Bv analysis

[DevianID]

Howdy.  Figured I'd start dumping my BV observations here.  Xotl mentioned they have BV changes in the pipeline, but maybe this article series helps someone.

[DevianID]

So first, the core weapon formula.  It has a lot going on, so let's identify it.

The basics is expected damage with a base 4 to hit at each range.  Then a multiplier for ammo/energy.  It isnt bad but has some issues.

So with a 4 base, a 5 damage medium laser expects 4 to hit at short, 6 at medium, 8 at long.  Hitting on a 4 is 33/36, 6 is 26/36, and 15/36 at long range.

This gives us 13.75 at short (3 hexes, 33/36 expected, 5 damage), 10.83 at medium, 6.25 at long.

Multiply by 1.5 for infinite ammo energy weapon, and we have 46.25 bv, which is where the 46 BV of a medium laser comes from.

Issues with this formula.  Accuracy priced here isnt consistent with accuracy priced elsewhere.  So while the range comparison part is good ( short range damage is better then long range damage), a pulse bonus accuracy is flatly cheaper within the formula then on gunnery skill increases or targeting comps.  So for consistency any accuracy a weapon provides should be priced outside of this formula.

Next is ammo.  Ammo weapons have a 1.2 multiplier instead of 1.5, and each ton of ammo is 1/8th of that.  This means, 2 tons of ammo equals 1.5x total multiplier, same as infinite shot energy weapons.  3 tons of ammo, and now the ammo based weapon is more expensive than an infinite shot weapon.

Ammo and weapon multiplier should be shot count moderated, not ammo ton moderated.  The formula gives one shot weapons a 20% multiplier on top of the 1.2 non-energy multiplier, so compared to an energy weapon a scale that goes from .24 for 1 shot to 1.5 for infinite shots makes much more sense.  Thus something like a medium chemical laser, with the above average 30 shots per ton of ammo, will pay more appropriately based on actual shots, while something like a hag30 with 4 shots total will pay less.

I like an exponential scale here approaching a limit of 1.5.  So 25 shots will be less then 30 shots, even if the multiplier is 1.48 versus 1.49 for example.

Finally, this formula misses a very big thing... The value of raw damage.  4 medium lasers have the same BV as 6 clan er small lasers, thanks to the medium lasers having 50% more range.  But, once both are im range, the raw damage of 6 ER small lasers is clearly better then 4 medium lasers.  Since the game isnt played on infinite maps, you will get into range.  No 'raw damage' addition means that 54 damage from heavy small lasers is cheaper then 3 LRM5s or 1 ER large laser.

At a minimum I would add the raw damage to all BV, so a medium laser would go from 46 to 51, making 4 medium lasers more then 10 damage at 18, and making 6 Clan ER smalls more then 4 medium lasers.  With further analysis, its possible the raw damage additive should be more.

[DevianID]

So some implications based on the weapon formula above.  Pulse lasers are too cheap.  Any one who tries pulse lasers out quickly realized they are just the most efficient thing period.  Taking the accuracy part of the formula out is a great solution to fix this, at least to make things consistent.  So a mpulse laser with 4/8/12 range with a -2, instead of costing 111, the weapon would cost 87, the value without accuracy, times 1.25 x 1.25, the same accuracy multiplier of a targeting computer.  If you also add the raw damage additive of 7 to the base, that would further get multiplied by accuracy.

Other things relating to ammo.  The elemental 2 shot SRM2 is more expensive then 4 OS srms.  That should be clear as an issue... The SRM2 with 2 shots has the same BV as 10 OS srms, and almost the same BV as an SRM2 with 50 shots.  This makes battle armor like the Rache have a jump of 4, with OS srm6, and its cheaper then the standard elemental with jump 3 and 2 shot srm2.  If the weapon had a shot multiplier in place of flat ammo, we could make a simple chart based on an exponential formula that would put 2 shots at ~35-40%, ECT ECT.

[DevianID]

This graph shows the current relationship between OS, infinite energy, and my proposed ammo curve.  We see shot count on the X axis, where ammo based weapons with few shots are very expensive topping out at 2.4 total multiplier (the weapon formula for ammo caps at 8 tons of ammo per weapon), while my proposed curve makes ammo based weapon cheaper then an equal number of one shot weapons, approaching a limit of 1.5 for infinite shots like energy weapons.

Purple is energy weapons, blue is OS weapons, red is my proposed formula, green is the limit for current ammo based weapons of a 2.4 multiplier with 8 tons of ammo, while the dark blue shows the starting value of ammo based weapons starting at 1.2+ammo tonnage as a baseline.

My proposed formula, which makes ammo based weapons never more expensive then infinite shot weapons, is
1.5-\left(.84^{x^{1.14}}\right)\cdot1.5
1.5- (.84^x^1.14)*1.5 where x equals the shot count

Edit: so to clear up potential confusion, the weapon would have its BV listed the same, but instead of Ammo BV and exception rules for OS and excessive ammo tonnage, it would be the formula above.

Thus a medium laser and a medium chemical laser would have the same listed BV of 46 for example.  But a medium chemical laser with 10 total shots of ammo per gun would then go from 46 to 42 BV if you apply my formula.  Meanwhile by the book a medium chem laser is only 37 and 1/3rd a ton of ammo is 1.5bv, which is why the overly high ammo shot count makes something like a medium chem laser too good, versus the 6 shot count LRM20, as paying for ammo in BV by ton versus shot is inconsistent.  That's why 12 medium chem lasers on a black hawk/nova are so much cheaper then 12 otherwise identical medium lasers, pricing BV by ammo tonnage instead of shot counts.

Edit: for streaks, as they have more ammo efficiency, streak shot count is doubled.  So 15 streak 6 ammo feeding 2 launchers will count as 15 each, not 7.5 each.  Also, for ammo with size modifiers, like precision, treat them as full bins not half, as the half ammo was added as a balancing effect.  Thus, 4 shots of precision on an AC20 will count as 10, the full normal amount, not 4 for the special ammo load.  For other ammo with BV changes, apply the value to the shot count.  So swarm-i will count as 1.1 times the shot count (6.6), instead of 2.25 BV more on a bin of LRM20 ammo.

Edit2: For those wondering, at 24 shots/weapon, pretty much every weapon is treated like an energy weapon (a 1.5 multiplier) and wont pay for shots past this number of turns of shooting.  The Gauss Rifle would go from 416 BV before ammo to 415 with only 24 shots for example.  So a gauss is at practically full value with 24 shots, and will only pay 1 BV to have 25+ shots per gun.  The formula caps out at ~24 shots as this feels to be the point where further ammo is excessive.  If this feels off, let me know!

[DevianID]

Now that we have unpacked the weapon BV calculator, and where the calculator needs adjustment to match the rest of the game/increase consistency, let's look at some other components of offensive and defensive BV multipliers.

For offensive multipliers, we have the speed factor.  This is fine, but the one concern here is the interaction between jump jets and MASC.  Now, Masc and Jump jets both do increase your threat profile, but the two are mutually exclusive.  The formula had units pay for both at the same time, by adding just 1 from masc, but this was changed in errata to use the maximum masc value as the run value every turn, while only using half your jump MP.  This makes a Shadow Cat omnimech have a speed factor of 15, 12 from masc +3 from jump jets.  This seems backwards--you can use the Jumpjets every turn, but masc only every other turn.  The shadow cat with 6/9(12)/6 should obviously have less of a speed factor then something that moves 8/12/6.

So a simple change is to only add half the MASC value, like how we add half jump MP.  So, a shadow cat would have 9+3 (JJ) +2 (3 extra MASC MP/2 rounded up the same as JJ).  14 may not be much less then 15, but this way a Shadow Cat pays less for speed factor then an Arctic Cheetah, which has 8/12 movement and 6 jumpjets, for a true 15 speed factor.

