Skill Advancement

[martinl]

Originally Posted by Varyon

I think the issue is more that putting [80] into any one skill makes you so incredibly competent (Default+25) that any encounter within your niche that the rest of the party can participate niche at all is a cakewalk for you, while anything that challenges you will wipe the floor with the rest of the party.

There are also some skills where high level skills will break the game. Someone with Strategy-30 and Tactics-30 can conquer the Roman empire with a very modest army, for example.

On the other hand lockpicking 30 doesn't really hurt most games at all - it is very situational.

When I GM I generally soft limit skills to 20 or so, unless the PC makes that skill a major focus in game. Exact limit depends somewhat on the game and character. (A DX 20 super would have a different limit than a DX10 gritty PI.)

[JulianLW]

Varyon, I think you're making a lot of the same points I'm making.

Originally Posted by Varyon

An unsupported DF Swashbuckler isn't in very good shape there, anyway. I suppose one with more points in Acrobatics has a better chance of pulling off an Acrobatic Dodge and Drop to get out of the way.

Instead of pumping more points into Broadsword, a smarter build for the Swashbuckler who wants to survive an area attack is to buy a higher level of Acrobatics, Enhanced Doge, and higher Basic Speed. Or Flying Leap or the Great Void power-up. Or have a large shield for cover. All on that DF template.

Originally Posted by Varyon

A DF Swashbuckler without a Ghost Weapon enchantment or similar is useless against those anyway. If he/she does have such, the one with Rapier-30 is in better shape than the one with Rapier-22 and slightly-higher secondary skills.

True. I was thinking of the Wizard-Hunter power-up, but that helps you fight through Shields and Force Domes. He'd need a magical friend who could deal with this problem - or an enchanted weapon.

Originally Posted by Varyon

The high-Rapier DF Swashbuckler has a better chance of both getting past the target's defenses and pulling off a strike to armor chinks/gaps. If those aren't options, and he/she lacks some sort of armor piercing enchantment, the more-balanced DF Swashbuckler is likely similarly useless (although if the opponent has high DR but not great grappling skills, the more-balanced character might have slightly higher Wrestling to restrain the target).

There's also just more ST - and power blow - to beat the DR. And not all DR has chinks.

Originally Posted by Varyon

A balanced DF Swashbuckler might have slightly higher Dodge, which would be an advantage here, but will largely be reliant on support from the rest of the party to get past such an encounter.

Yep: Higher Dodge. Which means a higher DX and/or higher HT or Enhanced Speed or Dodge. Or, for DF-style missile attacks, Shield skill or Parry Missile Weapons. And again, those teammates.

Originally Posted by Varyon

I don't think the DF Swashbuckler is terribly perceptive to start with, and is again reliant on the rest of the party to avoid getting backstabbed.

Buy Danger Sense or higher Per. And yet again, he'll need a friend....

Originally Posted by Varyon

If you're playing a game where "sword-guy" isn't a legitimate role in the party, you're SoL regardless. If it is a legitimate role, pumping all your points into a single skill can make you ridiculously competent in said role.

The point Gnome was originally making, I think, was that the Swashbuckler buys up Broadsword to ridiculously high levels and wipes up the floor with all the challenges so that the other templates are left twiddling their thumbs. And that's not true at all.

The Swashbuckler is a specialist, sure, but if he buys up Broadsword to ridiculously high levels, he's probably leaving himself vulnerable to some other really basic attack. I don't think, as Gnome was arguing, that "pumping all your points into one skill" (as you put it) lets him "graduate" beyond the monsters he's likely to encounter - unless your GM only throws a certain kind of melee monster at you.

And no level of Broadsword makes a Swashbuckler so competent that he doesn't need help, as you pointed out again here.

Originally Posted by Varyon

I think the issue is more that putting [80] into any one skill makes you so incredibly competent (Default+25) that any encounter within your niche that the rest of the party can participate at all is a cakewalk for you, while anything that challenges you will wipe the floor with the rest of the party. Sure, it's perfectly fine - desirable, even - that the Swashbuckler be better at melee combat than the Scout, but it's not so great if the rest of the party is functionally useless in melee combat so long as the Swashbuckler is around.

Personally, I wouldn't build such a character, and wouldn't allow one to be designed, without any sort of mechanical disincentives (just the classic GM veto). For those who would rather have mechanical disincentives, I'm certainly not going to tell them doing so is Hurting Wrong Fun.

