Improving contests via new rules

[seismic73]

Originally Posted by garyb

It seems everything in GURPS where 2 or more characters oppose each other there is a contest of skills. Why not for combat?

Originally Posted by The Speaker in Dreams

Now THIS is a darn cool idea! I like it very much - I think I'll have to implement it when I get a chance to see how it plays out. I've always wanted to try a margin of success - damage system in GURPS, but never could quite figure how to manage it. This ought to do quite nicely.

Originally Posted by S41NT

I just wanted to say today I got back to this issue and made a very very simple matlab program (LOL) to calculate the probability os an attack hitting (the attacker being successfull and the defender not) considering just the skill difference and the toHit chosen by the attacker (supposing the rest of his skill goes to a deceptive attack.

Hello,
Not sure if anyone will see this message, given that the last posting was more than a decade ago, but this thread was the most relevant one for the topic I want to discuss. I’m new to GURPS, but I’ve been spending months and months digging into the rules and figuring out the mechanics. I love the "realism" but I find the rules slow to play and a bit daunting to learn. I decided to try and see if I could make some simplifications without losing what is good about the game. I like the idea suggested above of using Quick Contests for all combat, and making full use of margins of victory.

Following S41NT above, I wrote a matlab computer program that calculates the probability of attacker success (as a number between 0 and 1) for different GURPS rules (such as, e.g., attack rolls in combat, and Quick Contests for magic) and then plots the results. The plots are in a matrix format which allows one to see what the probability of attack success is for different combinations of attack/defender effective skill. The plots can be found HERE . (I would have inserted the plots in this thread, but I don't see a way to do that. The matlab script I used to make these plots is in the same directory. Don't ask me why there are multiple copies... some Google weirdness...)

My plots:
1. Supernatural_rule16.pdf: A supernatural attack, which is a skill roll (including critical success/failure, along with the Rule of 16) followed by a Quick Contest.
2. Combat_v_dodge.pdf: A combat attack against a dodge defense.
3. Quick_contest_w_criticals.pdf: A Quick Contest that includes critical hits and misses for the attacker. (Same as #1, but no Rule of 16.)
4. Quick_contest_no_criticals.pdf: A simple Quick Contest with no critical hits/misses.
5. Diff_between_quick_contests.pdf: This is a difference plot of #4 minus #3.

#1, 3, and 4 use effective skill levels ranging from 3 to 25 for both attacker and defender, since these are realistic for GURPS characters. #2 uses a range of 3 to 25 for the attacker, but only 2 to 12 for the defender, since the latter range is typical of dodge scores (speed + 3).

For plot #1 we see the weird effects of the Rule of 16 (I really don’t like it). Having a skill level above 16 doesn’t do much for you when the defender’s skill is below that value, and is only slightly useful when the defender’s score is above 16. I think this ruins the realism of the game. If someone has a super-duper skill in something, and goes against average Joe Shmoe, they should be almost guaranteed success. That’s how it is in real-life combat, so why not in fantasy magic?

From plot #2 we see a really interesting result: improving one's dodge skill is pretty worthless, since a higher score only marginally decreases one's probability of being hit. Also, the probabilities seem wrong. My impression for real combat is that it is always easier to defend than to attack (contrary to what other's have posted). In professional boxing matches the statistics show that success in landing punches is rarely better than 40%. On the contrary, in this plot we see that as long as the attacker's score is relatively high, he has a probability of success well above 60%, regardless of the defender’s effective skill.

From plot #3 we see a plot that I like better. It is a basic Quick Contest (i.e., attacker and defender roll against their respective skills and calculate margins of success/failure, and then compare these) with critical successes/failures for the attacker factored in. This produces the diagonal constant-probabilities. What this says is that no matter what your effective skill level is, if you are matched with someone with the same effective skill you’ll tend to only be successful with your attack around 45% of the time. This is more in line with my analogy to boxing, above. It also seems to be more in line with common sense, where a few skill levels should make a BIG difference as to who pounds who into the dust (all other things being equal). There are a couple of other features of this plot that need to be explained. First, the probability of success never goes fully to 100% nor fully to 0% due to the possibility of critical successes/failures on the dice rolls. Second, there is a weird triangle of higher probability in the blue in the upper right side of the plot. This is due to the dynamics of GURPS rules for when an attacker with skill level between 15-18 attacks someone in 18+ range. The attacker still takes full advantage of higher probabilities of critical hits when he rolls 3, 4, 5, or 6. (Plot #5 brings out these weird features more clearly.)