As for improved jump jets, a 3/5/5 unit has a speed factor of 8, a 4/6/4 unit also has a speed factor of 8, while a 5/8/5 unit has a speed factor of 11.  3/5/5 is stronger than 4/6/4, so for unit with jump > walk MP, the extra JJ should not be halved.  This means that a 3/5/5 would pay normally for 3/5/3, giving a speed factor of 7, and then each additional jump MP from wings or Improved JJ would add directly to speed factor, meaning +2 for a 3/5/5 compared to 3/5/3.  This also means that a 3/5/4 unit will pay more then a 3/5/3 unit... the current formula prices 3/5/4 the same as 3/5/3 which feels incorrect.  This also means that with my change a 4/6/6 unit has a speed factor of 10, while a 4/6/5 unit will have a speed factor of 9 (the current formula treats them the same at speed factor of 9).  Alpha strike players are familiar with the power of 'jumpstrong', so its important to include higher then walk Jump units in the speed formula if we want a more correct speed formula.

Another example.  An oscscout is 8/12/8, for a speed factor of 16.  A 6/9/9 speed unit is 9+5, for a 14 total speed factor with the existing formula, but with my change to jump over walk MP, the 6/9/9 unit would pay 9+3 (jump up to walk/2)+3 (each jump over walk MP), for a speed factor of 15, making it slightly more expensive, but not as expensive as 8/12/8.

[DevianID]

On the defensive BV side, TMM currently scales linearly.  So tmm 1 is 1.1, TMM 2 is 1.2, ECT.  However, the hit chart isn't linear on 2d6 so we can update the TMM defensive multiplier a little bit.

If weapons start at 4 base, and we use the same value for TMM, then 4-5 is 33/36 (0 tmm) to 30/36, then 26/36, 21/36, 15/36, 10/36, 6/36, 3/36.

Thus the multiplier is 1 for 0 TMM
Tmm1 = 33/30= 1.1 (no change)
Tmm2 = 33/26 = 1.27 (up from 1.2)
Tmm3 = 1.57
Tmm4 = 2.2
Tmm5 = 3.3
Tmm6 = 5.5
Tmm7 = 11 (vtols or lams can hit 25 MP plus airborne)

So with this chart, something like a very fast, nearly impossible to hit vtol unit will pay much more for their evasion as a +6 or +7 is far more defense then the mere 1.6 or 1.7 multiplier the base chart uses.  With that said, these units also should get a discount from flak vulnerability... Vtols currently have a .7 multiplier for unit type vulnerability, so while the very fast vtols will have a very large defensive multiplier, they pay less then a mech would if it could get that high of a TMM.  (A 25+ MP vtol with an 11x multiplier for TMM only actually pays 7.7x with that .7 unit Vtol type multiplier).

For Masc, currently you pay for defensive TMM as if Masc was always on, but you can't have masc/superchargers on all the time (this doesnt apply to vtol jet boosters that have no risk like a supercharger).  As a compromise, if Masc would put the TMM to the next highest category, use the average of the 2 categories.  Thus, a Dasher with 10/15/(20) and TMM 4 or TMM 5, use the average of 2.2 and 3.3 for 2.75.

[DevianID]

Lets examine the heat modifier for weapons on the offensive side of things next.  This has all weapons, after 6 heat has been reached, cost 50%.  It also goes from highest to lowest weapon BV.  Both bits of this formula are off feeling.

First, if a weapon puts you at more than 6 heat over, that weapon should be discounted, as you cant fire it turn after turn.  The discount shouldnt start only for weapons when already over the 6 limit.  This is the 'Puma' example.  The Puma's 2 ER PPCs put the puma over 6 heat when firing, but the Puma pays 100% BV for both of these guns.  With my change, the ERPPC that puts the Puma at over 6 is discounted 50%, which should help fix the Puma/Adder primes heat pricing.  But by the same token, movement heat should not count, as standing still is an option with 0 heat, so you should not get a heat discount for running or jumping when you can always just stand still.  Thus, the rifleman should not get 2 heat for running to discount its weapons.

Next, weapons are currently subtracted by order of most BV of the weapon to least, but this doesnt make sense.  This would mean that a heavy large laser that puts you over 6 (paying full price for the heavy large laser), then discounts by 50% all other weapons, such as ER medium lasers or other low heat weapons like multiple LB10x which are less BV (but far more BV efficient) then the gun that put you overheat.  An ER PPC shouldnt ever discount a Gauss rifle in the heat category.

So I would have the heat formula subtract weapons in order of BV/heat.  Thus, Gauss Rifles would always get shot first, then an ERPPC, then ER mediums, and the Heavy Large laser would be calculated for heat last, it being the least efficient weapon.  The current order is ER PPC, Gauss, Heavy Large, ER Medium, which means some mechs get wildly inaccurate heat discounts on low heat guns.  The Rifleman for example gets heat discounts on the AC5s, but the AC5 should always be fired first, never discounted.

At the same time that some things are too cheap in the heat discount, most of the time weapons are too penalized.  So beyond the first weapon that brings a unit to over 6 heat getting a 50% discount (instead of the first weapon AFTER a unit goes over 6 heat), as a unit carries more and more weapons the discount goes up.  This works because the unit is already going to be shooting its most heat efficient BV weapons (see above), so the other weapons are paying 50% of their raw cost but never shooting.  Thus, the heat discount should go from 50% down 10% each time firing that weapon would hit a new shutdown threshold, capping at 20% (the value of a OS weapons, representing the death throw alpha strike value).

[DevianID]

The next item in the offensive and defensive BV formula is the 'tonnage' add on.  One assumes that this value is the melee value, with a stability value for defense on gyros.

For the tonnage modifier for offensive BV, I would put this into the defensive category since it scales by piloting.  A kick, if it were a weapon, would have a range of 2 (hex 0 and hex 1), with damage = tonnage/10, and a bonus for accuracy.  That comes out to about 70 for a 100 ton mech, without pricing the knock down or aimed location potential, so 100 (straight tonnage) feels close enough compared to 70 before the kick bonuses.  Charges scale with MP, for things like vehicles, so vehicles paying tonnage/2 feels fine as a charge is less usable then a kick.  Tonnage also deals damage back when defending from charges/DFAs, so thats another reason to put it here.

The gyro adding defensive value seems odd, as there is no comparative for this found on things like aerospace fighters, which also need to make piloting skill rolls to avoid falling.  The heavy duty gyro doesnt make the mech less likely to fall, only makes it less bad when the gyro takes damage.  So in all cases, paying for the gyro when all it does is harm the user is silly.  Its like paying BV for adding explosive ammo.  We should SUBTRACT the value of a gyro from the defense of a mech, as the gyro is a weakness that a vehicle doesnt pay for.  So a standard gyro should subtract tonnage/2 from defensive BV, while a heavy duty gyro should subtract tonnage/4.  By the same logic hips and legs should also be negative BV, but for the time being correcting gyros from adding to defense, when all they do is hinder defense.  Tonnage is a good mechanic for this, because the heavier a mech is, the more damage it takes when it falls.

I would add a 'stability' category here.  This stability defensive modifier is where I would put all the +/- piloting skill rolls.  So quads or leg AES would get a positive Defensive stability modifier, while small cockpits or heavy armor would get a negative modifier.  This 'bucket' for bonuses is mostly for consistency, to make sure all the bonuses and penalties are handled similiarly.  This multiplier to defense would also impact melee damage, which is why the offensive tonnage modifier was moved here.

Also, ammo defensive pricing needs adjustment.  Ammo protected by case is still a negative to a mechs defense, but currently ammo in the arms or torsos with case don't get a discount, despite damaging the mech and more significantly the pilot.  Yes, ammo in the arm of a clan mech wont destroy the whole mech, so the -BV shouldn't be as much as without case, but it should still be a negative.  Same with gauss, at only -1 BV per slot, gauss's drawback is undervalued in the defensive formula, especially with case.  Cascading explosions from AP gauss rifles on clan mechs will always kill mechs with 3 such explosions, so CASE should multiply the ammo negative BV adjustment by 1/3rd--if not more.

Finally, for offensive BV, I am for moving the 'tonnage' part from melee weapons also and unit tonnage to the defensive BV side of the equation.  This is mostly to make piloting exclusively scale off defensive BV, and gunnery exclusively scale off offensive BV.  Thus a melee mech with a hatchet wont pay extra for that hatchet with an increase in gunnery, but will pay extra for a hatchet with an increase in piloting, thanks to clearly separating the two.

