One of the things I really like about Fallout 4 which makes it stand out from Fallout 3 and Fallout: New Vegas is the integration of the settlement development and alliance system into the first person role playing game of the Fallout world. This has been, very much, the next logical step for this kind of game for a very long time and it seems clear to me that a lot of effort has been put into getting this off the ground. As with most new product developments, there is ample room for improvement and, with this in mind, I have some suggestions which I think are doable. I'm going to present this point by point but, to avoid this turning into the Great Wall of Text, I'm going to break it up by topic:

Minuteman Administration

It would make far more sense if the player, in the role of leader of the Minutemen, had administrative powers and some indicators of rank among the men. During the course of a game, it is possible for a more pragmatic player to accumulate a reduntantly enormous arsenal of "confiscated" weapons. Presently, these do very nicely as scrap for raw materials but I think it would be just devilishly fun if the player was faced with the choice of distributing the weapons according to a strategy which would require the presence of a command structure. This would require the player, in the role of leader of the Minutemen, to be able to:

1. Call a general assembly at a particular place and time (e.g. the Castle precisely 24 hours after the call is made)
   
   


2. Call a local assembly at a particular place and time (e.g. any settlement where the section is assigned precisely one hour after the call is made)
   
   


3. Identify different ranks and the distribution of office among the body of men. Just a little visible variation in the colours of the uniforms is all that's necessary or, better still, the addition of armor crafting station options to recolour various uniforms and clothes and armor pieces so the player can alter uniform appearance according to rank and loadout.
   
   


4. Trade gear to upgrade weapons and armour using rank or company-platoon-squad designation as a marker to ensure, for example, that every squad has a special weapon and every section has a heavy weapon.
   
   


5. Squads in the field must always have a corporal in attendance and sections in the field must always have at least two corporals (preferably three corporals) and one sergeant - which, as a logistical element, ensures predictability of distributed weapons and numbers in the event that the player decides to take a more active role in outfitting the Minutemen.
   
   


6. It would also make sense for the player to recruit minutemen replacements from amongst settlers and hand out promotions and demotions as necessary to organize minutemen.
   
   


7. An interesting variation on this theme would be the progression of each minuteman from recruit, to regular, and finally to veteran (perhaps indicated by a number of stars or chevrons in the title for the NPC) with the survival of each engagement. (with attendent increases in skills, depth of combat AI and overall toughness).
   
   


8. Squads and sections need to be named and identifiable so the player can keep track of the ones seeing more action. One way of doing this follows an alternative to maxing charisma to unlock the 20 settler limit by also allowing this limit to be unlocked if the player has assigned a section of minutemen to every settlement. Once a section is assigned to a settlement it could be automatically named after the settlement (e.g. the Spectacle Island Detachment) with squads assigned names like, e.g., Spectacle Island Alpha, Spectacle Island Bravo, Spectacle Island Charlie, and Spectacle Island Delta.
   
   


9. It might make more sense if the player progressed gradually toward Minuteman General rather than all at once. After all, it doesn't really make a lot of sense to call a leader of a company or even battalion sized force a "General" when Captain or Major would be more appropriate to the command level. This creates an opportunity for a very easy to implement chain of advancement based on the expansion of the minutemen which the player could be allowed to facilitate through the recruitment of settlers and assignment of Minutemen sections to various settlements.

On the subject of administrative labels: Label variations such as "Minuteman", Minuteman Corporal, Minuteman Sergeant, Minuteman Lieutenant, Minuteman Captain, etc. would be an appropriate way to keep track, although, the same thing could be achieved without ranks by assigning codes to identify squad, section, platoon and company assignment e.g. "Minuteman D3A2a" could designate corporal of the second squad, A-Section, Third Platoon, D-Company (D3A1a might be used to designate Lieutenant of Third Platoon, D-Company - at least until someone realised that it undermined signal encryption by providing too obvious a crib). Irrespective of the system in play, this kind of detail is an important command marker to tag who patrols with who and where; given that the Minuteman Leader will be wanting executive powers to give specific assignments to specific squads, sections and, perhaps later in the game, entire platoons. A big part of this is knowing that a body of men will have x automatic weapons, y long rifles and z heavy weapons because, given the current game mechanics, it's combined arms -specifically in the optimum combinations- which can make units, even small units, both stronger and more versatile.

Minuteman Executive Command

1. Time sensitive missions need to remain permanently on the HUD until discharged: either by completion or by delegation - otherwise they are too easy to miss (which says good things about the level of immersion in Fallout 4, by the way). I think that the fact that the same does not apply to other missions is a good way to signal that the mission in question is on a deadline.
   
   


2. The development of sufficient authority to delegate most of the radiant missions (from the pipboy, remotely), if desired, is something vitally important to preserving the whole "do what you want" nature of the open world theme and allow the player the option to either follow the radiant missions wherever they go or do her/his own thing and let someone-else deal with the radiant missions as they come up. I think that these radiant events make an important contribution to the broader backdrop in terms of events impacting on the settlements but there needs to be choice for the player's level of commitment/availability for radiant activities. Such options could be effectively implemented as benefits earned through specific achievements not necessarily tied to the Minuteman campaign and certainly not tied to progress in the main questlline.
   