Plot #4 is my favorite. It shows a straight-up Quick Contest, no critical hits or misses. The probabilities can go up to 100% or down to 0% in this case. There is no weird triangle of probability in the upper right side. The reason I like this is because I think it’s more realistic. If there is a HUGE difference in skill levels, then the attack should succeed automatically (in my opinion). If an 800 Lb. tiger tries to maul an old man in a wheelchair, it’s not going to miss. Short of divine intervention, the dude is a goner. Tigers don’t miss unless they are fighting something that can move more or less as fast as they can.

So, given the above results, here’s my proposal for different game mechanics: get rid of the usual attack/defend rules for combat. They differ from the Quick Contest paradigm and don’t seem to give realistic results. Replace all combat (hand-to-hand, distance, and magic) with Quick Contests. Also change the rules for critical hits/ misses. Make a critical hit occur when the margin of victory is 10 or greater, and a critical miss occurs when the margin of victory is -10 or less. The attacker would still use the same skill roll to attack as they do under current rules. For a dodge, the defender would use DX (perhaps with a modified). For a parry, the defender would roll against the skill level of their parrying weapon (again, maybe with a modifier). A similar rule would apply for blocking defense. Magic would also be done by Quick Contests with similar margin-of-victory rules to determine critical successes/failures.

I think these changes would make the rules simpler, more realistic, and lead to faster play. There would be no complicated Rule of 16. Critical hits/misses would be easier to calculate. And it would be a unified set of rules for all combat, whether physical or magical. It would, however, probably also produce a more deadly game, since skill levels differing by only a few points would have drastically different probabilities of success. But… c’est la vie… when dangerous creatures attack each other with full force and deadly intent, small skill or strength differences tend to decide the outcome, and it happens fast. This would make survival depend more on having a good strategy, as well as having a good GM who makes sure the bad guys are an appropriate match for the players.

I’m curious what experienced GURPS players think of my plots of the probabilities and my proposed rule changes. I haven't play-tested this, so maybe it's all a bad idea.

Thanks for reading through this (if you got this far)!!

[Sindri]

Originally Posted by seismic73

Hello,
Not sure if anyone will see this message

Everyone will see it, because when you post in an old thread it brings it up to the first page. You'd have been better served just making a new one.

Originally Posted by seismic73

#1, 3, and 4 use effective skill levels ranging from 3 to 25 for both attacker and defender, since these are realistic for GURPS characters. #2 uses a range of 3 to 25 for the attacker, but only 2 to 12 for the defender, since the latter range is typical of dodge scores (speed + 3).

This doesn't make any sense. First of all skill 25 is at best just barely realistic. For firearms its a level of skill where you realistically start being limited by some guns inherent spread without even bothering to use the sights, let alone aim.

Second these attacking skill and defense scores are neither comparable, nor do they demonstrate GURPS combat as she is played.

The counter to someone with a broadsword skill of 25 is... someone with a broadsword skill of 25 giving parry 16 (3 base +1 combat reflexes [since having combat reflexes is the default position among PCs and any worthy foe at skill 25.] +12 from skill) and either skill 24 or 26 would be more efficient. Someone with that doesn't bother to dodge except against attacks they can't parry.

In terms of actual dodge, my close quarters combat specialist in a game I'm playing in has a standard dodge of 16 (3 base +1 combat reflexes +7 Basic Speed +2 Acrobatic Dodge +3 retreat)

You want it to go at least up to 16, and since you're using matlab you might as well go past that too.

Originally Posted by seismic73

For plot #1 we see the weird effects of the Rule of 16 (I really don’t like it). Having a skill level above 16 doesn’t do much for you when the defender’s skill is below that value, and is only slightly useful when the defender’s score is above 16. I think this ruins the realism of the game. If someone has a super-duper skill in something, and goes against average Joe Shmoe, they should be almost guaranteed success. That’s how it is in real-life combat, so why not in fantasy magic?

Yes the rule of 16 is stupid. Its needlessly complicated, not an accurate genre simulation and is yet another thing that encourages rampant generalism with Magic.

