Skill Ratings & Point Value - Feedback Wanted

[nckestrel]

You're assuming both 1) equal number of units and 2) an even distribution of PV among units.  Both of which are very big assumptions.
If I had a force of 3 Urbanmechs and 1 Dire Wolf, where I put my Skill 3 is pretty significant. The PV of my force is not equally distributed.
If my opponent has 20 Stingers, and I have 3 Dire Wolves, each of us having a single Skill 3 unit is a pretty big advantage for me. My PV is equally distributed among my force, but one unit is far more significant among 3 than it is among 20.

[sadlerbw]

I’m with nckestrel on this one. Where you put a Low-skill pilot makes a big difference. There is no basis to say giving two equal PV forces one skill 3 pilot means they both have the same PV invested in pilots. PV cost for skill upgrades changes based on the unit the low-skill pilot is piloting, and getting rid of that idea and moving to a flat cost for skill increases regardless of the unit will almost certainly make balance worse.

Uniform is not necessarily balanced. I believe the goal here is to improve the gameplay balance of pilot skill changes, and going to a flat skill would not help that goal. I admit it would be simple, but it wouldn’t useful for balancing two forces based on PV.

[Tai Dai Cultist]

As a bit of update on my project to map LOS penalties based on every hex to every other hex on the standard battletech map:

Ranging was quick and easy.  Because for example, the range from hex X to hex Y is of course the same as hex Y to X.  Furthermore all ranges from hexes in row A to hexes in row B are the same as ranges from row A+1 to B+1 and etc.   If anyone's interested in seeing the band patterns for all 73,984 possible hex combinations I can post a spreadsheet just waiting for terrain penalties waiting to be filled in.  But I don't think a naked sheet does anyone any good without those penalties added in.

The insane part is drawing LOS across all ~74,000 combinations.  Unlike range, terrain penalties are not necessarily reciprocal.  No copypasta to speed up the process :(

A few days in to the process, I'm thinking I'm endorsing Joel47's of baking in terrain being in the way for a flat % of the shots.  If PV crunching can assume players with faster units can dictate range, I don't see any problem with assuming players will make use of terrain rather than taking pains to remain free of it for the purposes of evaluating the impact of Skill modification on PV.

[Elmoth]

He is actually right about both forces having the same % of the force invested in pilots (both having a skill 3) regardless of composition. If one choice is great and the other is very bad does not impact the truth in that sentence.

[DarkJaguar]

You're assuming both 1) equal number of units and 2) an even distribution of PV among units.  Both of which are very big assumptions.
If I had a force of 3 Urbanmechs and 1 Dire Wolf, where I put my Skill 3 is pretty significant. The PV of my force is not equally distributed.
If my opponent has 20 Stingers, and I have 3 Dire Wolves, each of us having a single Skill 3 unit is a pretty big advantage for me. My PV is equally distributed among my force, but one unit is far more significant among 3 than it is among 20.

I'm assuming we want an easy to implement system that is relatively difficult to cheese.

Basing any sort of skill buff on percentage of unit cost will ALWAYS favor lower PV units.  The game's rules as is favors swarm tactics (Even post rebalance).  This is just a drawback of what is effectively a dice pool game.  Whoever can roll more dice is statistically going to do better. The PV cost for pilots allows you to tune in where your high skill pilots are used.  A percentage based cost will tune it more in favor of low PV units.  A fixed cost will tune it in favor of high PV units (assuming the cost is reasonably high).

To the point Sadler made.

Uniform is not necessarily balanced. I believe the goal here is to improve the gameplay balance of pilot skill changes, and going to a flat skill would not help that goal. I admit it would be simple, but it wouldn’t useful for balancing two forces based on PV.

"balanced" in the median is not necessarily balanced at all.  People should stop looking for "balance" in a game that is impossible to balance in, and instead look for fairness.  Two equal PV forces will not be necessarily balanced right now even if everything is skill 4.  It comes down to force composition.  The rules as they are written allow for stuff to happen that is so heinous that the people who write the rules won't even address it, and instead simply recommend "walk away from that game".

