Harebrained Initiative (86 pages later)

[TL;DR? Weighing pros and cons, it's mostly pros, and the cons are arguable. I want to tweak how the "weight classes" are defined, though, which is why I'm really posting.]

It took a few weeks, but I finally finished reading this eighty-six page discussion of Harebrained Schemes' initiative system.

Regular BattleTech tosses a coin at the top of each round and then alternates between players, waiting to resolve weapons fire for all units en masse.

HBS BattleTech resolves each unit's weapons fire individually after it moves, and gives lighter weight classes the option to activate before heavier weight classes.

There's a lot to like about the HBS system:

• It mitigates the effects of always winning or losing the coin toss by making faster-activating units less powerful than slower-activating ones. 
• You can break up enemy activations (or bunch up your own) by reserving your early-acting units to later phases.  
• Your influence over activation order is intrinsic to force composition. 
• Bonus: It adds value to light and medium 'Mechs, which tabletop players often criticize as too weak relative to heavy and assault 'Mechs. 
• Meh: It matches the fiction better (events are rarely depicted as simultaneous - characters tend to take a hit from their enemy, then maneuver cleverly and fire a return volley). 

Other effects are less clear-cut, and I did see two substantive objections. 

Persistent Computerized Mech Worlds

For the rest of this to work, I need to distinguish the winningest human players from spoilers propped up by sockpuppets and robots...

Suppose that before each match, every player's current rating is degraded by some fraction according to their total number of matches to date (i.e., multiply by (n-1)/n), and degraded by some other fraction according to how many days (weeks) have elapsed since the last match; then after each match, their ratings increase according to their level of success, the opponent's rating, their total number of matches to date, and how frequently they encounter this particular opponent (ie, add success*opponent's rating/n/frequency).


Self-Organizing Factions:

Players rate each House (macro faction) on a scale from +100% to -100%. As they go about normal play, these ratings indicate how much the player's activity helps stabilize or destabilize each House. Each House has a baseline territory, and the players' aggregate activity will cause Houses to expand or contract (with diminishing returns) as a ratio of that baseline territory. This affects the availability of equipment, personnel, and other game assets within that House.

Global Events could be generated in a few ways, and should occur with some regularity. Some of these will allow for permanent changes to each House's baseline territory.

Normally, the player's material gains or losses are completely dependent on the player's own victories or defeats. However, players can choose to base a portion of their gains or losses on the success (in varying proportions) of up to eleven other players. This relationship is one-way, and does not diminish the assets of these other players.

The server will periodically map these connections. Each player's rating will be multiplied by whatever % they assigned to the eleven other players, and then added to those player's ratings, which is multiplied by those % and passed on, and so on. Each player's final value is then divided according to the player's factional loyalties to generate a rank (sergeant, colonel, marshall, etc) for each friendly faction. (The value might similarly be used to periodically put a bounty on the player's head for enemy players to collect.)(These values should possibly be divided by faction to start with, to prevent enemy officers manipulating another House's internal balance.) Rank determines how much influence each player's actions have on global equipment and events throughout the faction; highly ranked players may also have a great deal more specific regional control.

Large scale (or highly ranked) movement of players from one faction to another may be treated as defection, secession, or civil war.


Self-Balancing Battle Value:

The winningness of every 'Mech, the average lifespan between purchase and destruction, and the average cost of repairs per usage will be tracked continuously across all matches and players. 'Mech winningness will be weighted according to player winningness, with highly rated matches counting highly and low rated matches counting little. It will be used to periodically update 'Mech BattleValues, similar to how player ratings are updated. The updated BattleValues will be compared to a baseline (universal across all 'Mechs) ratio of lifetime value : lifetime cost, and used to rescale each 'Mech's purchase price and upkeep costs.