Originally Posted by seismic73

From plot #2 we see a really interesting result: improving one's dodge skill is pretty worthless, since a higher score only marginally decreases one's probability of being hit.

I assure you that that isn't true. Dodge is definitely the expensive but high quality option, that's why anyone bothers to get good at parry, but its very useful. Especially as the tech level goes up and getting hit becomes a really great way to get dead.

Sidenote: having a heatmap instead of raw probabilities makes it a pain to interpret.

Originally Posted by seismic73

Also, the probabilities seem wrong. My impression for real combat is that it is always easier to defend than to attack (contrary to what other's have posted). In professional boxing matches the statistics show that success in landing punches is rarely better than 40%. On the contrary, in this plot we see that as long as the attacker's score is relatively high, he has a probability of success well above 60%, regardless of the defender’s effective skill.21

We can play the personal impressions game all day long. I'll throw in my two cents though. People start out not really being able to defend at all absent All Out Defense and slowly develop the basic ability to take Attack maneuvers instead of switching between All Out Attack and All Out Defense.

Boxing is a pain to interact with mechanically. Its inherently hard to both handle the ridiculous length of boxing matches and the speed of actual fights in one ruleset. That said, boxers parry.

Originally Posted by seismic73

From plot #3 we see a plot that I like better. It is a basic Quick Contest (i.e., attacker and defender roll against their respective skills and calculate margins of success/failure, and then compare these) with critical successes/failures for the attacker factored in. This produces the diagonal constant-probabilities. What this says is that no matter what your effective skill level is, if you are matched with someone with the same effective skill you’ll tend to only be successful with your attack around 45% of the time. This is more in line with my analogy to boxing, above. It also seems to be more in line with common sense, where a few skill levels should make a BIG difference as to who pounds who into the dust (all other things being equal). There are a couple of other features of this plot that need to be explained. First, the probability of success never goes fully to 100% nor fully to 0% due to the possibility of critical successes/failures on the dice rolls. Second, there is a weird triangle of higher probability in the blue in the upper right side of the plot. This is due to the dynamics of GURPS rules for when an attacker with skill level between 15-18 attacks someone in 18+ range. The attacker still takes full advantage of higher probabilities of critical hits when he rolls 3, 4, 5, or 6. (Plot #5 brings out these weird features more clearly.)

We already have diagonal constant-probabilities. +2 skill on both sides gives the defender +1 parry and the attacker enough skill for -2 more penalties in deceptive attack, removing the parry bonus.

Originally Posted by seismic73

So, given the above results, here’s my proposal for different game mechanics: get rid of the usual attack/defend rules for combat. They differ from the Quick Contest paradigm and don’t seem to give realistic results. Replace all combat (hand-to-hand, distance, and magic) with Quick Contests. Also change the rules for critical hits/ misses. Make a critical hit occur when the margin of victory is 10 or greater, and a critical miss occurs when the margin of victory is -10 or less. The attacker would still use the same skill roll to attack as they do under current rules. For a dodge, the defender would use DX (perhaps with a modified). For a parry, the defender would roll against the skill level of their parrying weapon (again, maybe with a modifier). A similar rule would apply for blocking defense. Magic would also be done by Quick Contests with similar margin-of-victory rules to determine critical successes/failures.

I think these changes would make the rules simpler, more realistic, and lead to faster play. There would be no complicated Rule of 16. Critical hits/misses would be easier to calculate. And it would be a unified set of rules for all combat, whether physical or magical. It would, however, probably also produce a more deadly game, since skill levels differing by only a few points would have drastically different probabilities of success. But… c’est la vie… when dangerous creatures attack each other with full force and deadly intent, small skill or strength differences tend to decide the outcome, and it happens fast. This would make survival depend more on having a good strategy, as well as having a good GM who makes sure the bad guys are an appropriate match for the players.

The attack/defense paradigm does differ from quick contests. That's a good thing, because quick contests are not a good rule. Not that they're a bad rule, they're just fine. They're a mechanical fallback position, a generic tool that can be used for all sorts of skills. That sort of thing is a necessary object because you can't custom build a system for everything people can do. Anything you actually really care about though they're unsuited for because they simply do not have enough axes of differentiation. Being able to interact with attack, defense, damage and injury separately is what makes GURPS combat actually, you know, interesting to play through instead of people just hitting each other with their skill until they win and mostly only taking optional penalties if it multiplies an already significant skill gap between them and their enemy which is what direct skill vs skill with benefits for large margins tends to boil down to.