[DarkJaguar]

Okay, so I went and did some quick math.  Based COMPLETELY ARBITRARILY on Descronan's results quoted below, I have come up with the following values.

Skill 3 Pilot: +5 PV
Skill 2 Pilot: +15 PV
Skill 1 Pilot: +30 PV
Skill 0 Pilot: +50 PV

I came to these values by taking the MEDIAN size 4 battlemech at 49 PV (The average Size 4 Battlemech is 49.08 for reference), and then multiplying that number by the effectiveness difference in the referenced table.  At that point I had 4.41, 13.23, 28.91, and 53.41 as absolute percentages of the median size 4 mech.  From there I simply rounded to the nearest logical multiple of 5.

Again, these values were arbitrarily calculated.  I don't particularly care what they end up being, as long as they're flat across the board (PV agnostic), and a multiple of 5 or 10 for simplicity in application.  Also, you may notice that there are no provisions for skill 5+ pilots.  This is intentional.  If a scenario wants to use a skill 5 pilot, fine.  Otherwise, refunding points for lower skilled pilots just gives a way to min-max force composition and does not contribute to the "fun" of the game meaningfully in any other way.

Advantages of using these numbers
-Hard to game the system
-Balanced around the hardest hitting units in the game
-Sooooo easy to math
-Equitable
-seriously, please try and break this.

Disadvantages
-Imprecise
-Cheap units cost the same to level up as expensive units
-doesn't require a nifty table in the book that people have to reference

Wow! 8 Pages of comments. I haven't been able to read through all of them so my comments may be redundant.

First, skill mostly affects offensive value, not defensive. Yes, you can evade, but that rarely gives you a significant benefit. So this issue really comes down to an executive decision if the Offensive Value should be printed on the cards. If the answer is yes, then we can accurately attack the issue (pun intended). If not, then we are forced to use the PV of the entire unit.

Regardless, the majority are pointing out the odds of hitting reduce vs. the lowest likely target on the board. However, I think you've got that backwards. We should be looking at the odds of missing in comparison. Instead of starting with skill 4, start with skill 0 and work backwards.

Also, what is the average lifespan of a mech in combat? How many shots do you expect to take before your unit is combat ineffective? My bet is 2-3 rounds.

So my ideal cost table would look something like this:

Skill   PV Percent
0   209
1   159
2   127
3   109
4   100
5   92
6   77
7   58


This assumes 3 attacks based on the percentages below:
Skill   Combined
0   27.76
1   16.65
2   8.32
3   2.77
4   0
5   -2.77
6   -8.32
7   -16.65

[DarkJaguar]

.

[sadlerbw]

.

I think you certainly made your point there!  ^-^

I don’t mind trying to find instances I feel are broken using your suggestion, however I’m having a hard time figuring out what exactly you would consider broken. You ok if I just look for 1v1 matchups where Equal PV but different skills would likely produce lopsided fights? Would you rather I make up two sides with the same PV, and then give them both additional PV in skill increases in a way I think is unbalanced? Given some of the stuff you have mentioned in the last page or two, I’m honestly not sure what would, in your mind, count as evidence of a ‘broken’ matchup.

I’d rather not throw some examples together and have you come back and say, “well that doesn’t count because of X or Y.” So, help me understand what would convince you a proposed skill PV cost system was broken, and I’ll see if I can break it.

I mean, for me this alone would be compelling enough:
Ballistic jump platoon: 7PV base, 1/0/0, 6”j, 2a/1s. At skill 0 would be 57PV.
Mad Cat Mk. II 4: 51PV base, 4/4/4 ov2, 8”j, 9a/5s. It could skill up to 3 and still only be 56PV. To me that says the skill costs aren’t right. Maybe it doesn’t say that to you.