Purchase price and upkeep will have small random deviations on a 'Mech by 'Mech basis, will vary according to House loyalties, and will step incrementally toward their new calculated values over time. Availability of 'Mechs and parts will vary according to House loyalties, recent missions, and possibly other factors; there should be an overall tendency for a player's force to regress to House norms, thereby giving new players a chance against entrenched players.

Character generation for MechWarrior, Transformers, or Star Trek (part 2)

As a rule of thumb, characters should be as likely to lose their Individual Resource ('Mech, informant network, etc) as they are to lose their own arm, and as likely to have it replaced as they are to have their arm replaced.

When a resource is lost, roll to see how long it will take to replace (or if it will be replaced at all). Then reduce the resource score by 1 and, if necessary, reroll the specific item, rank, etc.

Items or resources gained via shared skills (such as Driver or Small Arms) indicate shared resources (such as a shared truck or common armory) which can be lost and replaced more easily.

The following skills, when allocated as an Individual Field, have a particular effect on resources:

Streetwise: ability to navigate the bottom of a social pyramid (common merchants and laborers, low-level military personnel, common criminals, whatever). The Individual Streetwise Field indicates cooperative relationships with spies or information brokers. The breadth & usefulness of this network is proportionate to Resources; how quickly the character can connect with a local network depends on skill. Expands rarity of this individual's Field items and shared items.

Bureaucracy: ability to take advantage of local laws and regulations, and efficacy when requisitioning supplies. The Individual Bureaucracy Field indicates a trusted place in the local bureaucracy or on-planet garrison. Resource points indicate how much intelligence and non-hazardous assistance (licenses and facility access vs. restricted information and rescue from legal or administrative difficulties) the character can get without risk; the speed and cost (in bribes or favors) of the assistance depends on skill. Expands legality of this individual's Field items and shared items.

Protocol: ability to navigate the top of a social pyramid, and to negotiate contracts. The Individual Protocol Field indicates a place in the interplanetary House hierarchy. Ability to improve the party's status or future assignments is proportional to Resources. Improves selectivity of this individual's Field items and shared items.

Leadership: ability to command NPCs, keep NPCs loyal, and expand the amount of situational bonuses party members can give each other. The Individual Leadership Field indicates higher rank, or a more prestigious lateral assignment, with authority proportional to Resources.

Administration: ability to keep an operation running smoothly. The Individual Administration Field indicates a land grant, business or other interest which generates surplus revenue in proportion to Resources. Elevates this individual's Field items and shared items into a pricier tier.

BattleMech [or Tank/Starfighter/Spaceship/Etc] Operations: ability to pilot the vehicle and operate its sensors, weapons and other systems. (Note that the "driver" common skill applies to ubiquitous everyman transportation, though [Vehicle Type] Operations may be a common skill as well in some settings.) Resource points indicates how much regular support the character's vehicle receives from a larger military. This may be a little high or low for the vehicle actually assigned.



Transform [Vehicle]: Transformers gain their "resource" bonus only when transformed. The bonus applies not only to Individual Field tasks, but also to any tasks which are dependent on (or characteristic of) their vehicle type.

Transform [Gestalt]: All characters on the sheet combine into a single larger robot, which receives the "gestalt" bonus to strength-based tasks and to firepower.

[The way that these Individual Fields affect the character's supply situation is inspired by the way that a user called Monbvol uses the Vehicle Trait in the 3rd edition MechWarrior RPG.]

Character generation for MechWarrior, Transformers, or Star Trek (part 1)

Multiple characters per player. The player has 10 points to divide between the Individual Fields of his characters; another 10 points to divide between the Shared Skills of his characters; another 10 for Background; another 10 for Resources. The points each character spreads across Individual Fields, Shared Skills, Background and Resources must combine to the same total as each other character. If four characters, then each character must total 10 points; if five characters, then each must total 8 points; if six characters, then four must total 7 points and the others must total 6 points.