The character you're describing - who is so hands-down better at his niche that the other players are useless? That guy is built for ONE SINGLE SCENARIO: opponents it's easy to get right up close to and hack with a sword - i.e. opponents made of flesh with low natural DR who are not using ranged weapons. He does great at that one scenario.

But the idea that buying up one skill really high to the exclusion of all the other options on your template is going to make your character supreme is, I think, a flawed strategy for most DF games. The guy who ignores buying more ST and Striking ST, a higher Per, a higher Speed and better Dodge, Danger Sense, HT, etc.... That guy will get killed quicker, in most games, I suspect, than the one with a more balanced point-spending strategy - who wants to survive a dose of poison, be able to dodge or block an attack he can't parry, be able to knock down a larger or a better-armored foe (or one without chinks), be able to sense danger, be able to run away faster, etc....

In most games, I strongly suspect, the guy who dumps all his CP into Broadsword is going to get hosed sooner than the guy who spreads his points around more evenly.

[kenclary]

Originally Posted by JulianLW

But the idea that buying up one skill really high to the exclusion of all the other options on your template is going to make your character supreme is, I think, a flawed strategy for most DF games.

DF games are (historically, across the hobby) where ultra-specialized min/max builds are most likely to succeed --- so the fact that, with GURPS, even in DF, they have huge weaknesses is a testament to GURPS' strengths.

Let alone all those other styles and genres where that style of build is even more silly...

[Gnome]

Originally Posted by JulianLW

The point Gnome was originally making, I think, was that the Swashbuckler buys up Broadsword to ridiculously high levels and wipes up the floor with all the challenges so that the other templates are left twiddling their thumbs. And that's not true at all.

Only, as you pointed out, when it's one challenge type (which happens to be by far the most common type in DF): things you can hit with a sword. The other characters still aren't twiddling their thumbs if they can do other things, like heal, etc. But any points they put in melee skills are mostly wasted, as well as any abilities that interact in a numerical way with melee skills (I'm thinking of buff spells like Shield or Grace, or debuffs like Clumsiness or whatever).

The Swashbuckler is a specialist, sure, but if he buys up Broadsword to ridiculously high levels, he's probably leaving himself vulnerable to some other really basic attack. I don't think, as Gnome was arguing, that "pumping all your points into one skill" (as you put it) lets him "graduate" beyond the monsters he's likely to encounter - unless your GM only throws a certain kind of melee monster at you.

As I said earlier, I'm speaking from my experience of playing DF games that include every kind of threat you can possible imagine, with multiple GMs, often using published monsters but many other scenarios as well. We usually have a typical balanced party that includes other archetypes. The melee fighters are specifically the ones whose one uber-high skill gets out of hand before the other types of abilities get out of hand. Sure, you want to buy Enhanced Dodge, Acrobatics, all of those other things. But in terms of point for point bang for your buck they tend to pale in comparison.

But I think this problem is easily solved by a number of methods mentioned in this thread (and I've tried more than one!): cap skills, require Unusual Background, or my current favorite which is to just use the 3e cost progression.

The character you're describing - who is so hands-down better at his niche that the other players are useless? That guy is built for ONE SINGLE SCENARIO: opponents it's easy to get right up close to and hack with a sword - i.e. opponents made of flesh with low natural DR who are not using ranged weapons. He does great at that one scenario.

But the idea that buying up one skill really high to the exclusion of all the other options on your template is going to make your character supreme is, I think, a flawed strategy for most DF games. The guy who ignores buying more ST and Striking ST, a higher Per, a higher Speed and better Dodge, Danger Sense, HT, etc.... That guy will get killed quicker, in most games, I suspect, than the one with a more balanced point-spending strategy - who wants to survive a dose of poison, be able to dodge or block an attack he can't parry, be able to knock down a larger or a better-armored foe (or one without chinks), be able to sense danger, be able to run away faster, etc....

In most games, I strongly suspect, the guy who dumps all his CP into Broadsword is going to get hosed sooner than the guy who spreads his points around more evenly.

Buying up skill is so cheap, you don't really have to ignore those other things. You could just lower DX by a level or two (for example) and spend another 20-40 more points in weapon skill.
What you're saying sounds like an article of faith about GURPS: "I'm sure if you spend more on this you'll have less left for that." Of course I can't argue with the arithmetic of that, I'm just saying some of those things get you less overall utility.