Oddly, I’m finding myself not entirely opposed to the idea of not giving refunds for raising skill above 4. The only reason I do that with skills is A) for thematic reasons, or B) to save PV on a unit that doesn’t really need good skill. It would still be nice to know roughly how aweful a super low-skill unit will be when planning forces, but I admit that raising skills over 4 is much less important in games I’ve played than lowering it below 4.

[Descronan]

Another idea/point. When calculating the TMM, the team introduced a new formula to address the cost increase. Seems to me, the same method should be used for skill as both affect the overall bell curve. Unfortunately, I haven't looked at the new PV formula enough to suggest anything more than that. I'll dig into it and see if I can make a more coherent statement.

Regardless, this all illustrates that an automated PV calculator is needed. The formulas for calculating PV are getting more complicated to the point where human error is almost guaranteed, which makes customization difficult to nearly impossible.

If you want simple, it will not be accurate. If you want accurate, it will not be simple.

[sadlerbw]

Out of curiosity, I decided to fire up the fight-o-matic and have the Skill 4 Mad Cat Mk. II 4 fight the Skill 0 Jump Infantry (Ballistic). This was using the 1v1, 10k repititions mode. The Mad Cat won every fight...all of them. No ties, no wins for the infantry. Mad cat 10,000 to nothing. So, for me that is enough to say a flat 50PV for Skill 0 is too much.

But why stop there!? I decided to give the infantry a more fair match, so I pulled up the stats for the Charger 1A1. It also has no damage outside of short range, and it moves 10" but doesn't jump. Also, it's only 18 PV. So, I left the infantry at Skill 0 and the Charger at Skill 4. The infantry managed to pull out a few wins and a couple of ties...and I do mean few. One round was 9987 Charger wins, 7 Infantry wins, and 6 ties. The other round was 9997 Charger wins, 1 Infantry wins, and 2 ties. So, in this case dropping the Skill on the infantry to 0 isn't even close to worth 11 PV!

What else can I try against these super-elite crunchies? How about something less beefy? Spider 10K it is! 17PV, 1/1/0 and only 3 health. It's fast, but damage and health are about as low as they can get and at skill 4, the infantry might have a chance. Well, they did have more of a chance, but still got stomped. Both fights ended up at around 6900 wins for the spider, 1300 wins for the infantry, and 1800 ties. So, even without the health advantage, the infantry at skill 0 aren't worth even 10PV.

So, lets go lower! Galleon GAL-100: 13PV at Skill 4, 2/1/0, 12" move, 4 health. Nope, still not enough. Galleon wins about 9700 times with about 200 ties and only 100 wins for the infantry. At this level, it's looking like lowering the skill of these infantry all the way to 0 isn't even worth 6PV! Fine, we go even lower.

How about we have them fight some other slightly-less-cheap infantry! How about the "AA Mechanized infantry Mechanized AA Infantry Point, 67th BattleMech Cluster, Iota Galaxy": 10PV, 1/1/0, 3 health, and 6"w move. Basically, they have one damage at medium and one extra health pip than the Skill 0 jump infantry. Is that enough to make Skill 0 a winner? Nope. 6500 wins for the AA Mech.Inf., 1900 ties, and about 1500 wins for the Skill 0 infantry. Well, now we have a slight problem, as making these infantry Skill 0 is worth less than 1 PV per decrease! These infantry suck so bad that there is no room for discrete steps between each skill decrease, even when fighting equally crappy units!

Thing is, this holds true for the vast majority of units that are under about 15 PV. Paying even 1PV per skill decrease is probably too expensive! Because of this alone, I'm willing to say, "no flat rate for skills. It must be somehow related to the PV or stats of the unit." I doubt I've changed your mind, since I don't think you see a reason why you would ever put a Skill 0 pilot in a 7PV unit, but hopefully you can see where I am coming from.