Using "Falco" as an example:

Individual Fields describe a broad area of expertise. Central tasks (e.g., Gunnery, Piloting, Sensor Operation) are made as trained rolls with full skill bonus. Fringe tasks normally associated with other skills (paperwork, repairs) are made as trained rolls but without any skill bonus. Field tasks which somehow relate to the character's Shared Skill (e.g., firing a 'Mech weapon carried like a giant pistol) receive the Shared Skill bonus as an additional bonus.

Shared Skills are basic training common to every character on the sheet (e.g., all four characters "Drive" as a 2-point trained roll). They also indicate the strongest attributes of each individual character. Any untrained roll which relies on the same attributes as one of that individual's Shared Skills receives the Shared Skill's points as a bonus (e.g., Small Arms requires steady hands and good eyes, so "Falco" gets a 2-point bonus when catching a thrown object, which also relies on steady hands and good eyes).

Background skills represent the characters' upbringing, affiliations, hobbies, or other specialized fields of knowledge. Social class or geography/climate indicate familiarity with associated culture and survival skills; affiliation skills represent knowledge of culture, language, and specific social hierarchy. Ridiculous future sports, future pop media genres, and specialized careers (gastroenterology, archaeology) are background too.

Resources indicate the quality of equipment all characters receive for any Shared Skill any character has, and the quality of equipment individual characters receive for their own Individual Fields. (E.g., all four characters receive a 1 point medkit and 1 point survival kit; "Falco" alone receives a 4 point 'Mech.)

2 points of skill is equivalent to "Green" MechWarrior skill.
4 points of skill is equivalent to "Regular" MechWarrior skill.
6 points of skill is equivalent to "Veteran" MechWarrior skill.
8 points of skill is equivalent to "Elite" MechWarrior skill.

[Edit: "Advanced Field" and "Basic Skills" renamed to "Individual Fields" and "Shared Skills," respectively.]

Apropos of Nothing

I have bunches of .txt files containing notes for writing an RPG, in the (increasingly hypothetical) event I should want to write an RPG. Those notes can be divided into certain broad categories. I wonder if the proportion of notes I have on each topic corresponds to the proportion of pages I would want an RPG book to devote to those topics. 

(Crunch week for me, so have some stream-of-consciousness.)

Instead of Vanilla 2d6

My work back here rolling "lower of two dice" made me wonder if I could emulate a 2d6 curve that way. " 7 + (lower of 2d8) -- (lower of 2d8) " turns out to be almost indistinguishable from a 2d6 curve! Were I actually using this die mechanic for something, I'd be tempted to have results of 0 and 14 "explode" with further "lower of 2d8" rolls.

Tabletop play seems to have a tension between providing interesting results (usually means increasing numbers of dice) vs. keeping record keeping reasonable (means keeping pip totals small). World of Darkness and Shadowrun solve this by rolling lots of dice and counting each die as a pip; Betrayal at House on the Hill is similar, in that its dice are numbered from 0 to 2. ...I guess my personal holy grail here would be a simple, manual die mechanic that'll produce a smooth 1/x curve.

2d6 is really easy to use on the tabletop. It does have some problems--cannot roll higher than 12 or lower than 2, "lucky" streaks are pretty common, and "unlikely" results may be more or less common than players think appropriate.

My first thought is that I can break "lucky" streaks by rolling lots of times and discounting all results until a given value has been discounted a certain number of times. (The quota for each value would be proportionate to the chance of rolling that value.) This would be terrible to execute manually but could work with a computerized die roller.

My second  thought is to replace 2d6 with 4d6 / 2. This makes middling results more likely and extreme results much less likely.

I'm not sure I'd want to use any of these as a straight replacement for 2d6 rolls in (say) BattleTech, but I think it might be interesting to designate individual pilots as cautious (4d6/2) or swingy (quotas).
(I'd like to eventually do my own computerization of BattleTech, and I have some thoughts on how I'd modify specific BattleTech mechanics, but that's not a project I can start soon.)

Clue Variant: Death Clue

Have I mentioned Death Clue before? I feel like I have.