Edit: putting numbers on everything for clarity.  Ammo is currently -15 defensive BV, but no minus if you have case.  With this change, since 3 ammo hits kill you case2 or not, ammo will be -5 minimum if protected by case in a location that wont kill you outright, like with an IS XL side torso.  Same with Gauss weapons, -1 per slot of gauss weapon, minimum -5.

[DevianID]

With all of the offensive and defensive bits of mech specific parts complete, we can look at other unit types.

The first is protomechs.  Protomechs dont have a tonnage additive for their melee attacks like mechs or vehicles, so they should have something to represent this. Proto's deal damage with frenzy, and also back to the attacker with DFAs.  Since protos can only damage units in the same hex, not 2 hexes like a mech kick, tonnage/2 (like a vehicle) is a good value for the combined melee potential.

Protomechs also have a modifier to their TMM, but dont actually increase their odds of being hit.  Instead, 3's and 11's miss, which is 4/36 slots.  So protomechs take only 88% of any hits, so it takes 1.125 times the hits to deal the same damage.  Thus, instead of giving protomechs +1 tmm in the defensive calculation, instead multiply by 1.125 the total of defensive BV.

The bits about ammo in protomechs is replaced by the weapon and shot curve from above.

[DevianID]

Vehicles on deck next.  With only having a focus on BV and not rules, only vtols stand out.  VTOL rotors take only 10% damage, so their effective health is 10x higher then the points paid, similar to how hardened armor also costs more.  Multiply the armor and structure points of a Vtol's rotor by 10.  So 2 points of armor in a rotor are priced as 20 points of armor.  The unit type multiplier and such are unchanged, so vtols will add additional points from rotor armor and structure, but then multiply their total defensive BV by .7.

Vehicles do not have an engine modifier, unlike mechs, as larger engines dont change the critical vulnerability.  Vehicle ICE and Fuelcell engines have fuel that explodes though, impacting defensive toughness.  Further, vehicles also have ammo critical slots that can explode, and case deletes the rear ammo.  So, if a vehicle mounts an ICE or fuel cell (or other engine with explosive fuel tank critical hit items), -15 defensive BV the same as ammo on a mech.  Further, -15 the defensive BV for a vehicle with ammo crits, or -5 for vehicles with ammo and CASE systems.  Since all ammo on a tank is stored together, this -15 for carrying ammo only applies one time.  The exception is for gauss weapons and other explosive weapons.  These can be hit in any location on a weapon hit crit, so require additional negative BV seperate from the 1 ammo crit.  Divide the explosive components into 2 categories, the first being items that can explode for equal or more internal structure of the location on a tank.  These items count as -15 defensive BV.  If the item explodes for less then the IS of the vehicles location, then it is -1 defensive BV (-1 BV per slot, to be consistent with mechs, which is 1 hittable slot on vehicles).

[DevianID]

Infantry is next.  Infantry have a lot going on.  The base rule for defensive BV is just 1.5 per trooper, modified for TMM like normal.  However, infantry have a 90% damage reduction from energy weapons, and take 3.5x damage from machineguns, but a 2x damage from taking hits outside of cover.  Pending calculating all weapons effect on infantry and finding the average damage reduction, infantry should have a unit type multiplier the same as vehicles to represent their total bonus and negative damage reductions.  An early estimate is a 1.2 multiplier, as infantry recieve more damage reduction they they take in bonus damage--unlike say a hovercraft that has a .7 multiplier as they are disabled quicker then their armor would indicate.

1.5 points for defense tracks the same as internal structure for mechs and vehicles, but infantry cant take any critical damage, and can never be destroyed early from hit locations or critical effects, so 2.5 points (like armor on mechs and vehicles) is the appropriate rate per trooper seeing as you must eliminate each and every troop to destroy the infantry unit.

For offensive BV, the antimech skill is based on weapon damage in the current formula, but it should be put in defensive and scaled off antimech skill.  If an infantry unit has anti-mech abilities, like mechs pay seperately for physical attacks, infantry should pay based on the antimech attack itself.  Instead, infantry should pay in the offensive category for the ability to shoot at range 0.  Infantry with a 1/2/3 range attack, unlike mechs, actually can attack 4 different hex ranges with a shooting attack, the hex they are in.  To calculate the range 0 shooting BV bonus for infantry, simply add the damage the infantry does at range 0 to the offensive BV.  For example, if a rifle platoon deals 11 average damage, then add 11 to the offensive BV for range 0 shooting attacks.

Also, infantry take no movement penalty for moving/jumping.  This is worth at least a +1 bonus to accuracy, same as a targeting computer or AES.  Following consistent attack bonuses, apply a 1.25 modifier to infantry shooting attacks to account for their bonus accuracy from movement.

For the antimech portion of infantry, tonnage (like for the melee on mechs) isnt appropriate as all infantry work the same regardless of size.  Instead, a 4 damage leg attack roughly cooresponds to a kick from a 20 ton mech dealing 4 damage, but with only 1 hex of influence (range 0), unlike a mech with range 0+1.  Thus, 10 BV should be added for infantry that can make anti-mech attacks (20 equavilent tonnage/2 for reduced area).  In this way, antimech attacks which always do a flat 4 damage on leg attacks wont be scaled off their shooting damage.  This will address the imbalance between too cheap infantry that dont make anti-infantry, and too expensive shooting weapons paying twice, the second time for anti-mech ability on the same guns.

Edit: A prelim glance at weapon groups and their average infantry damage versus average base damage
mguns 7, 2 base
flamers 14, 2 base
lasers 1, 5.33 base
ACs 1.25, 9.25 base
srms/streaks 2, 5.33 base
lrms 2.5, 7.5 base
LBX 2.25, 5.55 base
Ultra 3, 13 base
Pulse lasers 3, 6 base
Gauss 2, 15 base
ER lasers 1, 5.33 base
3.55 damage, 6.94 base

So a prelim is 1.95 toughness bonus.  The infantry will lose this toughness bonus when caught out of cover, as then the 2x multiplier kicks in.  If infantry are in cover 1 time in 3, that is still a net toughness bonus...  My estimate of a 1.2 modifier is pretty conservative in this case, as putting a number on how often infantry will be in cover is difficult, but 1.2 would be roughly 1 in 5, which feels more then fair when in reality infantry will probably never leave cover.

[DevianID]

Connected to infantry is battle armor.

Like infantry, battle armor share a few things.  The current formula prices antimech attacks off of shooting weapons, meaning battle armor without guns but with antimech ability dont pay.  Likewise, battlearmor with lots of arm mounted weapons pay too much.

Instead anti-mech ability should be priced at 10 BV, and put in the defensive side (as it scales off anti-mech skill, not gunnery).

Battlearmor have a range 0 attack, like infantry, which is one further hex then the same weapon on other units can shoot.  Like infantry, battlearmor should pay BV equal to their range 0 shooting ability, so a 5 strong unit of small laser battle armor should pay 9 BV for the ability to shoot at 0 hexes (3 damage, cluster on 5, so 9 average damage), compared to the same small laser found on battlemechs who lack the ability to shoot at 0 hexes.

Ammunition BV should be handled with the same formula used by mechs, meaning a weapon with 2 shots will have a decrease in BV, and weapons that dont track ammo (most other battle armor weapons) will not have an ammo modifier.  This ensures an AP gauss on a battle armor unit is the same cost as the AP gauss on mechs, other then potential ammo differences.

Battle armor ignore movement/jump penalties, so like infantry they should pay 1.25 more for their bonus accuracy when compared to a mech.

On the defensive side, the last point of health on an elemental is cheaper then the rest.  However, battle armor do not take critical damage, so this last point should be priced at 2.5 bv per point like the rest.

[wundergoat]

Awesome writeup.  I'll have to dive in deeper, but I had a few early comments.