[Elmoth]

Wow. Infantry really are bad  in AS

[Valkerie]

Out of curiosity, I decided to fire up the fight-o-matic and have the Skill 4 Mad Cat Mk. II 4 fight the Skill 0 Jump Infantry (Ballistic). This was using the 1v1, 10k repititions mode. The Mad Cat won every fight...all of them. No ties, no wins for the infantry. Mad cat 10,000 to nothing. So, for me that is enough to say a flat 50PV for Skill 0 is too much.

I have to agree with this.  I would never pay that much for a skill 0 unit.

While there should be increased PV costs for going skill 3 or lower, they shouldn't be so expensive that a player would never consider it.  To me, the PV increase for skill should make a player think hard on what units to bring to the battle.  Enough to weigh the costs and benefits, but not so much that there doesn't appear to be any pay off.  "Do I spend the points to bring my Warhammer down to skill 1 and keep the Jenner, or only go to skill 3 with the Warhammer and bring a Dervish instead?  What will better serve me this outing?"

[Joel47]

...fire up the fight-o-matic...

 :toofunny:

[ianpelgrim]

I think that pv related to unit pv is the best for skill rating.
Since your changing things. Why not make pilot specials worth a number of skill ratings.
So you will see more pilot specials in lighter units. And army value reflects pilot specials.

[DarkJaguar]

Out of curiosity, I decided to fire up the fight-o-matic and have the Skill 4 Mad Cat Mk. II 4 fight the Skill 0 Jump Infantry (Ballistic). This was using the 1v1, 10k repititions mode. The Mad Cat won every fight...all of them. No ties, no wins for the infantry. Mad cat 10,000 to nothing. So, for me that is enough to say a flat 50PV for Skill 0 is too much.

This is actually functioning how I intended it.  The purpose of a flat rate skill boost isn't to make it so that units that should never be able to be a serious threat to another one suddenly are.  it's to make the pilot skill detached from the unit's cost so that it's consistent.

The flat rates I quoted are tied in to the median cost of an assault mech, so quite purposefully, anything that costs LESS than that average cost will have diminishing returns from a pilot upgrade.  My design goals and features intended are as follows.

• Simple to implement
• 2nd Grade Math
• consistent for all play styles
• difficult to successfully exploit

Thing is, this holds true for the vast majority of units that are under about 15 PV. Paying even 1PV per skill decrease is probably too expensive! Because of this alone, I'm willing to say, "no flat rate for skills. It must be somehow related to the PV or stats of the unit." I doubt I've changed your mind, since I don't think you see a reason why you would ever put a Skill 0 pilot in a 7PV unit, but hopefully you can see where I am coming from.

You are correct in that you have not changed my mind.  Thank you for your time in doing these experiments, but if anything, you have further reinforced my faith in the system, that you were unable to break it in these scenarios.  While I'm certain that in the end few people will come around, remember that the flat PV wasn't designed to be balanced between units, it was designed to be fair between players.

Pilot skill does not factor into a unit's defensive capability outside of specific advanced rules and tactical uses that your simulator cannot account for.  The other end of extremes would be something like a Skill 4 pilot in http://masterunitlist.info/Unit/Details/5832/turkina-z and a skill 0 pilot in http://masterunitlist.info/Unit/Details/708/cougar-e .  That little cougar, in a flat out simulation should NEVER win using basic rules, even though the PV is the same, using flat rates.  This isn't a flaw, the Turkina simply outclasses the cougar, and the player who put 50 points into the cougar made a poor tactical decision.

That's not to say I'm opposed to suggestions, but keep in mind that trying to balance these numbers is A) impossible, and B) MUST account for extremes, and not median cases.  So what is a Mad Dog E with Sandblaster, Long-Range Master, and Sniper worth?  Because Skill 0 lets the player use up to three SPA's totaling 6 SPA points.  That Mad Dog pilot will need to roll -1+TMM at long range, or 2+TMM at extreme for 4 and 3 points of damage respectively.  Meanwhile any normal opposing pilot of skill 4 would need to roll a 10 at long or 12 at extreme.