[D10]

Originally Posted by seismic73

I’m new to GURPS, but I’ve been spending months and months digging into the rules and figuring out the mechanics. I love the "realism" but I find the rules slow to play and a bit daunting to learn.

Welcome, and dont worry, thats pretty standard, as you play more and learn more rules and options the game will get smoother and easier.

Originally Posted by seismic73

I decided to try and see if I could make some simplifications without losing what is good about the game. I like the idea suggested above of using Quick Contests for all combat, and making full use of margins of victory.

I recommend against it.

Originally Posted by seismic73

From plot #2 we see a really interesting result: improving one's dodge skill is pretty worthless, since a higher score only marginally decreases one's probability of being hit. Also, the probabilities seem wrong. My impression for real combat is that it is always easier to defend than to attack (contrary to what other's have posted). In professional boxing matches the statistics show that success in landing punches is rarely better than 40%. On the contrary, in this plot we see that as long as the attacker's score is relatively high, he has a probability of success well above 60%, regardless of the defender’s effective skill.

I advise against trying to extrapolate boxing statistics to GURPS combat, or MMA for that matter, as deaths would be really common if we simply started applying gurps rules to these combats. Yes, in gurps its 'easier' to defend than it is to attack, but this can change very quickly. Someone who gets hit with a critical strike can lose half his dodge score very quickly. Being attacked from an angle can severely cripple your defensive capabilities. Having used your retreat against someone else that turn will result in a diminished defense capacity for the remainder of the turn, and being attacked by several attackers will be bad for almost everyone.

Originally Posted by seismic73

This produces the diagonal constant-probabilities. What this says is that no matter what your effective skill level is, if you are matched with someone with the same effective skill you’ll tend to only be successful with your attack around 45% of the time. This is more in line with my analogy to boxing, above.

Aside from empowering your boxing analogy, I fail to see how this system would be in any way desirable, or an improvement, over the current one.

Originally Posted by seismic73

It also seems to be more in line with common sense, where a few skill levels should make a BIG difference as to who pounds who into the dust (all other things being equal).

Increasing a few skill levels already makes a big difference, what is the problem here?

Originally Posted by seismic73

Plot #4 is my favorite. It shows a straight-up Quick Contest, no critical hits or misses. The probabilities can go up to 100% or down to 0% in this case. There is no weird triangle of probability in the upper right side. The reason I like this is because I think it’s more realistic. If there is a HUGE difference in skill levels, then the attack should succeed automatically (in my opinion). If an 800 Lb. tiger tries to maul an old man in a wheelchair, it’s not going to miss. Short of divine intervention, the dude is a goner. Tigers don’t miss unless they are fighting something that can move more or less as fast as they can.

I disagree with your basic premisse that if a tiger attacks an old man in a wheelchair 'its not going to miss'. Maybe over the course of a combat you would say that it is almost impossible for the tiger not to maul the old man, but to say that the only way the tiger can miss ONE hit is by divine intervention seems a bit much. What happened to critical failures? Even an assassin standing by a combatose person can be stupid and drop his knife.

Now I completely understand trying to change the game to suit your subjective feelings of 'feels right', but I would advise against changing such a solid, well balanced system merely on that basis. For one, you will be on your own with regards to rules feedback, and the combat system might be the best conflict resolution system in all of the GURPS tool box.

Originally Posted by seismic73

So, given the above results, here’s my proposal for different game mechanics: get rid of the usual attack/defend rules for combat. They differ from the Quick Contest paradigm and don’t seem to give realistic results.

Realistic results is a subjective concept. For me GURPS combat has more than enough and I certainly dont use all the options at my disposal, such as the ones for lasting damage to limbs despite them not being crippled, and etc.

GURPS aspires towards realism, but it will never truly give us realism. Sure you can change the rules so that at the end of the day you feel like it is 'more realist', but never in more than 7 years of playing GURPS 4e I have felt that the results being yielded by the system at my disposal were something id bother changing because 'they were not realistic'.

Originally Posted by seismic73

Replace all combat (hand-to-hand, distance, and magic) with Quick Contests.