In regular Clue, you set three cards aside at the beginning of the game, and the players try to deduce what those three cards are by figuring out what cards are in each others' hands.

In Death Clue, each player is dealt three cards and the name of ANOTHER player, tries to deduce what THAT player is holding in their hand, and is required to lie (pass) when THAT player asks for clues. Goal is to "kill" THAT player by guessing their hand, but you can also "kill" whoever's hunting you. Everybody (usually) survives to the endgame and the endgame is pretty quick.

"Resistance: Avalon" is my Sunday group's go-to game. It's fast, easy, a little bit of strategy and lots of persuasion. For us (~5 or 6 players) it tends to be one-sided, and I'm pondering how to change that.

Current notion is to hybridize it with Death Clue while preserving the talky/cooperative aspects. (Don't know how. Might not be possible.)

Initiative

The fastest ways I've seen to resolve initiative are preplotting (secretly record all unit declarations simultaneously) and the playing card method (each unit acts when its assigned card is drawn).

What I'm going to propose is not as fast.

0. Before the game, assign point values to all units (say, BV2/100, round up). Multiply all units in the lower valued force by a factor which brings their total even with the higher valued force. Round any unit valued at less than a single point to tenths of a point.
1. Some units may be required to declare before standard initiative begins. They do so now. If both players have such units, the players alternate as in step 4.
2. At the beginning of any phase in which initiative matters, each player may designate any enemy units as "exempt" from standard initiative.
3. Each player secretly records declarations for as many of his own non-exempt units (at least one, assuming he has at least one non-exempt unit) as he wants, and tallies up their point values. Both players reveal and resolve those declarations.
4. The player whose tally is lowest openly declares actions until his tally exceeds his opponent's. Repeat until all non-exempt units have been declared.
5. Declare actions for "exempt" units, alternating as in step 4. (Given that players will tend to "exempt" enemies they don't care about, these could be resolved quickly and simultaneously.)

Simple Fantasy Halfbreeds

Progress Quest is a weird and satirical yet very authentic experience. As my Double Wookie Battle-Felon climbed to level 8, I got to looking at the game's racial types. They're really not that far off from what D&D-style games actually do. I've tried before to distill D&D3.X's sub-races into more efficient templates; looking back at that, I now realize that race and alignment can be collapsed directly into the base ability scores.

• "Elf"-ish ability score = dexterity, agility, foot speed, personal beauty, vision, hearing, affinity for & resistance to musical / mental magic.
• "Dwarf"-ish ability score = physical health and resilience, sprints, sense of position / direction / elevation / orientation, mechanical engineering / architecture, affinity for & resistance to runic (artifact-based) magic.
• "Orc"-ish ability score = size, strength, stamina, sense of smell, affinity for & resistance to blood magic.
• [Elemental] affinity score = terrain-specific survival & peasant skills, discipline, piety, ethereal awareness, affinity for & resistance to pact-based magic. (Choose an element: fire, water, earth, wind, wood, desert/void, light, dark, etc.)
• "Society" ability score = domestication, class, literacy, common lore, device skills, teamwork, leadership, loyalty, sensing motives, affinity for civilization & technology.
• "Detachment" ability score = coolness in the face of danger and horror, resistance to plight of others, resistance to magical healing and resurrection. (Also replaces stereotypical Cthulhu-esque insanity stats.)

Balancing the three bloodlines creates a human; raising the highest stat to double the second-highest creates a pure ("double") breed; in between lie all the shades of half-breed; terrain-specific [Elemental] breeds can give way to pure elemental sprites; "society" can indicate "high" or "low" breeds; hobbits, pixies, gnomes and goblins amount to "short" breeds; and any of the various animal-faced races can be emulated by relabeling Elf, Dwarf or Orc to the animal type. Use either your "society" score or the best-shared of the first four stats to try communicating in someone else's language.