[SC_Dave]

After lurking on this thread for a while a few points come to mind:

1. PV has been tweaked to try to balance things better. But there are some match ups, no matter how cheap you made skill upgrades, where certain units just cannot compete with others. (For example, you could give a 2" moving, rifle armed, foot infantry platoon a skill upgrade to 0 for free and in a one-on-one match up against a skill 4 Trebuchet it should still rightly lose every time. The infantry move too slowly to control the range & can't shoot as far as a Trebuchet.)

2. A system where it is simple to calculate the cost of skill upgrades is desired.

3. Some suggest linking the cost of skill upgrades to offensive values. This does seem to make sense as piloting skill rolls do not exist in Alpha Strike & PV already takes into account defensive factors such as armor & speed. It does however conflict with point 2, as various special abilities, overheat etc influence a units offensive ability. It would seem to me that such a system would require something akin to tailoring skill upgrade costs to each unit. But as noted in point 1, sometimes in a 1-v-1 situation you will never make it fair.

After considering the above for a while I have come up with an alternative idea. I won't spell it out in detail here as I have previously been advised that doing so could potentially stop it being used (unsolicited ideas etc).

I will however suggest that people consider the way Special Pilot Abilities have been treated. Some of these affect to-hit numbers. It is however recognised that SPAs are simply too difficult to cost in PV. Instead SPAs have their own system where they cost between 1 - 4 points or are granted through the building of formations.   

[sadlerbw]

Getting back to my "foot infantry are terrible" demonstration earlier: Yep. Some units are so terrible that making them pay anything is over-costed. My contention was, because the game does have a reasonable number of very low PV units, we really need a sliding scale based on PV, or offensive power, or whatever so that you don't get screwed quite so bad if you end up with low-skill 'pilots' stuck in bad units. Flat PV cost for a skill level is just too punishing for anything that isn't a Clan Assault Mech. While not as bad as foot infantry, there have been several Clan Kahns who were elementals. Imagine trying to play a game with a Skill 0 point of Elementals with Kahn Licoln Osis in it and having to pay 40-50PV to put him on the board. You can kiss PV balance goodbye. Want to bring some 7th Kommando with your Wolf's Dragoons? Not if there is a flat PV cost for elite infantry you don't! And what about that Ghost Bear Zeta Galaxy? That is a Galaxy made entirely of Elementals, and two clusters are Elite, one is Veteran, and one is Regular. I guess you should only play the regular cluster if you want any hope of PV balance?

I guess my point is that this game include famous, elite units that aren't assault mechs. If you try to include these units with a flat PV, I don't see how it would work. Trying to include units in a thematic game that are low-skill AND low-PV isn't going to work with flat PV. The skill cost needs to be tied, at least vaguely, to the potential power of a unit or it will screw up PV balancing. Even if you give both sides the same number and level of upgraded pilots, it only works if both forces have about the same number of really powerful units. If I'm playing the Davion Light Guards, and you are playing the 1st Sword of Light, I'm going to get stomped if we both get the same number of pilot skill upgrades, and those upgrades cost the same amount. If the biggest mech I have to upgrade is a light, and you have access to Heavies and Assaults, it just isn't going to work.

But hey, agree to disagree and all that. If the powers that be ended up going with your suggestion, I wouldn't feel like someone peed in my cheerios. I'd just bring more assault mechs!

