile; Col 285 Sector XT-F b12-7;	Yes	0.9
Col 285 Sector AP-F b12-2	30.07	3	3	Col 285 Sector YT-F b12-2; Col 285 Sector YT-F b12-0; Col 285 Sector YT-F b12-5;	No	0.7
Col 285 Sector US-H b11-1	30.12	3	3	Col 285 Sector AF-P c6-1; Col 285 Sector YT-Q c5-12; Col 285 Sector YT-Q c5-10;	No	-1

A couple of chances for unavailable control systems to push into the reserves list here

-------------------------------------------

Calculating Thor
Read 156 control systems and 839 targets.

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
Col 285 Sector TH-C b13-3	22.35	2	2	Vaipacnali; HIP 17897;	No	0.5
Unktety	25.36	1	1*	HIP 19198;	Yes	1
Col 285 Sector HL-O c6-15	38.45	1	1	Col 285 Sector SI-T d3-89;	No	0.3
Kalkaduna	38.66	2	2	HIP 21548; Col 285 Sector LL-E b12-5;	Yes	0.8
HIP 16798	39.53	1	1	Col 285 Sector MQ-F b11-3;	No	-0.5
Col 285 Sector NQ-F b11-0	40.04	2	2	Col 285 Sector NQ-F b11-1; Col 285 Sector NQ-F b11-2;	No	-0.7
Col 285 Sector QW-D b12-1	40.27	2	2	Col 285 Sector QW-D b12-2; Col 285 Sector QW-D b12-3;	No	-0.1
HIP 18513	40.47	1	0.5	Col 285 Sector LB-O c6-3;	No	-0.6
Col 285 Sector HF-G b11-5	41.27	4	3.5	HIP 21548; Col 285 Sector CF-Q c5-3; Col 285 Sector HF-G b11-4; etc	No	0.5
HIP 22031	41.34	3	1.5	HIP 21548; Col 285 Sector LL-E b12-5; Col 285 Sector CF-Q c5-3;	No	0.6
HIP 18271	41.46	1	1	Col 285 Sector QX-U d2-48;	No	-0.5
Col 285 Sector FU-H b10-2	41.48	1	1	Col 285 Sector IA-G b11-2;	No	0.2
Col 285 Sector FU-H b10-3	41.50	1	1	HIP 20807;	No	0.1
Col 285 Sector LL-E b12-6	42.06	1	0.2	Col 285 Sector LL-E b12-5;	No	0.8
Col 285 Sector SM-C b13-4	42.10	1	1	Col 285 Sector JG-O c6-18;	No	0.2

Another recovery check needed at Unktety, and as is usual at Thor, this prediction is very dependent on keeping track of which Control systems are untapped from last week.

-------------------------------------------

Calculating Raijin
Read 162 control systems and 830 targets.

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
Aurus	25.72	7	7*	Pegasi Sector PI-S b4-0; Pegasi Sector KH-V c2-20; Iduni; etc	Yes	0.1
Pegasi Sector KC-U b3-3	26.68	1	1	Pegasi Sector KC-U b3-2;	No	-0.8
Pegasi Sector NY-O a7-4	26.72	11	11	HIP 115162; Vistnero; Pegasi Sector MY-O a7-5; etc	No	0.3
Pegasi Sector OE-N a8-1	26.88	10	10	Pegasi Sector TK-L a9-2; Pegasi Sector HH-U b3-2; Pegasi Sector PE-N a8-3; etc	No	-0.3
Ngolite	26.89	9	9	Pegasi Sector JH-U b3-8; HIP 113535; Pegasi Sector RE-N a8-0; etc	Yes	0.8
Pegasi Sector EB-W b2-7	26.93	6	5.7	Pegasi Sector IH-U b3-4; HIP 115162; Pegasi Sector DG-X c1-6; etc	No	-0
Pegasi Sector QZ-M a8-0	26.94	1	1	Pegasi Sector HH-U b3-4;	No	-0.8
HIP 116052	26.96	4	4	Pegasi Sector LC-U b3-3; Pegasi Sector GW-W d1-112; Wanmi; etc	No	0.1

Recovery check needed at Aurus but this is looking pretty solid otherwise.

-------------------------------------------

Calculating Hadad
Read 153 control systems and 839 targets.