This instantly breaks the game and makes magic users and ranged attackers godlike beings capable of vandalizing lesser beings with their 30ish skill level habilities. With this implemented, you will either have to arbitrarily cap all skill levels in your world or accept that a wizard with lets say, mass daze, at skill level 30 can basically knock out rooms full of the most healthy and resilient people in the world with barely any chance of resistance.

Originally Posted by seismic73

Also change the rules for critical hits/ misses. Make a critical hit occur when the margin of victory is 10 or greater, and a critical miss occurs when the margin of victory is -10 or less.

The incentive to raise a single skill to a ridiculously high level is already very high for combat/magic. By doing this, you increase this incentive by orders of magnitude and risk entering an endless chain or arbitrary fixes to prevent the initial change from breaking the game.

Originally Posted by seismic73

The attacker would still use the same skill roll to attack as they do under current rules. For a dodge, the defender would use DX (perhaps with a modified). For a parry, the defender would roll against the skill level of their parrying weapon (again, maybe with a modifier). A similar rule would apply for blocking defense. Magic would also be done by Quick Contests with similar margin-of-victory rules to determine critical successes/failures.

I think these changes would make the rules simpler, more realistic, and lead to faster play. There would be no complicated Rule of 16.

Frankly, its not a complicated rule. Theres nothing complex about it other than having to be aware of it.

Originally Posted by seismic73

Critical hits/misses would be easier to calculate.

No they wouldnt, because now instead of knowing if I scored a critical hit before my enemy defends I need to wait for his roll, we both gotta calculate our margin of success and then calculate the delta between the margins. It seems a lot less easy to calculate to me.

Originally Posted by seismic73

And it would be a unified set of rules for all combat, whether physical or magical. It would, however, probably also produce a more deadly game, since skill levels differing by only a few points would have drastically different probabilities of success. But… c’est la vie… when dangerous creatures attack each other with full force and deadly intent, small skill or strength differences tend to decide the outcome, and it happens fast. This would make survival depend more on having a good strategy, as well as having a good GM who makes sure the bad guys are an appropriate match for the players.

To me, it just seems like this set of rules disempowers players from trying to face mighty foes and creates an arms race for ridiculously high levels in combat skills. With your system, instead of every attack/spell being an interesting contest it becomes a cakewalk for the one-skill joes.

Originally Posted by seismic73

I’m curious what experienced GURPS players think of my plots of the probabilities and my proposed rule changes. I haven't play-tested this, so maybe it's all a bad idea.

Thanks for reading through this (if you got this far)!!

I have played over 400 sessions of GURPS in a DFesque setup. I have seen the game being strained to its limits, with 1100 points PCs, skills at the range of 25-30, etc. The game is exceptional, and I think if you give the current system a chance, you will find that the outcomes it produces are extremelly well suited in terms of fun and realism. Sure there will be a quirky situation here and there where you personally felt something was not right, but after seeing it in action and basically stress testing the system to its limits, im very satisfied with it and I think it produces more realistic, and more fun results, than the proposed alternatives.

In fact, I get scared whenever this topic comes up because I love the current combat system so much. My point, if its not obvious by now, is that this is entirely a matter of opinion, and you might risk breaking the game in way I can't even antecipate by pursuing your desire for a better game.

My personal recommendation is to give the combat system a chance and play it out a bit. When I first started playing I found a bit bothersome that sometimes combats would take almost an entire game session to finish, or even more than one. But after getting used to it, I just feel like its part of the game, and a good one at that.

[seismic73]

Originally Posted by Sindri

In terms of actual dodge, my close quarters combat specialist in a game I'm playing in has a standard dodge of 16 (3 base +1 combat reflexes +7 Basic Speed +2 Acrobatic Dodge +3 retreat)

You want it to go at least up to 16, and since you're using matlab you might as well go past that too.

I didn't realize Dodge scores go up that high. If I ever re-post the plots, I'll make sure the x-axis range go much higher. Thanks.

Originally Posted by Sindri

Sidenote: having a heatmap instead of raw probabilities makes it a pain to interpret.

I tried contour plots, surface plots, etc. I thought a table of values would be less easy to interpret, but I can post those in the future if someone wants to see them.

Originally Posted by Sindri

Boxing is a pain to interact with mechanically. Its inherently hard to both handle the ridiculous length of boxing matches and the speed of actual fights in one ruleset. That said, boxers parry.