Pulse lasers are too cheap...So a mpulse laser with 4/8/12 range with a -2, instead of costing 111, the weapon would cost 87, the value without accuracy, times 1.25 x 1.25, the same accuracy multiplier of a targeting computer.

The tcomp multiplier is kind of a cludgy way adjust any weapon bv.  In my own explorations I played with setting the base TH# to 6 and 8, then normalizing the bv of various ranged back so ISMLs (and any other x/2x/3x range weapons) would keep existing BV.  This moved the damage*accuracy calls to a part of the bell curve where accuracy mods were more impactful and more realistic.  Using 6 gave good results, putting the cLPL somewhere around 370bv.  I'll have to look, but I think base 8 put it above that of a cERPPC.  I think this approach is better since it doesn't just affect the pulse bonus, but accuracy penalties (like on MRMs and HLs) and funky ranges and min ranges (snppcs, isLRMs)

MASC and jump jets I agree with your ultimate output, but I don't think you can have MASC count for half effect by default.  The issue is uptime isn't as important on the offensive side of the BV formulas, and merely having the capability to move somewhere provides value.  Now, I do agree that the 100% uptime assumption makes no sense on the defensive side.

I don't buy that 3/5/5 is better than 4/6/4 by default on offense.  You have more movement options on more accurate movement modes on 4/6/4.  Alpha Strike JJStrong isn't relevant, as that is a defensive effect and BV draws a clear distinction between the offensive and defensive sides of the house.

For ammo bins/explosive equipment, one of the issues I have with BV is the credit given for stuff that can explode and kill you is per slot, regardless if that slot is padded or sitting alone.  I would take that explosive possibility and factor in probability.  Very crudely, you could multiply the slots by probability.  Taking the MAD-3R, DRG-1N, and CDR-3R for instance, rather than being credited for 1, 4, and 4 explosive bits, it would be 12, 20, and 26 respectively.  Obviously this needs tuning.

[DevianID]

The piloting and gunnery skill increase table is a mixed bag.  Above, care was taken to split items that scale off of piloting into defensive, and items that scale off gunnery into offensive.  The intention is to be able to price shooting weapons off the offensive BV, and piloting skill items off the defensive BV.

20% increase for gunnery has been demonstrated to be a fair value, that drops off, for the regular skill table, thanks to simulations done by others such as DFA; 20% of an entire unit's BV was always worth the increase except when stacking accuracy from pulse weapons to below 0 base to hit values.  This is due to the bell curve making accuracy past 7 to hit increasingly valuable, and accuracy below 7 decreasingly valuable, until all attacks autohit at 2 or less.

Targeting computers provide a flat +25% weapon attack modifier, so the first point of gunnery likewise should match this value, adding +25% weapon attack, to keep parity.  Unlike the standard chart though, which priced further increases as addititive (.2 + .2 = .4), a more accurate gunnery score should be multiplicative with itself.

Thus, to price accuracy consistantly each gunnery increase is a 1.25 multiplier to the offensive BV.  This means the first gunnery increase of 1.25 offensive BV is probably cheaper for almost all units then the old increase in gunnery from 4 to 3 using +20% of the entire unit.  This is intentional to ensure targeting computers and pulse lasers are not flatly better sources of accuracy.  The second gunnery increase, from 3 to 2, is another 1.25 multiplier, so now at 1.5625 total modifer going from 4 gunnery to 2, we start to see diminishing returns to stacking accuracy.  The end goal is to make 0 gunnery very expensive, a total 2.44 multipler to offensive BV, to prevent low skill units from becoming too advantageous.

For piloting skill, from a defensive point most piloting skill rolls will take place on a 6: A 5 piloting base, and +1 for taking enough damage to cause a piloting skill roll.  This means you pass an average PSR 26/36 times.  The first point of piloting increases the pass rate from 26/36, to 30/36.  So piloting, going from a 26 to a 30, is 1.15 better.  So each point of piloting should increase the defensive BV (which includes melee/antimech attacks) by 1.15.  This only applies to the defensive BV, unlike the normal rule which is a 10% increase to the entire unit.  By targeting the defensive portion, we more accurately target what piloting skill is doing.  On the flipside, a worse pilot takes the odds of passing a PSR from 26 to 21.  Now the defensive BV is reduced, so multiply the defensive BV by .81 for worse pilots.

This should only apply to mechs and aerospace fighters, the units that can fail a PSR roll from damage.  For all other units, that do not really use piloting, divide these bonuses multipiers for piloting/antimech skill in half as the skill is much less valuable to them.  Tanks, for example, almost never use piloting/driving skill, so they would only pay 1.075 for a driving increase, and .905 for a driving decrease.

Edit: Wundergoat thanks for checking it out!  So far its a little rambling, I plan on cleanup passes and examples in the future.

I agree the tcomp multiplier is a bit cludgy, but it is by the book--one of the few things I kept so far, mostly because I needed to pick SOME value to normalize accuracy against.  Pulse accuracy by the book is cheapest, skill increase accuracy is the most expensive, so for the time being the median value (used by tcomps, aes, and a few others), was chosen to be the 'price' of accuracy.  I do get what you are saying by revaluing weapon BV to 6 base instead of 4, but 6 base means short range (26/36) versus long range (6/36) is valuing short range at over 4 times the value of long range.  While everyone agrees short range is more valuable then long range, long range isnt so bad to be only worth 23% (and thus only cost 23%).  Base 4 to hits values short versus long at 45%, which feels correct.

As for 3/5/5 versus 4/6/4, the 6 run provides only a single hex that a 3/5/5 cant reach.  When you map out all the hexes and facings a 4/6/4 can have, versus a 3/5/5, the 3/5/5 has about 30 hexes more then 4/6/4, and of course 36 hexes in the 5th ring can be any facing.  So yeah, 3/5/5 has mathmatically a significant number of hexes over 4/6/4 that the 3/5/5 can threaten from.

For masc, 100% uptime on the offensive side is unreasonable.  In an earlier discussion on the topic, using a 5 turn snapshot, you can masc 'safely' 3 out of 5 turns, so I gave masc a 60% uptime based on that 5 turn snapshot.  The snapshot is a more accuracte uptime for masc then 50%, but 50% round up is the lazy solution(because it is what the book currently uses for jump jets, so we can just borrow from an existing source for consistency).  Also, the point of 8/12/6 having the same offensive threat of a 6/9/(12)/6 movement is proof to me that Masc should not count at 100% offensive uptime, because it IS worse then an 8/12/6 speed, so unlike the RAW formula, the MASC formula NEEDS to show this.

Finally, for ammo risk, I totally agree... but poor ammo placement runs into issues on valuation.  Simulation results on ammo bin contributions to mech death do show that ammo is a massive risk, more then -15 defensive BV, BUT on the other hand players can play around this.  You can shoot the ammo, or dump the ammo.  So if you have a big torso bin vulnerability, and the player evaluates its not worth carrying (mguns are pretty common for this), turn 1 you can dump the ammo.  Now, you recieved a BV discount contingent on you carrying that ammo until it detonates, but dumped it or fired it before it got hit.  For this reason, we cant use the actual risk of said ammo as its BV valuation to provide big -BV discount.  -15 is about the most discount you can give for a vulnerability you can easily offload on turn 1, and already you can come out ahead by taking mgun ammo for -15 at a cost of less then 15 BV, netting you positive BV.

[DevianID]

An example for mechs, using my observations on BV.