"Detachment," "Society," and the ratio between the two offer a more concrete and nuanced view of "alignment" than the standard good-evil / law-chaos layout does. They don't address the character's goals or commitment to the party, but those things can make or break a campaign, and really ought to be hashed out with the group before play begins.

Without complex character builds, where will players turn to flex their stat-wonkery? I think it would be useful to treat lair design & stock scenarios the way character creation has been treated - point builds, random generation, class/level systems, with a few new options or ideas costed out in each splatbook to help sell more splatbooks. Can't hurt to turn player attention to something that's routinely in short supply, and to have your books always trying to address new & different aspects of an encounter.

PS: I prefer how D12s roll to how D20s roll. So, instead of rolling 1D20 against dodge+armor (or whatever action you're taking), roll 2D12 and assign one die result to accuracy and the other to potency.

Play a Short Battalion as Easily as a Lance

Provided that you're not too keen on 1:1 accuracy with the fictional setting, there is a very simple way to preserve the "feel" of BattleTech while fielding seven times as many miniatures:

Assign each location on regular 'Mech sheets to a separate figure.

Ignoring the head, this turns a lance into four squads of seven (twenty-eight figures total), and uses basically all the same rules as regular BattleTech. No need to roll on the hit location table, obviously, since each figure will target other figures individually; jump jets apply only to whichever figure has 'em on its crit table; heat is shared across all units in the squad; use 60-degree firing arcs, no rotation, for all figures; critical hits largely the same, except shoulder/hip crits immobilize the corresponding figure; figures occupy the entirety of whichever hex they're in, and can block LOS; ammo explosions apply full damage to the hex they're in, then apply half the excess to adjacent hexes; limbs mass 10% of the "full" 'Mech's mass and torsos mass 20%; physical attacks and weapons attacks are handled in the same phase, since each figure can only do one or the other anyways; each figure can do whatever physical attacks the location could normally, except that torso locations do 1/3rd normal charging damage.

Variations:

1. Instead of making a second roll to check for critical hits, treat a roll of doubles on any appropriate To-Hit roll as a single successful crit.
2. A squad deactivates when its "Center" is destroyed.
3. Jump MP replaces ground MP 1:1 instead of being a separate movement mode.
4. "Leg" figures can move anywhere they could normally; torsos can only move towards "legs," and cannot move beyond legs; arms must also move towards legs but can also continue beyond them.
5. Split-location equipment (ammo, autocannon spillover crits, Artemis, etc) is removed from its current location and considered to be mounted outside the armor on the rear hexside of whichever figure carries the primary/bulk of the associated weapon.
6. Legs get zero free facing changes; torsos get one free facing change; arms get two free facing changes.
7. Figures (for purpose of city fighting or physical attack damage) determine their mass by summing all equipment and armor and structure etc. in the location. Assume split-location equipment spreads mass evenly across all crits, and that the engine puts 1/3rd of its mass in the center torso and splits the remaining 2/3rd between the legs.
8. The "Center" figure absorbs the head location's armor & structure and weapons, but ignores pilot effects. 
9. Instead of the "short/medium/long" "+0/+2/+4" brackets & modifiers applying to the range between shooter and target, assume all weapons are always in range, and apply them instead to the target's movement in place of the standard Target Movement Modifier table. Green pilots subtract one hex from their effective movement for this purpose; veteran pilots add one. 

Edit, November 16, 2012:

• Alternate to variations 2 & 4: the center must remain within (Flank MP) hexes of at least one leg; each side torso must remain within (Flank MP) hexes of the center (or the hex the center was destroyed in); each arm must remain within (Flank MP) hexes of the associated side torso (or the hex the side was destroyed in).
• Alternate to variation 8: each item in the head gets assigned to different non-center location; when a non-center location is destroyed, the whole team suffers effects as though the corresponding head item had suffered a critical hit. 