I believe you on the "pain to interact with mechanically". I disagree, however, on the "boxers parry". For whatever it's worth: I box and have competed in the past. We dodge and block (absorb the punches on our gloves and arms, which are kept close to the body), but we don't parry.

Sindri, thanks for taking the time to read my post and for your advice. I'll stick to the standard rules, as you suggest.

Originally Posted by D10

I disagree with your basic premisse that if a tiger attacks an old man in a wheelchair 'its not going to miss'. Maybe over the course of a combat you would say that it is almost impossible for the tiger not to maul the old man, but to say that the only way the tiger can miss ONE hit is by divine intervention seems a bit much. What happened to critical failures? Even an assassin standing by a combatose person can be stupid and drop his knife.

The rules state somewhere that simple tasks shouldn't require a roll. Depending on how one defines a "simple task", a tiger killing an invalid, or an assassin killing a comatose patient might be a "simple task". That's how I'd rather play it, but I get your point.

Originally Posted by D10

Now I completely understand trying to change the game to suit your subjective feelings of 'feels right', but I would advise against changing such a solid, well balanced system merely on that basis. For one, you will be on your own with regards to rules feedback, and the combat system might be the best conflict resolution system in all of the GURPS tool box.

Yeah, you and Sindri have convinced me to abandon this idea.

Originally Posted by D10

My personal recommendation is to give the combat system a chance and play it out a bit. When I first started playing I found a bit bothersome that sometimes combats would take almost an entire game session to finish, or even more than one. But after getting used to it, I just feel like its part of the game, and a good one at that.

D10, thanks for reading my post and giving me your advice. I can see how my ideas would basically require re-engineering the whole game. I'll stick with the standard rules and see how it plays.

[Ulzgoroth]

Originally Posted by seismic73

I disagree, however, on the "boxers parry". For whatever it's worth: I box and have competed in the past. We dodge and block (absorb the punches on our gloves and arms, which are kept close to the body), but we don't parry.

I'm not sure what system of terminology you're using, but in GURPS that cannot possibly be blocking (blocking is a thing you do with a shield or occasionally a cloak) and definitely is parrying.

Originally Posted by seismic73

The rules state somewhere that simple tasks shouldn't require a roll. Depending on how one defines a "simple task", a tiger killing an invalid, or an assassin killing a comatose patient might be a "simple task". That's how I'd rather play it, but I get your point.

Do you know about Telegraphic Attack? It's the core GURPS mechanic for easily hitting things that you're not worried about defending themselves.

[seismic73]

I see what you mean. By GURPS terminology I would indeed be "parrying". (But no boxer would ever use the word "parry" to describe what they do.)

[Sindri]

Originally Posted by seismic73

I tried contour plots, surface plots, etc. I thought a table of values would be less easy to interpret, but I can post those in the future if someone wants to see them.

To be fair, tables of values don't give as immediate an impression of what's going on. They are, however easier to check for accuracy.

Originally Posted by seismic73

I believe you on the "pain to interact with mechanically". I disagree, however, on the "boxers parry". For whatever it's worth: I box and have competed in the past. We dodge and block (absorb the punches on our gloves and arms, which are kept close to the body), but we don't parry.

In GURPS unarmed combat skills give a parry, and that parry is what is raised by increased skill in the relevant unarmed combat skill and that score is what is used for active defenses that involve contact with the arms or hands. Boxing specifically even gets improved retreats like fencing skills.

A block is the active defense a shield makes, nothing else makes a block and shields never parry. For that matter there isn't actually any fundamental mechanical difference between a block and a parry. There's as much mechanical variation inside the world of parries as there is between a block and a typical parry. It would not be crazy if in a future edition they were given the same name and the differences treated as exceptions like fencing weapons have.

I realize that this may not line up with boxing terminology, but it doesn't line up with swordfighting terminology either and all active defenses done with a sword are parries. My point is just that if two boxers square off (hopefully outside the ring) in GURPS when one of them makes a successful attack the other is probably going to try to parry because parry scores rise faster than dodge.

Which is not to say that boxers can't dodge, martial arts training is a good excuse for increasing DX and to a lesser extent HT (moreso Fit than HT but still) and actual combat experience (or just cinematic kindness towards martial arts training) will give Combat Reflexes so they can have a higher dodge than normal.

Originally Posted by seismic73

Sindri, thanks for taking the time to read my post and for your advice. I'll stick to the standard rules, as you suggest.