[DarkJaguar]

Getting back to my "foot infantry are terrible" demonstration earlier: Yep. Some units are so terrible that making them pay anything is over-costed. My contention was, because the game does have a reasonable number of very low PV units, we really need a sliding scale based on PV, or offensive power, or whatever so that you don't get screwed quite so bad if you end up with low-skill 'pilots' stuck in bad units. Flat PV cost for a skill level is just too punishing for anything that isn't a Clan Assault Mech. While not as bad as foot infantry, there have been several Clan Kahns who were elementals. Imagine trying to play a game with a Skill 0 point of Elementals with Kahn Licoln Osis in it and having to pay 40-50PV to put him on the board. You can kiss PV balance goodbye. Want to bring some 7th Kommando with your Wolf's Dragoons? Not if there is a flat PV cost for elite infantry you don't! And what about that Ghost Bear Zeta Galaxy? That is a Galaxy made entirely of Elementals, and two clusters are Elite, one is Veteran, and one is Regular. I guess you should only play the regular cluster if you want any hope of PV balance?

I guess my point is that this game include famous, elite units that aren't assault mechs. If you try to include these units with a flat PV, I don't see how it would work. Trying to include units in a thematic game that are low-skill AND low-PV isn't going to work with flat PV. The skill cost needs to be tied, at least vaguely, to the potential power of a unit or it will screw up PV balancing. Even if you give both sides the same number and level of upgraded pilots, it only works if both forces have about the same number of really powerful units. If I'm playing the Davion Light Guards, and you are playing the 1st Sword of Light, I'm going to get stomped if we both get the same number of pilot skill upgrades, and those upgrades cost the same amount. If the biggest mech I have to upgrade is a light, and you have access to Heavies and Assaults, it just isn't going to work.

But hey, agree to disagree and all that. If the powers that be ended up going with your suggestion, I wouldn't feel like someone peed in my cheerios. I'd just bring more assault mechs!

You bring up an interesting thought here.  What role does setting have in balance?  Should it even be factored in?  There aren't any rules in the book prescribing force composition based on faction.  There's not even anything that dictates play style other than a brief mention of zellbrigen.  Even that is very loose in its application.  That being said, does the unwritten code of faction play have a place in this discussion?

I agree that the lore and setting is very important to the game, but when those setting and lore excerpts have no rules written for them, I think they really ought to be completely discounted from the discussion.  Yes, it sucks that a player who wanted to make a "canon" Clan Invasion Era Smoke Jaguar unit would be screwed on PV, that would be the decision they made.  If their group wanted to house rule that they were going to ignore PV cost for skill up in favor of making a purely compositionally canon fight, they could approach it that way.  Nothing in the rules however requires players to do such though.

That's not to say that there aren't drawbacks and shortcomings of the system, just that I think the benefits outweigh them more than any other proposed system thus far.  There are too many problems with the game's core rules and bloat to even attempt to balance it as is, so it's best to ignore balance in favor of consistency.  Any system based on a formula will be exploitable, and will emphasize certain units.

For what it's worth, I use a TON of battle armor in play.  i am very familiar with them, and I would never spend 50 points to make one a skill 0 pilot.  There would never be a situation in which it would be worth it, but that being said, I also don't think there is really any situation in which I would spend ANY PV on upgrading a battle armor though, excepting having 1-3 points left over that I can literally not spend anywhere else. 

This is not a fault necessarily of the unit, and more that there is no significant benefit to upgrading the pilot on them outside of a negligible reduction in to hit rolls (at the ranges in which BA operate).  The SPAs in the book simply do not benefit battle armor in any meaningful way, and thus upgrading a unit that can only (generally) operate in the short and medium bracket where to-hit rolls range from 4-8 to hit is really not worth any cost.

As it stands the only units tangibly worth upgrading pilots in are Mechs and Aerospace fighters, and unless the pilot abilities evolve to give real benefits to other units as well, they should likewise be discounted from this discussion and simply focus on BM and AS unit types.

[Joel47]

A flat cost, while trivial to implement, is an oversimplification. It would be worse, balance-wise, than the current system. If the argument for ignoring the imbalance is, essentially, "don't bring those units," then the rule is dictating how people should play (assuming they want a fair battle).

While a simple system is a goal, it is not (I believe) the goal. We want to find a system that better balances the extremes, without being something I would enjoy requiring complex math. I suspect it will be a small-to-moderate alteration of the existing table.