[Tymathee]

In regards to the "Sword Guy Problem", I think a decent solution is to simply disallow the use of selecting Hit Locations when attacking. Big skill soaks penalties to hit easily, making exploiting the Hit Location rules all too simple. I might recommend only allowing the use of Hit Locations when it would be dramatically appropriate; "Aim for the dragon's glowing heart!".

Of course I suspect a lot of GURPS GMs would not want to sacrifice realism for the sake of game balance. I take no issue with it, but others might. It's just one way to handle the issue of out of control melee skill levels.

One problem is it really poo-poos on Techniques (if they're even permitted in the first place) that involve targeting Hit Locations. A compromise might be that you have to buy into Techniques with points in order to use Hit Locations. Works as a pseudo-Unusual Background tax.

[Emerikol]

So I think the game can be broken if you have players intent on breaking it. This is why I asked the question. I've dealt with some expert rules lawyers over the years so I understand how they think.

So there are a variety of approaches to "fix" the problem.
1. Just get player buyin not to do this... For me this works but is unsatisfying.

2. Make the cost increase for advancement of the same effectiveness. This is absolutely realistic in my opinion. There is a diminishing returns for improving a skill with training. Then the question becomes what is the best approach.

3. Another approach is enforced breadth. You do this by requiring something like a pyramid. So you always must have more skills at cp 1 than you do at cp 2 and so forth all the way up. I'd probably just use a term like rank to represent movement up the skill table. So an IQ-3 very hard skill or an IQ easy skill are both rank 1 when you spend that first point. The downsides of this approach are that PCs will start taking skills just to grow their pyramid and they'd become jack of all trades in some pretty weird stuff.


So my sympathies likely lay with #2 above but I'm still not sure the exact right way to do this. I've mentioned that everything over 20 costs it level-20 in addition to the base +4. You could also just keep doubling the cost so it's +4 then +8 then +16 then +32 and at some point PCs will stop improving that skill.

[coronatiger]

In one of our campaigns, I would classify Mattea and Ilzo as combat monsters by this time. We just had a big fight, and on most of her attacks, Mattea had Bow-33 including Acc (28 on Move and Attack), before taking penalties for range and darkness. (She could take -3 in addition, to be able to shoot every second.)

Ilzo held off the main charge single handedly for most of the fight, while Mattea returned fire on the opponent's archers, as well as keeping climbers from reaching our ridge top.

Even though Leopold and Va'lyndra have put lots of points in non-combat related skills and abilities, they were crucial to our success. They helped keep the ridge top clear, and Lyn even picked off a couple of archers.

Mattea depends on her mobility. If Lyn and Leo hadn't been there, she would likely have joined Ilzo for back-to-back fighting, losing the advantage of holding the high ground, as well as being stuck in melee range.

So, my point is: Even though Ilzo and Mattea at times got kills every round, and none of us expected Leopold and Va'lyndra to be nearly as efficient, our two less combaty friends had important tasks during the fight.

And to generalize, having a high-skill fighter ensures that the party has someone to occupy the opponents while the rest do something important, such as unlocking the gate so everyone can escape, solving the riddle to deactivate the constructs trying to kill them, or just watch the fighter's back so he doesn't get swarmed.

[kenclary]

Originally Posted by coronatiger

So, my point is: Even though Ilzo and Mattea at times got kills every round, and none of us expected Leopold and Va'lyndra to be nearly as efficient, our two less combaty friends had important tasks during the fight.

And to generalize, having a high-skill fighter ensures that the party has someone to occupy the opponents while the rest do something important, such as unlocking the gate so everyone can escape, solving the riddle to deactivate the constructs trying to kill them, or just watch the fighter's back so he doesn't get swarmed.

"Sword Guy" can't be in two places at once. Being the best at a particular task doesn't give you the ability to always be the one doing it. Nor does it guaranteee that success will always be about attacking or parrying with a sword.

It's not broken; in fact it's suboptimal. ("It" being Sword Guy.)

[bocephus]

Originally Posted by Emerikol

So I think the game can be broken if you have players intent on breaking it. This is why I asked the question. I've dealt with some expert rules lawyers over the years so I understand how they think.

So there are a variety of approaches to "fix" the problem.
1. Just get player buyin not to do this... For me this works but is unsatisfying.

2. Make the cost increase for advancement of the same effectiveness. This is absolutely realistic in my opinion. There is a diminishing returns for improving a skill with training. Then the question becomes what is the best approach.