Edit, August 1, 2019:

• Don't know when this dates to, but I just found an old note suggesting that each unit track heat individually and units in base contact can share cooling capacity. (Harsh.) Possibly as an alternative to the movement restrictions mentioned earlier.

Mini Campaign System

Some of these rules will look specific to BattleTech (because I'm trying to manipulate 'Mech stats into matching the fluff better), but the basic structure could probably be adapted to other game universes.

Before choosing forces you'll need to calculate three or four values for every 'Mech that will be available to the players:

Production Cost = [(cruising speed)*(armor points)+(jump MP)*(armor mass)/(mass of individual jet)+(heat dissipation)*(maximum possible heat)+(weapon and equipment BV)]/10
Ready Maintenance = 3*(number of weapons, equipment, jump jets and excess actuators) + 2*(maximum possible heat)
Active Maintenance = 1.33*(armor mass)^2/(engine mass)
Repair Cost = (C-Bill price)/100,000

These values are permanent traits of a given design and do not change during play. They're scaled so that a typical 2866-3050AD 'Mech battalion costs 3600 production points and is able to ready 90% of its force, deploy 67% an average distance, repair 22%, and replace 3% every turn.

At the beginning of every turn the players generate 3 objectives. Each objective has a random value and sits a random equidistance from both players. Roll 2d6 for each objective to determine distance, and roll 2d6*20 to determine their values.

There are two ways to play this:

Bean Counting Way: At the beginning of each turn, players add 3600 points (assuming that's how many they bought forces with initially) to their unused points from the previous turn. They must pay the "ready" cost for every unit they wish to use this turn. They must also pay an amount equal to the "distance" roll x "active" cost for every unit assigned to a scout force or a main force. Units which were inactive all turn can be repaired at the end of the turn. New units can be purchased for 8x their "Production" cost.

Fast Way: Before play begins, subtract the sum of each force's "active" costs from half the sum of its "production" cost; record the difference on a token. The force receives this many bonus points each turn (or, if negative, must pay that many points each turn). Use the 2d6*20 roll to determine the equidistance to each objective: if it comes up doubles, then players must pay 2x "active" cost to assign a unit to a scout force or main force; otherwise, pay 1x "active" cost. Deactivating an undeployed unit generates points equal to its "ready" cost. Units that were neither deactivated nor deployed get repaired for free at the end of turn. "Objective" points must be spent to purchase new units (at 1x "production" cost) the same turn they are won.

Now, roll 1d6 to determine the type of objective:

1-2 Stand Up Fight (whoever destroys the most units (powered down units count half) in 10 turns)
3-4 Territory (whoever has the most hexes closer to his ground units than to opposing ground units at the end of turn 10)
5-6 Grab (divide points into 4 equal lots on the middle of the board: any 'mech can carry points (on its back) equal to its tonnage, but if it doesn't have hands another 'mech would have to put them there; weapons fire destroys 1 point per point of damage) 

Once the objectives have been generated, both players secretly assign their units to any of three roles (scout, main body, rear guard) at any of the three objectives. The objectives are then resolved in any order the players choose. Resolve one site entirely before moving on to the next.

Any unit that exits off a neutral map edge is considered to have powered down or fled (as appropriate). 

Scouts

Determine maps randomly, then lay them single file, narrow end to narrow end. Both players reveal their scout detachment. The force with higher average speed gets to choose its home edge. Disregard any units still on the board after 10 rounds.

If scouts advance off the opposing edge, sum their Cruising MP (plus the range of any Active Probes, plus 1 for every .25 tons of "sensors"). The opposing player then matches that total by revealing Main Body units and summing their Cruising MP (plus the range of any ECMs on the revealed units). The "EW Suite" on a revealed RVN-1X counts as 4 points of Scout probe (even when deployed with the Main Body) and 6 points of Main Body ECM (even when deployed as a Scout). If the opposing player runs out of Main Body units, begin revealing Rear Guard. If the opposing player has no Main Body or Rear Guard, then the Scouts win the Objective.