Dasher H (Currently 779 BV, now 1006)This unit, IMHO, was too cheap before.  This feels like a positive change.
Heavy Small Laser (currently 15)
  Heavy Small 19 base from damage/range, +6 (value of raw damage), *.8 (inverse of 1.25 accuracy multiplier for a penalty to hit) = 25*.8 = 20 BV

Offensive Calculation:
Weapons
9 Heavy Small Lasers (20* each)
Tcomp (1.25 multiplier)
Speed Factor (was 20 with always on MASC, now 18) 2.72
=612 (was 566)
Heat Pass:  22 HS+6 OV=28.  Weapons fire=27, 28>27 so no heat discount

Defensive Calculation:
Armor: 95 (no change)
Structure: 37.125 (no change)
Gyro Vulnerability: -10 (was +10)
Tonnage (melee): 20
TMM 4, +5 with MASC, so averaged speed factor 2.2+3.3 instead of always on MASC= 2.775 multiplier (was 1.5 for a 5! TMM)
 142.125 * 2.775 = 394.39 (was 213.118)

New BV for Dasher H = 1006

BV for Puma/Adder Prime (Was 2083, now 1788.2275) The puma felt too expensive before, it certainly is enjoying the reworked heat discount
ERPPC BV Was 412, now 427
Flamer BV was 6, now 8

Offensive BV
427*1.25+427/2*1.25+8/2
Heat 22+6=28, Second PPC =30, flamer = 33, so both discounted 50% as 28<30/33<36 (36 is 14 heat/shutdown for further heat discounts)
Speed Factor 1.5
804.625 * 1.5 = 1206.9375 (was 1602)
Defensive BV
  370.25 * 1.57 (TMM3) = 581.29 (was 481)

Rifleman 3N (was 1039, now 1097) The rifleman slightly abused the old heat system, so its a little more expensive with the scaled heat system
Large Laser was 123, now 131
Medium Laser was 46, now 51
AC5 was 70+ammo, now 93* before ammo, 84.6 after 10 shot multiplier applied

Offensive BV
Heat efficiency
131/8= 16
51/3= 17
93->84.6/1 = 84.6, (ammo efficiency of 10 shots per gun (1.3648/1.5 multiplier, ~91% effectiveness for only having 10 shots)

So Ac5s shoot first as the most heat efficient by far, then medium lasers, and finally large lasers

10+6=16, 1+1+3+3+8+8 =24
AC5+AC5+Mlas+Mlas+Large Laser+Large Laser (40% for hitting 14 heat over)
84.6+84.6+51+51+131+52.4 (131*.4 heat discount) = 454.2
454.2 * Speed Factor 1.12 = 508.7 BV (was 482)

Defensive BV
463.5*1.27 (TMM2)=588.645 (was 556.2)

Small Laser Elementals: (was 447, now 655) Elementals were always very hard to destroy, so while their offense went down overall thanks to fixing 2 shot weapon pricing, their defense is more accurate to how hard these are to remove on the table
Small Laser was 9 now 12
SRM2 was 21+ammo, now 29*, reduced because 2 shots (.4785/1.5 modifier)=9.25
Autorifle was 1.59, now 2.11

Offensive BV:
SLAS+SRM2+Autorifle
(reminder that weapons that dont have a cluster gain a cluster modifier of .633 in a squad of 5)
12*.633 (BA cluster)+9.25+2.11*.633 (BA cluster)=18.18
Range 0 damage = 3*.633+2.8+.52*.633= 5.03
18.18+5.03= 23.2, *1.25 (battle armor foot movement accuracy bonus) *1.25 (Battle armor jump accuracy bonus)=34.6875
Speed factor 5 = 2 (walk+1 for 'run') +1 (1 jump /2 round up for jump = walk) + 2 (each point of jump in excess of walk MP) (Was speed factor 3)
27.75*.88 (SF4)= 25.03 With reevaluated SF the offense is 34.6875 (was 30.122)

Defensive BV
2.5*11 (total health) = 27.5 (was 26)
10 BV (antimech attack possible) *1.25 (battle armor accuracy bonus) = 10 (edit: accuracy handled under weapon BV)

37.5*TMM3 modifier 1.57= 58.875 (was 33.8)
Offense+Defense *7 for squad of 5 multiplier = 93.9625*7 = 654.9, for 655

EDIT: accounting for the infantry speed factor bonus below for changing hexes for free yielding more speed then accounted for, the elemental's speed factor will go from 4 to 5, which is the same as a 2/3/3 jump unit.  Further, since 3 jump is more valuable then 3 foot MP, the jump elementals will take an additional 1.25 accuracy bonus (so 2 accuracy bonuses total) for their jump ability.  This puts their offensive modifier up to 34.6875, instead of 25.03.  This increases their BV by 67.6, for a new BV of 655.  I plan on looking at the antimech attack portion next, as originally I multiplying the antimech attack ability by battle armor accuracy, but that was incorrect as the base 10 BV formula makes use of already included a -2 to hit, so battle armor shouldn't double pay.  The basic battle armor formula also multiplies anti-mech attack by squad size, and squad size does impact accuracy to a small extent... at 10BV*7 for 5 troops, this physical attack is ending up costing as much as a 70 ton kick.  Does a 4 damage + crit attack equal 14 damage from a mech kick?  Im thinking 10 BV per trooper may have been too high, and 5 BV might be more appropriate for battle armor.  At 5 BV per trooper instead of 10, a 5 squad BA leg attack at 35 BV would be a 7 damage kick from a 35 ton mech in terms of BV equal. 

[BATTLEMASTER]

Good analysis!  How does an under-armored 'mech like the SHD-2D change?

[DevianID]

Good analysis!  How does an under-armored 'mech like the SHD-2D change?

Well the weapons would go up, like on everything, because of putting a cost on the value of damage (instead of just damage at range).

On the defense side, the TMM multiplier goes from 1.3 to 1.57, so the 369 BV goes to 446.

Offense, the BV of weapons becomes
AC5, 93 BV base, subtract the ammo shot count discount from 20 shots (.84^[shots^1.14])=.46% discount, so still 93 BV.
SRM2, 29 BV base, with 50 shots no shot discount=29BV
LRM5, 60 BV base, with 24 shots = 0 BV discount, = 60 BV
Mlas 51
No heat discount, generates 13 heat has 14 sinks.

Weapons 93+29+29+60+51+51=310, multiplier 1.63 (SF)=510.19 (was 529.75)

Final BV = 956 (from 899).

Notably, I haven't touched armor costs, so while the 2D does go up, most things go up unless they have a glaring weakness that wasnt being addressed (the heat on a Puma Prime, for example).  So most fast things will increase in price with the higher TMM modifiers, but the shadowhawk 2D will go up less then the normal shadowhawk, as the normal shadowhawk has more armor to multiply from its speed.  So while a 2D goes up 57 BV, the SHD2H goes up 103 BV (383O 783D, 1167 BV total, was 1064).  The Dasher H went up to 1006 for example, and I do feel a lot better with a Shadowhawk 2D at 956 versus a 1006 Dasher H, compared to before a 899 Shadowhawk 2D versus a 779! dasher H.  Now the SHD2D is cheaper then a Dasher H, which feels correct to me.

[DevianID]

Some further thoughts on infantry movement.  I was having trouble reconciling the speed factor on infantry versus mechs, as infantry dont run but do get free facing changes.  So the first example that came to mind was the annihilator versus 3 MP motorized movement.  In the speed formula, both count for the same thing.  However, when you plot out the hexes that an annihilator can move into, and the number of hexes the 3 MP infantry can move into, it becomes obvious that there is no contest the 'speed factor' of the infantry are much higher--as high as 3 jump MP in the absence of heavy woods, as infantry MP can move into light woods with no penalty unlike mechs.  So a 2/3/3 movement 'annihilator' unit would be speed 3+2=5, able to reach 36 hexes, the same as the infantry if there was no terrain. 

A 2/3/3 mech is speed 5

So if 3 foot MP currently is speed factor 3,

A 1 foot MP 3 jump infantry is speed factor 4 for me currently with my proposed JJ movement changes, showing the speed factor needs to be higher for the infantry, as 1/3 infantry are as good as 2/3/3 mechs in terms of hexes threatened.

Thus infantry speed factor needs to go up making 3 foot MP better then 3 mech running MP (because it is), and 1/3 infantry jump MP equal to 2/3/3 mech MP.  With the effective 'run' a 1 MP infantry would count as 2 for speed factor, as a 1 MP infantry can move into 6 total hexes while a 1/2 speed vehicle can move 1 back, and 4 forward, for 5 hexes.  Pretty close.  A 3 MP foot unit, is 36 hexes for infantry, and a 3/5 has 41 hexes it can reach.  A 3/4 speed unit has only 23 hexes it can reach, making 5 MP the more correct speed factor for 3 MP infantry, and 2 MP infantry with 18 hexes is right between speed 3/4 at 23 hexes and 2/3 at 13 hexes.