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
Muruidooges	18.37	6	6*	Col 285 Sector RH-B b14-2; Col 285 Sector VN-Z b14-0; Montioch; etc	Yes	0.6
Bi Dhorora	18.47	5	4.2	Col 285 Sector VN-Z b14-0; Montioch; Omumba; etc	Yes	0.3
H Puppis	25.93	1	1*	Col 285 Sector SX-Z b14-1;	Yes	0.2
Neites	25.94	3	3	Col 285 Sector TS-Z b14-5; Col 285 Sector TS-Z b14-4; Col 285 Sector TS-Z b14-0;	Yes	0.8
Col 285 Sector VT-R d4-78	26.76	1	1	Col 285 Sector YY-X b15-8;	No	-0
Col 285 Sector YY-X b15-1	29.55	1	1	Col 285 Sector OC-L c8-11;	No	0.1
Col 285 Sector OM-B b14-7	29.57	1	1	Col 285 Sector JW-M c7-11;	No	0.8
Col 285 Sector XN-Z b14-5	32.13	8	8	Col 285 Sector VS-Z b14-1; Col 285 Sector KW-M c7-24; Col 285 Sector XN-Z b14-1; etc	No	-0.8
Col 285 Sector KW-M c7-20	32.23	8	7.1	Col 285 Sector VS-Z b14-1; Col 285 Sector KW-M c7-17; Col 285 Sector KW-M c7-24; etc	No	-0.7
Col 285 Sector PB-D b13-5	32.34	3	3	Col 285 Sector TH-B b14-1; Col 285 Sector GQ-O c6-19; Col 285 Sector TH-B b14-0;	No	-0.9

A couple more for the Recovery timer checks here. Muruidooges and Bi Dhorora are very solidly targeted (though with some overlap) but the planned attacks on H Puppis and Neites could be cut off by retaking the rest of TS-Z subsector as well, so I'm expecting this one to end on the "revise later" list anyway.

 

[Planewalker]

This is so wild.

 

[Dev Gnoll]

Looking at what we cleared this week, the predictions for Hadad really are out of date :}.
Still going to guess it's back to Bi Dhorora and Muruidooges

 

[Aleks Zuno]

Looking at what we cleared this week, the predictions for Hadad really are out of date :}.
Still going to guess it's back to Bi Dhorora and Muruidooges

INIV are doing what we can to make them as outdated as possible, of course! At minimum we will make M. Hadad shrink by evicting more than five Control systems each cycle, and it is well that those choices can also be made in a way which helps keep them out of populated systems, in effect downgrading some of those Alerts to uninhabited and moving them further from the Maelstrom. With thanks to Commanders who are defending the inhabited attacks, I hope those cases are now modest enough to rely on their completion for the moment, and I suggest holding Bi Dhorora as an Alert and leave Muruidooges to go to Invasion.

The defended Yan Zangata will be another regular on these lists for the moment, though notably it will be soon time to guard the nearby HIP 28150 once out of Recovery, which will in turn erode the possible Yan Zangata attackers next time. Speaking of HIP 28150, I have just noticed its proximity to Col 285 Sector JW-M c7-11 above, so I think JW-M c7-11 may become our closing target this cycle (another open-space harvest!).

We are also making an effort to record some of these eviction difficulties now, with a few interesting initial hypotheses/conclusions! With difficulty given in units of Cyclops samples:

• The completion requirement also depends on the distance to Sol, not just the distance to the Maelstrom. We can see this in records for week 20, where RX-Z b14-0 and VN-Z b14-7 have almost identical Maelstrom distances (27.41 and 27.48) but very different difficulty (342–360 and 423–450), where the unexpected higher number can be explained by considering the Sol distances (177.06 and 191.8).
• Holding the Sol distance constant, it appears to be a simple exponential decay as Maelstrom distance increases. The least wrong shape estimate thus far for Maelstrom distance x seems to be 460000 e⁻⁰⋅³ˣ, and this seems to have a vertical shift based on Sol distance. We do not have enough results to suggest a shift expression, though it seems to be around +230 at most for ~177 Ly, +280 for ~183 Ly, and +320 for ~195 Ly. Those numbers are particular to M. Hadad (~202 Ly), so any normalisation involved would need also to be deduced before that can be used elsewhere.
• There seems to be a fixed difficulty lower limit of 360. We can see this when comparing RX-Z b14-0 last cycle to NR-B b14-0, which we have just now taken this cycle. Both are almost exactly 177 Ly from Sol, and both had 360 difficulty, however their Maelstrom distances are notably different (27.41 and 30.36). Using the above shape and supposing that the former result was on the cusp of estimating 360 anyway, the latter then estimates around 280 while in reality still requiring 360 (288 samples moved it 80%).