[DarkJaguar]

A flat cost, while trivial to implement, is an oversimplification. It would be worse, balance-wise, than the current system. If the argument for ignoring the imbalance is, essentially, "don't bring those units," then the rule is dictating how people should play (assuming they want a fair battle).

While a simple system is a goal, it is not (I believe) the goal. We want to find a system that better balances the extremes, without being something I would enjoy requiring complex math. I suspect it will be a small-to-moderate alteration of the existing table.

I just don't think there is any way to make it balanced across everything.  No matter WHAT you do, something will be screwed or disproportionately favored, so IMHO it makes most sense to "balance" it at the units that will be able to make the most benefit out of it.  Sure, under a flat cost system Firemoths and elementals wont be able to justify the cost of a skill increase. However  when the alternative is a system that can be fairly easily exploited by edge case units, requires referencing a table to figure out, and really doesn't do anything to "balance" the game, what is the point?

[nckestrel]

Enough arguing.  You are not convincing each other, and you don’t have to. Make your arguments for what you want and stop worrying about what other people want. We get your arguments, move on or be done.

[sadlerbw]

Ok, back to numbers then! I messed around with some spot-checks rather than doing the grand 'fight everything' mode. Partially because I didn't want too many potential outliers to get smoothed out, and partially because I didn't have time to sit here and let it run! Anyway, I was picking some semi-random matchups to see what sort of things happened with my pet "skill increases cost 12% of base PV per increase" idea. Again, it reinforced that ANY sufficiently simple formula is not going to produce tight results. Even just moving one step to Skill 3 vs. Skill 4 produced a pretty hefty swing. I won't go through ALL the numbers here, but here are some things that were notable from all the staring at numbers:

- 12% was generally a bit low for Skill 3 and Skill 2, but the outliers could still put up 2:1 win ratios, or loose just as badly.

- Skill 0 was feast or famine. The PV I was looking at (35 in this case) ended up making most of the Skill 0 units lights, or very low damage or durability mediums. Many of the fights ended up coming down to whether the Skill 4 unit had enough damage output to drop the Skill 0 unit with one lucky shot. If it didn't, then the Skill 0 would tend to run away with the wins.

- Down at Skill 0, 12% of PV actually worked out reasonably well with the units I was spot-checking. Now, there were some severe outliers where they either stomped or were stomped comprehensively, but the blowouts weren't all going one way. Crazy losses were happening as much as crazy wins. There were even several very even matchups (Dark Crow Standard [skill 4] vs. Sentinel 3L [skill 0] or Bellerephon 1X [Skill 0] were super-close.)

- The speedy little buggers like the Fire Moth G should fear anything with Skill 2 or less that has enough damage to one-shot it...because that seems to happen a lot. Actually, I could generalize that to: fear anything with enough damage to one-shot you and enough Skill to offset your TMM down to an effective 2 or less.

- Pure damage numbers weren't terribly consistent as a predictor of comparative power between units. The Fire Moth G (4/4/0), Wolverine 6K (3/3/0), Avalanche 10 (2/3/2), and Dark Crow (Standard) (4/4/3) are all 35PV at Skill 4. Judging by Offensive value, if we drop them all to Skill 0 the win order should be Dark Crow -> Fire Moth -> Avalanche -> Wolverine. It isn't. What actually happens at Skill 0 is this: Dark Crow -> Wolverine -> Avalanche -> Fire Moth. The Fire Moth gets utterly destroyed by everything else, the Wolverine has about a 3:1 win ratio over the Avalanche, and while the Dark Crow does have about a 2.5:1 win advantage against the Wolverine, there are also around 3000 ties between the two. The Wolverines Armor and Structure are doing a whole lot for its value in a fight, and the damage numbers just can't capture that.