Now, to check if there is a scaling issue with more MP, what about 5 MP infantry, or 7 MP battle armor.  These currently have speed factor of 5 and 7 respectively, and like the above example have much more hex reach with 5 infantry MP then a 3/5 mech/tank would.  A 5 MP infantry unit has 90 hexes, and 7 MP has 168 hexes.  A 4/6 has 66, 5/7 has 96,  5/8 has 131 hexes, a 6/9 mech has 173, 7/11 has 275 hexes.  So we see that 7 infantry MP aligns closest with 6/9 MP, and 5 infantry MP aligns with speed 5/7 as that would be 90 (infantry) vs 96.

At the end of calculating the infantry speed equals, the numbers show that for 1 infantry MP, their 'run' for speed factor is 2, and for infantry faster then 1 MP, their 'run' is 2 higher, as it doesnt scale at 50% more, instead it is 2 higher--a 3 infantry MP = 5 mech run, 5 infantry MP = 7 mech run, 7 infantry MP = 9 mech run.

How does this work with jump MP.  Well jump is better then non jump, so a 3 jump MP squad is more useful then 3 foot MP, despite both hitting the same number of hexes outside of terrain--aka the same speed factor.  So infantry that can jump should have an additional accuracy bonus factor (infantry currently i gave a 1.25 multiplier since they dont take a movement penalty for walking).  Thus, 3 Jump infantry will be offensively 1.25 more then 3 foot infantry, which are in turn 1.25 offensively better then a 3/5 speed mech who has to pay +1 for walking or +2 for running, while sharing the same speed profile as 3 infantry in terms of hexes able to move to.

Edit: I edited the sample elemental a few posts above.  But the new question is how much BV really is an antimech attack worth.  I had it valued at 10 but in a squad of 5 that was 70 BV with the squad multiplier, which is equal to the physical attack of a 70 ton mech, but you also cant shoot when you make a battle armor anti-mech attack.  So while 14 damage +psr from a 70 ton mech and 4 damage + crit chance is rougly in the ball park, you lose shooting.  Im thinking the squad multipler hitting the antimech attack is too much... conventional infantry dont have a squad multiplier, so I dont want battle armor paying more then conventional infantry for the same attack with the same drawback.  Thoughts?

[wundergoat]

I still gotta get some presentable weapon math together for this discussion, but I ran into more BV issues. Ferro-lam armor has a 1.2 multiplier on it, which is far too low. Not only is a 20% damage reduction the bare minimum that ferro lam will get you, a 20% reduction translates to 25% more effective armor.  The multiplier should probably be around 1.4 to 1.5, if not higher.  For reference, reflective and reactive are at 1.5, effectively figuring they apply their damage reduction half the time.

The other thing was apparently LVSPLs currently not only pay nothing for their accuracy bonus, they are actually slightly cheaper than would be expected for just their basic damage.

[BATTLEMASTER]

HVACs also don't suffer a BV penalty for being randomly explodey.  I checked it a few months ago with the Heavy Metal weapon BV calculator.

[DevianID]

Yeah so far I have gone through the total warfare stuff, I havent done writeups for Tac ops.  Ultras and RACs are both off in BV, due to jam chance.  HVACs like-wise should get the 'defensive' reduction for being explodium, like ammo does.  Specialty ammo also is off, as things like precision ammo dont cost BV they only modify shot count... and ammo BV is very flawed versus my shot count multiplier.

Armor especially is off, like wundergoat mentioned.  Hardened is only a 2x, but the -2 to tacs and 18 head armor are not costed like they should be.  Specific solutions are available though, like hardened/ferrolam points in excess of 9 in the head will cost more since weapons that deal 12+ damage pay a BV multiplier for dealing that damage, meaning extra head armor needs to cost much more to balance that equation.  On the agenda is a list of all items.

For the weapon math wundergoat, what are your thoughts on the 'value of existing' damage add-on... is it enough or should it be more.  Currently I have it as the weapons maximum raw damage, because 4 medium lasers and 6 er small lasers have the same BV in the core weapon formula which is just damage x range.  However, 6 clan ER Small Lasers is better then 4 medium lasers.  1 medium laser, of course, is better then 1 clan ER small, and 1 medium laser does cost more then 1 CERSL.  But 6 ER smalls for 30 damage is stronger in a brawl then 4 mediums, so for the time being I am adding raw damage to all weapon values, as damage is undervalued in the existing formula (looking at you, 54 damage dasher H!). 

[wundergoat]

I think you’re trying to address a real issue, namely some weapons have very low BV due to poor range, but you can mitigate the issue with mobility and thus get great efficiency.  The question is how to account for it.  ML > cERSL, and with range control 4xML should beat 6xERSL, but lose if the cERSLs have range control, so my inkling is that this should be accounted for in the speed factor, but accounting for the potential on a weapon makes some sense.  Adding the base damage to BV certainly addresses short range weapon spam but it makes me wonder if we need to be looking at short range vs long range value as well.  When I looked at weapons before I used the ML as a baseline and used shift factors for S/M/L damage to hold the ML’s BV steady as I played with the base TH#, but now I’m wondering if that was correct.

One final thought for now - BV is inherently an empirical calculation and while more detailed math can get our BV model closer to true value, ultimately everything needs to be calced out and tested.  I wonder if there is some way to get megamek to run hundreds of bot games automatically and parse out good data from the logs.

[DevianID]

The speed factor does a good job for multiplying range advantage, but the issue is that a 3 damage range 21 LRM5 is 7x the cost as a small laser range 3, but 7 small lasers absolutely shred in point blank.  In an actual game with limited turn time and mapsheets, the game trends towards raw damage since you cant kite forever.  So no matter what value you give to the speed factor, its not addressing that 21 damage in small lasers is better then 3 damage at range 21, as speed factor multiplies both--see the Dasher H versus the LRM dasher.  With my '+damage' to BV, the BV for both does change.  The base BV is 65 for 3 damage at 7/14/21, which would go to 68.  The base BV of a small laser is 9, going to 12.

So pre 'value of damage' addition, its 7 smalls.  The 'value of damage existing' change makes this 68 v 12 is 5.67.  This is a step in the right direction, but im not sure its enough or not.

[DevianID]

Further analysis on the value of damage, versus the list value of armor.

This is more as a proof of concept of the 'value of damage' part of weapon BV that needs to be added.  As a reminder, I currently have it as +raw damage to the weapons BV, as I know it needs to be SOMETHING, but I havent done the calculations to see what this 'value of damage' is in the BV formula.

A medium laser is 46BV.  1 point of armor is 2.5 BV.  With 3/5 speed, an offensive speed factor of 1.0 and defensive modifier of only 1.2 in existing BV rules, at medium range, the medium laser needs 8's to hit said +2TMM unit.  If we assume 5 turns of shooting only, the medium laser will deal 10.4 damage.  This is 31.25 BV worth of armor on that TMM2 unit.  But if we play more then 5 turns of battletech, which is very likely, say 10 turns where the medium laser is shooting/12 turns total, well now that medium laser deals 62.5 points of damage worth of armor bubbles.  The medium laser has parity after shooting 7.36 times.  AKA after 7 shots with a medium laser, damage is worth more then armor.  If the medium laser shoots 7 or less times in a game, armor is worth more then the damage from a medium laser at skill 4.  As games go more then 7 turns by default, weapon damage is undervalued so long as sufficiently normal game turns are played.

With my proposed change of +5 BV to a medium laser, because it does 5 damage, and 1.27x multiplier for TMM2, now those 10.4 points of damage in 5 turns deals 33 BV in damage, and in 10 turns its 66 BV worth of damage for a 51 BV gun. This is 7.7 turns shooting the medium laser.  Its a minor change, but hopefully a step in the right direction.  However, adding only +5 BV for 5 damage doesnt change the turn count very much (less then .4 turns), so we still have the same issue.