3. Another approach is enforced breadth. You do this by requiring something like a pyramid. So you always must have more skills at cp 1 than you do at cp 2 and so forth all the way up. I'd probably just use a term like rank to represent movement up the skill table. So an IQ-3 very hard skill or an IQ easy skill are both rank 1 when you spend that first point. The downsides of this approach are that PCs will start taking skills just to grow their pyramid and they'd become jack of all trades in some pretty weird stuff.


So my sympathies likely lay with #2 above but I'm still not sure the exact right way to do this. I've mentioned that everything over 20 costs it level-20 in addition to the base +4. You could also just keep doubling the cost so it's +4 then +8 then +16 then +32 and at some point PCs will stop improving that skill.

Another, possibly simpler, option is to limit Crits to a max of 6 or less. If you follow the "exceed your skill by 10" rule a 40 skill would be a crit on most swings.

No matter how good your skill anything higher than a 6 is just a hit. This makes a 40 sword skill much much less useful. The math just doesn't pay back the points. Once you get to skill 20, to absorb a reasonable number of penalties, more is something you will grow into figuring out the sweet spot of whats legitimately enough to do the job.

There is a slight caveat to Range weapons, especially muscle powered ones because distances add significant penalties very quickly. But this is a function of how the mechanics work.

Yet another option is that as your skill proggession stays the same but the time to train increases. Instead of the generally accepted 200 hours to raise a skill (or something like it) once you get to Stat+4 or what ever you choose, start adding 100 hours each skill level. So its not the cost that is the investment, but the game time required. This will probably not work in games where there isn't some semblance of time tracking, or downtime that reflects healing, shopping, gathering supplies, travel.... but even if you are a generous GM and allow 4 hours of training a day, thats 25 days in game. Once it goes up it takes a month and a half, then two months, 2.5months, 3 months, 3.5... your character in vestment is the same in points but time determines the ability to raise the skill not just having cash in your wallet.

Lastly an option I came up with for a house rule. sort of what you said but not doubling. This was discussed at great length in another thread you can find by looking for posts that have this text.

Skill level cost variation (this replaces the standard version, its not an addition to it):
Scale them by one instead of exponential. Just adding one to each subsequent skill increase, you pay additional 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, and on (the points spent intersect at the 6th increase then start to become cost prohibitive) early on its even a touch cheaper, but definitely covers increasing skill once you are at that high skill level. If it wasn't that I use a tool that is pre configured for GURPS I think I would probably implement this right off now that someone pointed it out.

VH progression GURPS 4e 1, 2, 4, 8, 12, 16, 20, 24, 28, 32, 36
VH progression proposed 1, 2, 4, 7, 11, 16, 22, 29, 37, 46, 56
VH progression math: 1, 1+1, 2+2, 4+3, 7+4, 11+5, 16+6, 22+7, 29+8, 37+9, 46+10, 56+11, etc....

Its basically the same point investment up to:
Attribute +7 for easy
Attribute +6 for average
Attribute +5 for hard
Attribute +4 for VH

After this point it starts getting really expensive and a +40 would simply be untenable. Your game has to be balanced for this, and you can allow the addition of techniques for martial abilities to get players up into that 20ish sweet spot. But this is mechanically much easier to control if your not using one of the software build options, they usually hard code skill level progression

[Emerikol]

Originally Posted by bocephus

Skill level cost variation (this replaces the standard version, its not an addition to it):
Scale them by one instead of exponential. Just adding one to each subsequent skill increase, you pay additional 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, and on (the points spent intersect at the 6th increase then start to become cost prohibitive) early on its even a touch cheaper, but definitely covers increasing skill once you are at that high skill level. If it wasn't that I use a tool that is pre configured for GURPS I think I would probably implement this right off now that someone pointed it out.

VH progression GURPS 4e 1, 2, 4, 8, 12, 16, 20, 24, 28, 32, 36
VH progression proposed 1, 2, 4, 7, 11, 16, 22, 29, 37, 46, 56
VH progression math: 1, 1+1, 2+2, 4+3, 7+4, 11+5, 16+6, 22+7, 29+8, 37+9, 46+10, 56+11, etc....

Its basically the same point investment up to:
Attribute +7 for easy
Attribute +6 for average
Attribute +5 for hard
Attribute +4 for VH

After this point it starts getting really expensive and a +40 would simply be untenable. Your game has to be balanced for this, and you can allow the addition of techniques for martial abilities to get players up into that 20ish sweet spot. But this is mechanically much easier to control if your not using one of the software build options, they usually hard code skill level progression

The above looks really good.