   


3. Once the player becomes leader ("General") of the minutemen (on any command level), it would make a lot more sense for the player to be able to assign squads or sections to specific settlements (which would function to obviate player attendance at the settlement in question in the event of an attack).
   
   


4. When the number of linked settlements reaches 10 or more, it is quite plausible for the player, in the role of Minuteman leader, on or beyond the Lieutenant command level, to be able to delegate a radiant settlement defense mission to a squad or, if in command of more than 20 settlements, a section of Minutemen.
   
   


5. When the number of linked settlements reaches 20 or more, it is quite plausible for the player, in the role of Minuteman leader, on or beyond the Captain command level, to be able to delegate a radiant punitive/pre-emtive strike mission to a section of Minutemen.
   
   


6. When an attack is pending on a settlement which doesn't need player intervention, because the settlement already has a section of minutemen assigned by the player (as per point #3 above), it is still necessary to inform the player as a courtesy (in case the player would like to come across and watch the fireworks or, perhaps, even participate) and also to aid the player in keeping track of which minutemen detachments are getting the most action (and, thus, have a greater need for fresh troop replacements). Such notifications will not always be effective unless delivered as a messagebox which has to be dismissed by the player before it can disappear from the HUD. If there is one thing about war that never changes, it's attrition.
   
   


7. Most of Rome's war was not punic. Likewise, the frequency of punitive/pre-emtive strike missions needs to be an order of magnitude less than the frequency of settlement defense missions if this mechanic is to make sense.
   
   

Kidnappings

1. Likewise, kidnappings need to be at least an order of magnitude, preferably two orders of magnitude, less frequent than settlement defense missions.
   
   


2. In the case of kidnappings, I've heard tell that there is a statistical 48 hour deadline irrespective of what the kidnapper claims - and there is always a significant delay between when a kidnapping takes place and when it is reported - so if this kind of mission is going to make sense, there needs to be an uncertain deadline with 65 percent of deadlines falling between 12 and 48 hours tailing off so that 95% of the deadlines fall between 3 hours and 72 hours (i.e. Poisson distribution) without notice to the player of exactly how much time is available. Given the difficult circumstances and challenging execution, you can see why something like this needs to be treated as a special or very rare mission, no?
   
   


3. One other thing I've heard about kidnappings is that paying the ransom is the surest way to get the captive killed. The kidnappers, in nearly all cases, simply take the money, kill the captive and run. If this is the case, I don't think anyone's being done any favours by propagating the myth that a kidnap victim has any likelihood of survival if the ransom is paid - even in a work of fiction; bearing in mind Monkshood, The Elder Scrolls and water under the bridge. In this very specific regard, the deception involved in real kidnappings is simply too serious to leave any room for credibility - even if that credibility comes from a sense of form and conventionality drawn from works of fiction - as was once implied about the portrayal of Monkshood in a certain computer game. Perhaps I am arguing the number of angels dancing on the head of a pin or perhaps this might be a good reason for making sure that the consequences attached to the "pay ransom" option communicates the correct message on this point. Either way, I thought this was worth bringing into the open.
   
   


4. As special radiant missions, kidnappings can also afford to be cast as mysteries. After all, wouldn't it make so much more sense if the location of the captive was a mystery to be solved (rather than something known by all and sundry)?
   
   


5. Clues could include such details as whether the assailants were raiders, gunners or supermutants, the number of captives taken, the bodies of dead assailants left in the field (as, during the heat of battle, some settlers might actually misidentify assailants and be too distraught to check the bodies afterwoods), etc. On extremely rare occasions, one of the bodies might have a note naming the location if the snatch operation was a complete bungle (most likely with supermutants and least likely with gunners - but rare in all cases other than Institute snatches where any "note" would not exist as more than quantum states in a chip which may or may not be recoverable from a dead synth). The most important clues would involve trace evidence and types of item uniquely associated with a particular location.
   
   


6. If the multiple of player character Intelligence and player character level is less than 300, it would make sense that this evidence would have to be collected and brought back to a minutemen lab previously cobbled together and staffed by the player with dedicated researchers who, together, could pool their knowledge to tell the player roughly where to look, rather pinpointing the precise location. For example instead of saying where the captive is being held, point out that the assailants seem to have spent a lot of time, recently in the vicinity of suchandsuch ruin. This, of course, is not the location, but just the regional center marker for which the base housing the type of kidnapper (gunner, supermutant or raider) that also houses the captive, happens to be the closest - leaving it up to the player to figure out which fast-travel marker this would be.
   
   


7. This kind of approach might seem to be not much more involved than the settler marking the captive's location on your map but I think it would make more sense and make the allotted time somewhat more valuable to the success of the mission. Especially if the researchers took roughly an hour to decide on the location.
   