The same calculation with a PPC at 10 damage (using the base BV system and my sample TMM change, which is minor at only TMM2) shows a PPC needs 14 shots either way, hit or miss, to earn 176 BV of armor in the base system.  So while the PPC does have 2x the range at 18 versus 9, that 2x range doesnt translate to it getting 2x the shots.  Even the most generous estimate of these 3/5 speed units will have the attacker gaining 2 hexes a turn, as the defender walks 3 back while the attacker runs 5 forward.  This means the medium laser would close on the PPC wielding 3/5 mech, like an aweseome, in 5 turns at most, and 2 turns if the PPC mech isnt moving.  This, of course, is less then the difference in weapon effectiveness with the awesome needing 14 turns to 'earn its keep' while the medium laser needs ~7.  By the time the PPC mech has earned its 14 turns to get value, the medium laser is somewhere between 17% and 56% better with +5 BV per medium laser, depending on how quickly the medium laser can close in, versus 22%-63%. 

Paying for accuracy and closing into short range, of course, is almost always worth it in these sorts of calculations involving medium range or similiar +2 hit penalties, as a +2 TMM and any movement or range mods push the to-hit to 8 or above, making accuracy at a minimum worth 40% going from 8 to 7, for only 25% more--and only getting better above 8s to hit.  However, the value of accuracy is not in question at this moment... suffice it to say, accuracy is ALWAYS worth it on an average game at least 2 times.  An elite 2 gunner is always worth the BV when shooting versus a 4 gunner.  Only at very high stacking accuracy bonuses will this not be true--pulse lasers are mostly wasted on a 0 gunner for example.

At short range, thus 6s to hit, the end result of closing in with the enemy and then standing still to deliver maximum damage, a unit hits 26/36 times versus TMM2.  This puts the turn parity in base btech BV at 8.12 turn for a PPC base BV rules and 8.11 turns for a PPC with +10 BV for 10 damage.  A medium laser needs 4.25 turns base versus 4.45 turns with my system.  Considering the short range difference for these 6s to hit is only a gap of 3, which is 1-2 turns of movement at most, we see again that even by adding +damage to the BV of weapons it isnt enough, the shorter range higher damage weapons are still better without an outside factor that makes crossing 3 hexes take multiple turns, which is not a standard map or gravity setting.

So, the value of damage added to weapons needs to be MORE then just +damage.  If a medium laser is 46, increasing it to 51 isnt enough, raw damage regardless of range is still worth more then this.  At +2x damage to BV, we see the 6 to hit medium laser at defensive BV parity in 4.88 turns, while the PPC is hitting turn parity at 8.5 turns.  At +4x damage to BV, its 5.76 versus 9.4.  So we are running into an issue that the absolute range difference is not catching up raw damage very quickly, seeing as the absolute turn difference between the 2 weapons is 2 turns at short range, but the value of the damage of the medium laser is more then this... the medium laser is closing the distance and out damaging the PPC still.  Damage is still better even at a 4x damage BV addon or 10x BV addon.

We all knew the shorter range medium laser was an apex weapon on standard maps, but the math is tough.  A theoritical medium laser only awesome would smash the PPC awesome in almost any simulation you run where the 2 maneuver against each other and they have the same BV.

As an aside, on the topic of setting accuracy to 1.25x multipliers, versus armor multipliers from TMM/defense.  When we apply the 1.25 multiplier, compared to the value of speed/defense TMM on base 4, we see the following relationship.

+1 Accuracy 1.25x  +1 TMM 1.1x
+2 Accuracy 1.5625x +2 TMM 1.27x
+3 Accuacy 1.9531x +3 TMM 1.57x
+4 Accuracy 2.4414x +4 TMM 2.2x
+5 Accuracy 3.0516x +5 TMM 3.3x
+6 Accuracy 3.8147x +6 TMM 5.5x
+7 Accuracy 4.7684x +7 TMM 11x

So I really like this relationship, as a unit with TMM+1 likely wont be getting that +1 TMM very much, so the 1.1 cost is less then the +1 Accuracy.  A 3 gunner will have to pay more to match the +1 TMM something like a 2/3 annihilator pays for, as the Annihilator likely wont be countering the +1 gunnery very much and +1 accuracy will always be useful anywhere on the 2d6 curve.  This flips around the time a unit is paying for gunnery 0 + a targeting computer or other effect, versus TMM5.  Something with a TMM5 bonus likely will be maintaining a high TMM, while the accuracy side has hit diminishing returns--something with a +5 bonus to hit will often be autohitting things with less TMM (A +5 to hit is base -1 to hit, so versus targets in short range with TMM3 or less the shots are all auto hits, thus shooting at TMM2 is wasteful), wasting points in excess accuracy, while TMM is exponentially more powerful past to hit's of 8 (4 base +4 penalty to hit).

the tl;dr for the above is that something like a PPC, from range 18 to 2, with a like enemy closing in at only 2 hexes a turn, expects about 4 hits in those 9 turns.  A weapon with half the range, and thus double the damage by BV, will get about 1.72 hits in those 9 turns, on the last 4 turns when the range 9 weapon is in range.  But there is 2x as much damage from half range, so its effectively 3.5 hits, compared to the PPC who shot for +5 turns more then the collection of half range guns.  Thus, now in close range, the next turn the medium range guns have outpaced the PPC after 10 turns of play.  Adding to the BV is troublesome, as this trend continues if you get 4 medium lasers for the price of a PPC, or 3 medium lasers for the price of a PPC; 3-1 pricing (a 96 bv medium laser versus 276 bv PPC) versus the existing 4-1 pricing (46 bv versus 176 bv) sees the mediums catchup in 4 turns instead of 1, which is still too fast as that is 13 turns of PPC shots versus 8 turns of 3x medium laser shots.  5 turns is the absolutely maximum reasonable turn difference for a unit to close from 18 to 9 hexes, while the minimum is only 2 turns.  So, the +damage addon to weapons is looking like +10x damage for BV.  Thus, a small laser needs to go from 9 bv to 39, while a medium laser needs to go from 46 to 96, to keep the ratio of damage to range at least to the minumum reasonable ratio for fairness.  I would love some thoughts on this, or other math showing exactly how many small lasers it take to equal a medium laser versus medium laser to PPC on a reasonable map.

[DevianID]

I ran some simple trials testing PPCs versus medium lasers.  The awesome with PPCs was losing non-stop, even versus only 8 medium lasers.  The lasers just get in range so fast.  The large laser version faired better (the charger challenger SB).  The worst part is, the target was just a hunchback 4P, not even anything custom.  The 11 medium laser BV equal version of an awesome dominated so much it wasnt worth running the trial a bunch, so I switched to the hunchback 4P to see if even at 8 medium lasers the raw damage was better then range.  The hunchback beat the large laser challenger more then it lost, and the large laser challenger is the same BV as an awesome, 400+ more then a hunchback 4p.

The small laser awesome similarly crushed the PPC awesome.  I could only test(fit) 26 small lasers, so it was a 1302 BV awesome variant, but once it got in range the 78 small laser damage carved up the 3 PPC awesome very quickly.  There just wasnt enough turns outside of 3 hexes for the PPCs to deal enough damage to overcome the 33 v 78 damage difference in close range.

The simulation data is just to reinforce with datapoints what I was saying about the value of raw damage being unrepresented in a weapons BV.  In a bounded field of the maps in btech, you will get in range with shorter range weapons before you are destroyed at range, when the fight is fair between the total weapon BV.  I just wanted to confirm with some simulation data that, yes, equal BV in 4 medium lasers does yield better results then the same BV in 1 PPC, and this is still true at 3 ML to 1 PPC like my numbers in the prior post posited.

Obviously, terrain such as swamp bog will favor the PPC... but terrain such as all heavy woods favors the medium laser, so while we can pick out scenarios and maps that change the core math, it still doesnt change the core issue of needing to add 'the value of damage' into a weapon's calculation, which appears in the limited data and calculations done so far to start at +10x damage in BV to all weapons.  This would make the ratio for the medium laser to small laser, currently at 1 to 5, down to 1 to 2.5, and the medium laser to PPC from 4 to 1, down to 2.9 to 1.