For each 2.5 tons of communication gear that advances off the opposing edge, the player may ask one yes/no question about the opponent's Main Body or Rear Guard.

Any scouts that retreat off a friendly edge join the Main Body action, but do not count as "revealed" units or contribute to the Main Body's average speed.

Main Body

After scouts finish revealing units, both players secretly decide which of their unrevealed units they want to commit to the Main Body, which they want to send after the enemy's Rear Guard and which they want to send back to their own Rear Guard (units assigned to Rear Guard do not have to pay the "active" maintenance cost). Both players then reveal all Main Body units and any Rear Guard units (like artillery or Mobile HQs) that affect the Main Body engagement.

For each 2.5 tons of communication gear a player has in the Scout force or Rear Guard, the player may choose a single mapsheet. Determine the other mapsheets randomly. The Main Body with higher average speed gets to choose its home edge. Resolve the Objective as described earlier.

If a player sent Main Body units after the opponent's Rear Guard, then lay out a second map for that battle. The Rear Guard player chooses all mapsheets, reveals all forces, and deploys on the map. The attacking player chooses home edge. Play this battle simultaneous with the Main Body battle but do not end until one side or the other has surrendered or withdrawn from the Rear Guard maps.

If a player wins the Objective or advances units off the opposing map edge, then that player may choose to pursue any defeated or retreating opposing forces.

Rear Guard

Lay the maps single file, narrow end to narrow end. If the fleeing player already laid out a second map for the Main Body battle, then use those maps; the victorious survivors of that secondary battle chooses home edge and deploys their surviving forces opposite the pursuing player's home edge. If the defending Rear Guard has not yet been attacked, then that player chooses all maps and home edges and deploys the Rear Guard on the board (unrevealed units originally assigned to Rear Guard may deploy as Hidden Units). Otherwise, determine mapsheets randomly, and the force with higher average speed chooses home edges.

Fleeing units enter from the pursuing player's home edge. Pursuing units enter from the same edge on the following turn. The battle ends after ten turns (if not earlier).

Base Defense

To attack your opponent's resources directly, assign units to Scout, Main Body and Rear Guard like a normal objective. Then:

Bean Counting Way: travel the highest distance rolled for this turn x2 to destroy/steal points or inactive units worth 5x the total Cruising MP of your Main Body.
Fast Way: travel the highest distance rolled for this turn +2 to destroy/steal inactive units worth 5x the total Cruising MP of your Main Body.

Your opponent can assign "Base Defense" forces to Scout, Main Body and Rear Guard as well; however, the defender can reinforce the "Base Defense" Main Body by calling back Rear Guards assigned to other objectives. To do this, the Rear Guard unit cannot have been involved in a battle, and must pay the "distance * active maintenance" cost for whichever objective they were originally assigned to. (The Defender can also pay to "ready" otherwise unassigned units.)

VARIATIONS

Obviously my equations at the top there are aimed at 2866-3050 play, but higher tech shouldn't be too hard. Do maybe 6000 point battalions for the century before and after; count CASE as the equivalent mass of standard armor; make Gauss weapons add their damage to both heat and heat dissipation counts; I'm not sure how to account for gyro or cockpit variants. 

Could add forward bases to shorten travel distances and to avoid keeping your eggs (points) in one basket, or have your resource points travel with the units themselves. Another thing I considered was that any resource points that didn't get used by the end of a turn would move themselves 1d6 distance towards the opposing player, and would require their own set of "Base Defense."

It'd be easy enough to reflect the quality of the battalion's technicians, or their relationship with House quartermasters, by giving the player more or less resource points than were used for purchasing the initial force.

Raising the value of the Objectives, or lowering the price of replacement 'Mechs, would encourage more decisive battles (and fewer retreats).

[Edit, August 24: I've changed the "Ready Maintenance" equation so that LRM-5 boating no longer breaks even.]