 

[Ian Doncaster]

Looking at the recaptures: the ones in Taranis and Oya so far don't change the predictions; if Hyades Sector ZZ-O b6-1 is finished off it weakens the attack on HIP 26274 but not enough to prevent it, so no need to issue a revised list there either.

Revised prediction for Hadad:

Target	Distance	Controls	Confidence	Attackers	Inhabited?	Advance
Muruidooges	18.37	6	6*	Col 285 Sector RH-B b14-2; Col 285 Sector VN-Z b14-0; Montioch; etc	Yes	0.6
Bi Dhorora	18.47	5	4.2	Col 285 Sector VN-Z b14-0; Montioch; Omumba; etc	Yes	0.3
Col 285 Sector OM-B b14-7	29.57	1	0.2	Col 285 Sector JW-M c7-11;	No	0.8
Col 285 Sector XN-Z b14-5	32.13	8	8	Col 285 Sector VS-Z b14-1; Col 285 Sector KW-M c7-24; Col 285 Sector XN-Z b14-1; etc	No	-0.8
Col 285 Sector KW-M c7-20	32.23	8	7.1	Col 285 Sector VS-Z b14-1; Col 285 Sector KW-M c7-17; Col 285 Sector KW-M c7-24; etc	No	-0.7
Col 285 Sector PB-D b13-5	32.34	3	3	Col 285 Sector TH-B b14-1; Col 285 Sector GQ-O c6-19; Col 285 Sector TH-B b14-0;	No	-0.9
Col 285 Sector MR-M c7-22	32.82	1	1	Col 285 Sector UC-B b14-4;	No	-0.7
Col 285 Sector WD-Y b15-7	32.82	1	1	Col 285 Sector YY-X b15-4;	No	0.3
Col 285 Sector SX-Z b14-8	32.84	1	1	Col 285 Sector YY-X b15-2;	No	-0
Col 285 Sector TX-Z b14-2	32.86	1	1	Col 285 Sector TX-Z b14-0;	No	-0.4
Heikegar	32.86	1	1	Col 285 Sector NR-B b14-0;	Yes	0.9

I've manually marked OM-B as low confidence based on the intent to cut it off at JW-M. If the counter-attack on JW-M succeeds, we'll need to read the table without the OM-B line when assessing prediction success.

EDIT: with NR-B b14-0 cut off, Heikegar is safe, the next extremely unlikely reserve is the uninhabited L2 Puppis system at 33.07 LY.

 

[Aleks Zuno]

Heikegar

32.86

1

1

Col 285 Sector NR-B b14-0;

Yes

0.9

I had wondered whether preventing too many Alerts closer to the Maelstrom could then endanger a farther populated system. That said, aside from being well beyond fifth place, we already cleared NR-B b14-0 preemptively!

 

[Ian Doncaster]

I had wondered whether preventing too many Alerts closer to the Maelstrom could then endanger a farther populated system. That said, aside from being well beyond fifth place, we already cleared NR-B b14-0 preemptively!

Probably not too much of a risk in the short term, by the looks of things.

In the long-term the recaptured uninhabited controls are going to be alerted again, so if you're taking them out on their first control week to maximise cooldowns, once you get to 30 of those broken links (6 weeks * 5 alerts) Hadad will start putting all its effort into trying to rebuild those links for further attacks. Or if they have favourable difficulty levels, you could block the inner 15 of them at the Alert stage and generate that stalemate with a faster cooldown cycle and slightly higher average difficulty to maintain.

In practice, since in the short term Muruidooges and Bi Dhorora are likely to be cleared at either Alert or first Invasion week for a 3 or 4-week cycle, and have too many deep controls targeting them, you probably won't quite need 30 others.


(This is where the original Thargoid budgeting was a lot more balanced in their favour - if blocking an inhabited alert generated four uninhabited ones elsewhere, it'd make it a lot more practical for them to get around this sort of walling-off given sufficient time)

 

[yttrbio]

Only 4 alerts for Hadad this week, and I'm not sure why. Col 285 Sector YY-X b15-4, Col 285 Sector YY-X b15-2 and Col 285 Sector TX-Z b14-0 should all have been able to put out alerts, and none did.