As an aside, the median simulation turn time was about 12 turns with the hunchback versus awesome/challenger SB.  Some 5-6 turn and 20 turn outliers, when one side got headshot or 2 damaged units poked each other with sticks and harsh language for 5 turns did appear but it was pretty consistent that the units closed in, got into short range, and someone died in 12 turns.  With pilots better then 4 gunnery, this number would go down, I just thought it was an interesting data point for about low long it takes to destroy an 80 ton energy zombie with 40 damage a turn in medium lasers.  My ammo curve put 12 shots at about 94% as effective as infinite shots, so the ammo curve was working pretty well considering how few outliers beyond 12 turns there were.

[DevianID]

Just updating this post since some community members wondered where my BV errata musings were.

From other places, and on the topic of the value of damage, in the months since the last update ive settled on the following.

Remove the 1.5 multiplier in the current weapon BV formula.  Instead, add ideal damage *3.  Add the ammo multiplier found above.  For Jam weapons, take the salvo shot count and multiply by the exponent .895.

The jam exponent means 15 shots would become 12, and 10 become 8.6, which should accurately reflect the odds of shot availability versus shots lost due to weapon jam.

This formula means a medium laser calculates as thus: 5*3*33/36+ 5*3*26/36+ 5*3*15/36... Then add 5*3 for the value of damage.   The total is thus 46, so medium lasers see no change, but small lasers will go up and large lasers/ppcs will go down.  Keeping the medium laser the same BV was a suggestion by wundergoat that made sense, while also greatly emphasisizing the value of raw damage as outlined by the posts above.

Other things mentioned by me elsewhere but not here is many weapons currently having the wrong BV.  Artillery never got updated BV when its damage tripled, VSPLs have always been wrong as they use the lowest damage instead of more damage at short range, ECT.  Also lots of errata on weapons, like removing the +1 penalty to hit with snub nosed artillery cannons didn't adjust the weapons BV, so they are undercosted.  But all that weapon stuff is 'fixed' if weapon errata would be adopted, simply as a measure of updating the BV of weapons in errata finally.  That stuff isn't a 'lets update the BV formula', it's 'the number listed is wrong using the old formula and should be changed'.

The next thing im considering is what AE damage multipliers should be, and Flak modifiers.  AE and Flak are currently free, but AE especially does bonus damage as it can hit 2 enemy units in the same hex.  It also has bonus effects versus some unit types.  Same with flak.

Off the top of my head 1.5 feels appropriate as an upcharge for AE damage, not even considering the accuracy the weapons usually get (which, like pulse would be a seperate modifier).

Flak is another question though.  While, say, an LBX loses damage via the cluster roll to gain flak and LB accuracy, other weapons like the silver bullet gauss and HAG dont have variant ammo with FLK, they get flak for free.  But flak is only useful against a very small subset of units, so the question is... How many times shooting a HAG or whatever is it getting the flak bonus?  Or, instead of increasing Flak weapons, should we reduce the defensive TMM value of units with a flak weakness?

[DevianID]

While the question of flak is out fishing for replies, another thing that has come up many times now is jump jets versus running as it relates to TMM in defensive BV calculations.  Jumping is easy TMM, you get it regardless of terrain or facing, meaning a 5\8/5 can always get a +3 tmm with jump jets, but a 5/8/0 mech must run and turn only 1 time the whole turn, limiting them to just a few hexes.  So 30 hexes of available places with jump, versus only 20 with an 8 run speed, and those 20 reduce to just 1 possible hex if terrain eats 1 MP.  Something like 2 elevation changes instead make your max run evasion only +2.

So what is the fair TMM for running then?  Subtracting .5 isn't nuanced enough, as a 5/8 should have a harder time then a 6/9 to get a +3.  We can do a step system, so you take a mechs maximum TMM if running, and grade each mech based on where they fall.  Basically a speed factor for defense.  Run speeds of 1 or 2, for example, have a 0 tmm, but run speeds 3 and 4 have a 1 maximum tmm, yet clearly a unit with 4 tmm has an easier time generating a +1 compared to a 3 max move unit.  Further, units that dont pay for turns can just zig-zag for TMM, so they can always get their max possible TMM--same as jump units.

The proposed run TMM would thus look like:
3mp .33tmm
4 .66tmm
5 1.33tmm
6 1.66tmm
7 2.25tmm
8 2.5tmm
9 2.75tmm
10 3.125tmm
11 3.25tmm
12 3.375tmm
13 3.5tmm
14 3.675tmm
15 3.75tmm
16 3.875tmm
17 4 tmm
18 4.125tmm
19 4.25tmm
20 4.375 tmm
21 4.5tmm
22 4.625 tmm
23 4.75tmm
24 4.825tmm
25 5.125tmm
26+ .125 more tmm, up to 6 max at 32+

This provides a gradient for each step, with a bump each time a new TMM plateau is hit.  So now a 10mp unit won't have a +4 tmm multiplier, it will have a 3.11 hmm multiplier, illustrating that 10mp is very unlikely to get a +4 modifier, while a 7 jump unit will always get a +4 modifier when it wants.  This should accurately reflect how much higher a TMM a jumping unit has versus a running unit.

Thoughts?  Anyone want to plop down some mechs on a random single map and see what TMM they can hit at a given speed over the course of, say, 10 turns?  I'm curious if my estimate on average TMM is close to accurate on the common map types.  Not worried about if the moves are 'good', just trying to crowdsource average 10 turn TMM across a handful of the classic maps.

[BATTLEMASTER]

What about the AMM relative to the TMM? Should that affect BV?  I'm thinking a BV discount for units where the AMM is greater than its maximum TMM for a given movement mode.  Though this would only affect anything with 4 MP run or less and 4 MP jump or less.

Or is that all considered somewhere else?

[DevianID]

What about the AMM relative to the TMM? Should that affect BV?  I'm thinking a BV discount for units where the AMM is greater than its maximum TMM for a given movement mode.  Though this would only affect anything with 4 MP run or less and 4 MP jump or less.

Or is that all considered somewhere else?

Thats a good question!  AMM and TMM are, per the existing rules, separated by offensive and defensive BV.  So defensively, while jumping makes your offense accuracy worse, on the defense side at least you dont care about how bad your accuracy is, only your total defense.  No adjustment needed here I think, a bad jump that doesnt really add much survivability also wont cost if it doesnt actually make you harder to hit.

I have not done a deep dive into figuring out the exact equation 'speed factor' is calculated, other then to identify certain movement that is not accurate to the listed speed factor (like masc counts as always on, which is inaccurate across an entire game). 

But this is where the BV discount you mention already exists!  So, for example, a unit with 5 MP running in the current rules has a speed factor BV multiplier of 1.0.  It generates and costs equal, and is the Baseline in the rulebook.  A 4 total MP has a .88, so something with only 4 MP gets a weapon discount in BV right now. 

In my proposed errata, if a unit ignores some movement TMM, it counts as an accuracy buff... thus infantry, who currently pay nothing for 0 AMM, would now pay as if they had a +1 accuracy in my errata, +1 more if they have to jump to get that speed factor.

I also increase the effective speed units that dont pay to turn are counted at.  This is done by calculating how many hexes a 'traditional' unit that pays for turns can get to at a listed speed, and how many hexes a unit with free turns can get to.  So, a unit with 1 walk MP can move into 2 hexes, but infantry with 1 MP can move into 6 hexes, making them faster by speed factor then a 1 MP non-infantry unit.  In the existing formula jump units likewise have a higher speed factor then non-jump, and can enter more total hexes then a ground based unit which justifies their higher Speed rating. 

I would not give a unit a discount for just having jump jets though, because while they are +3 AMM, you can always choose to not use them.  We dont want to give optional things a discount.  IE at only 1 jump MP, which is -2 net accuracy to defense, if we discounted that they would be cheaper then a unit with 0 jump.  The option to jump should always add value, as all other things equal having 1 jump MP is better then not having it.  But it wont cost very much in terms of speed factor.