   


8. I would add to this that kidnappings need to be tailored to the faction that commits them:
   
   A) The Institute wouldn't be looking for ransom - you'd simply have a limited time to find the captive before the Insititute got around to doing what it does - which may well take place in the field.
   
   Supermutants would be expected to grab multiple captives with the intention of beating them into what one sees in meatbags or maybe just consuming them outright - either way, meat rather than caps would have to be the driving motive behind supermutant kidnappings.
   
   C) Of all the factions, raiders would be most interested in caps for the promised "return" of a captive.
   
   D) Gunners are well enough organised to be aware of the fact that they don't need caps if they can just take what they want - which is the whole point of the paramilitary getup. Gunner kidnappings would only make sense to me if, in spite of issuing an unlikely ransom, they had an ulterior motive such as drawing the leader of a rival faction out into the open where it is easy to gun someone down.

Raids

1. Raids on settlements need to begin outside the defense perimeter and work their way inwards. If a player happens to be present at a closed settlement and the hostiles make their initial appearance inside the walls it makes no sense - especially in the case of supermutants and ghouls because none of the foregoing could possibly disguise themselves as wandering settlers or provisioners; certainly not well enough to get past the gate and only the Institute has a
   
   Spoiler
   

   teleporter

   
   
   .
   
   


2. Gates need a bit more work. There's not much point in a gate if anyone and anything can open it. Gates need to be set up so they can be barred (like some doors) and the bar on the gate needs to be set up so that a gatekeeper can be assigned to it from amongst the settlers so that the gate only remains barred to hostiles. Conversely, settlements which have a barred gate should not be networked via a provisioner until the bar on the gate is assigned a gatekeeper.
   
   


3. Gatekeepers would also be much more effective if, at the commencement of an attack, they made sure the gate was barred before going off-duty and taking up a strictly non-combatant (ie. run and hide) role until the conclusion of battle.
   
   


4. Raids need to take place in at least two stages - perhaps three.
   
   A) The opening, and optional, stage would be, for the sake of playability, limited to a 5-10 minute bombardment by missiles and mini-nukes if the attackers have them and are so inclined. This would be more the case with Institute and gunners whereas it would be highly unllikely with raiders and supermutants while impossible with ghouls. Moreover, bombardments do not have individual ordinance targeting anything other than an arbitrary point in the applied fire pattern. This does not preclude the practice of "walking" a fire pattern to drive enemy troops in a particular direction or trap them in a noose (if you have brought enough batteries to bear). But I suspect that this scale of tactics might be a little big for the Fallout worlds at their present state of development.
   
   The middlegame stage is the external attack in which the vanguard comes into firing range and begins shooting at anyone in line of sight. This is not intended to break the defense but to conceal the real action which is to bring up a fatman to break the gate, missiles to break either gate ot bar depending on which goes first, or bring up a gauss rifle to take out the bar on the gate.
   
   C) The endgame stage (also the most important stage) is the breach when, having forced a gate, the fighting propagates inside the battlements. This is where one side or the other comes into a huge advantage depending on whether the defenders consider the breach inevitabie or preventable.


5. Advanced raid tactics would involve the use of a few attackers in power armour, who do not enter the fray until the breach is made, and who are there only there to punch through the fire zones, at the beginning of the Endgame stage of a raid, with the objective of bringing down generators and any turrets still in operation when the power goes down - while allowing the regular attackers to clear the outer defenses and keep the settlers and Minutemen busy as they push inwards.
   
   

Defense Strength versus Defense Rating

Defense rating is not the same as defense strength. Having said this, I can't think of a reliable way of approximating defense strength for a given settlement without the use of calculations complex enough to stall the rendering and probably requiring functions beyond the current limitations of the engine. Elements with a profound impact on defense strength include:

1. Shortness or brevity of defense perimeter (limiting breadth of approach)


2. Area inside defense perimeter (variable effect depending on stage: dilutes impact of bombardment but makes the elimination of cover from post-breach action more difficult)


3. Placement and completeness of walls


4. Angle and proportion of overlap of fields of fire from multiple turrets


5. Presence of fire corridors at points of access


6. Nature and situation of terrain relative to the architecture and layout of the settlement and its defenses

Until someone can say something more than just "it's too complicated" it's probably best not to worry too much about it and leave well enough alone. In any case, this point is just for the benefit of those who, llike myself, are wondering if there isn't a better method of calculating defense.

Anyways,, food for thought with loads of spelling and grammatical errors thrown in for FREE

Updated some of this stuff into the Settlement Critics thread.

It does sound as though you want this game to feature much more strategy game features. I'm not opposed to it but very little here seems feasible, and does sound like it would require a large chunk of devs for future games. Some of them would be great, especially the attack-phases, would give you a good reason to set up artillery at settlements if they are the exclusive method of retaliating against the first phase.

Free bump!

I concur. Many of these ideas would offer improvement, but implementation would be nigh impossible.