So in general, 12% looked like it wasn't quite enough cost for Skill 3 and 2, but cumulatively wasn't too terrible at Skills 1 and 0. I need to do some more spot checking at a higher PV so I can include some of the monster assaults though, especially to see if Skills 1 and 0 drop off more when everything at that skill level isn't forced to be relatively light. If I had to pick a single percentage for each skill increase I'd have to go for more than 12% at this point. I may try 15% next.

[Joel47]

Interesting. I wonder if the powers-that-be would accept the slight added complexity of a column like:

PV	Skill 3	Skill 2 or better
33-36	5	4

(That's just one row. I used 15% for the skill 3 cost and 12% for the Skill 2.)

[sadlerbw]

Ran some 54PV monsters through the 12% test. I used these units: Devastator 2, Mad Cat Mk. II Std., Annihilator 1G, and Timber Wolf Prime. I thought about going over 54 PV, but outliers gonna outlie, so I figured I should stay away from the tippy-top of the PV list.

Anyway, I compared them to a variety of units that would also cost 54PV if skill upgrades were 12% of base PV per level. I tried to pick one fast unit, one heavily armored unit, one unit with loads of firepower, and one relatively balanced unit for Skills 3,2 and 0. I skipped Skill 1 because it's results were close to what happened at 0 last time, and I was short on time. Anyway, here are my observations with these new units:

- When the low-skill units aren't all forced to be speedy low-armor units, or units with virtually no damage, it does make a difference. The results were different in meaningful ways this time around.

- Skill 3 units actually ended up being too expensive this time around! The Timber Wolf Prime was the most balanced against Skill 3 units, but the others simply had too much armor/structure and too much firepower for the 1-point skill upgrade to translate into wins. Massive speed did help, but again these 54PV guys have so much firepower that the 1-or-2-shot effect still came into play. The Annihilator was a pure force of nature, even at Skill 4, unless you put it up against something that could drop it in two or three shots.

- Skill 2 was similar to Skill 3. Only difference is that when the Skill 4's won, they tended to win even bigger.

- Skill 0 was less all-or-nothing than it was at the lower PV level. The Annihilator still wrecked everything convincingly, but the Timber Wolf was fairly even, and the Devastator and Mad Cat Mk. II both had at least one matchup that was close. However, Skill 0 still ended up being over-priced at 12% of base PV.

So, in this case, 12% was actually too expensive for anything with sufficient amounts of armor and structure! I don't know what to say about that. At the low end, 12% wasn't enough. At the high end, it was too much. I was hoping not to bring algebra into this, but I'm not sure that is going to work! Oh, and just for reference, these were the opponents I chose at the different skill levels:

Skill 3: Highlander 732, Arcas 2, Atlas RS, Black Lanner H, Warhammer IIC 11

Skill 2: Battlemaster 4S, King Crab 0000, Fenris Prime, Pariah A

Skill 0: Banshee 6S, Cestus 7A, Dragonfly F, Gunsmith.

[Joel47]

What are you doing for range & cover algorithms?

[sadlerbw]

Pretty much stock from the version I'm using: range is random, woods percent 10/30/50, cover percent 10/30/50

I could change woods and cover but it doesn't make much of a difference. Fights just generally end faster. I'm also OK with random range. Random makes a nice way to test general combat power.

[iamfanboy]

So, uh.

This guy.

AV-OU Avatar.

He's basically a walking C3M unit, frail, slow, and weak - unless it happens to be underwater, which is a super corner case.

Why is it 45 points after the the recent PV change? It was 29 PV before.

I was planning on going through my old C3 article and updating it so I was clicking on the links, only to be confronted by this... bit... of unusual point costing.

[nckestrel]

It’s assumed you are using it underwater.  If you are not, you choose poorly.  Fortunately, that one is an Omni. Fix it :).
I believe the increase is due to the spreadsheet fixing that TOR, unlike most special ability damage, is added to base damage and therefore needs to cost PV.  It would be 30 without the TOR.