Well far be it from me to discourage house rules. That said the GURPS combat rules have a lot of emergent properties that one might not get from just reading the text so here especially you'll want concrete experience. For example I mentioned Deceptive Attack and Ulzgoroth mentioned its converse of Telegraphic Attack and those are really key.

Once you've got that under your belt, then you can start to think about things T-Bone's FEND ( http://www.gamesdiner.com/fend-fully...ses-gurps-4e3e ) and many others.

[Tomsdad]

Originally Posted by seismic73

I see what you mean. By GURPS terminology I would indeed be "parrying". (But no boxer would ever use the word "parry" to describe what they do.)

Just to echo what some others have said, I would really recommend running some combats with the system as is to get a feel for what it's like in play. I would also recommend slowly adding in the the various combat options.

e.g the ones already mentioned Deceptive attack, Telegraphic attack, but also defensive attack, committed attack, and there's more after that even!

big tip though don't try and swallow the entire system in one go, you'll choke!


Personal GURPS confession time. I brought 3rd edition in the very early 00's and having flicked through it was a bit non-plussed and it just sat and gathered dust.

I didn't get 4th ed until 2006. Again I flicked through it and went 'meh' and on the shelf it went. There it stayed for about 6 months until I pulled it out on a whim (and a thread on a forum) and ran a couple of combats with it and well it clicked and I have never looked back!


So what kind of combat would you like to run? maybe we can help out with some worked examples, showing how different options and combat choices interact. Also what the rules actually represent, e.g. the boxers block or shoulder roll being a parry in GURPS terms. So you know how to represent in GURPS combat what you are envisioning for your games.

[D10]

Originally Posted by seismic73

The rules state somewhere that simple tasks shouldn't require a roll. Depending on how one defines a "simple task", a tiger killing an invalid, or an assassin killing a comatose patient might be a "simple task". That's how I'd rather play it, but I get your point.

In my table, rolling for this kind of simple situation is encouraged (specially when it comes to combat actions or spellcasting) preciselly because of critical failures. Not that we are interested in checking for crits in everything, but if a player critically fails he might burn his luck to roll twice more thus not fail, and then if another challenge/situation comes up where he would want/need his luck before the cooldown he might be in trouble.

So as Luck rises in importance, so does situations where critically failing can impose severe penalties.

For instance, altho hitting a sleeping person might be trivial, one of the results in the critical fail table is that you strain your shoulder for 30 minutes. That can be a real bummer depending on the situation, and by allowing the PC to simply kill the sleeping person without the small chance of that happening, you end up removing a risk inherent to combat actions that has a function in balancing the game.

Another example is the "you hit yourself result", these crit fail table results almost single handedly balance characters with ridiculously high melle damage.

I certainly wouldnt require the player to roll all his attacks and defenses in a training or sparring situation, but if he wants to break a door, slit a throat, etc. I think you need to make the roll happen in order to value the luck trait.

Originally Posted by seismic73

D10, thanks for reading my post and giving me your advice. I can see how my ideas would basically require re-engineering the whole game. I'll stick with the standard rules and see how it plays.

You are welcome

[tbone]

Originally Posted by Sindri

Once you've got that under your belt, then you can start to think about things T-Bone's FEND ( http://www.gamesdiner.com/fend-fully...ses-gurps-4e3e ) and many others.

I was just about to mention that; thanks. (Actually, it appears up early in the thread too – many years and one long-discontinued URL ago.)

This FEND is a look at using full skill (not Skill/2+3, etc.) for both attack and defense rolls. But to be clear, it's offered as a "just for the fun of it" experimental thing; I don't use it. More importantly, it doesn't quite address the actual topic. It suggests using full skill levels for defense, yes – but still following GURPS-like separate attack rolls and defense rolls, not Contests.

So. Setting that aside... Why not use Contests for combat attack/defense? For all the reasons everyone has given. For me, it boils down to this (which maybe someone has already expressed more skillfully):

Contests are usually used as a really easy way to play out a single opposed task, or a complex series of opposed tasks abstracted into a single roll for simplicity.

Combat doesn't fit either of those: it's not a single opposed task, and gamers – bless their violent hearts – generally don't want to abstract combat into some simple single roll. Most gamers just want to play things out blow by blow, with tactics and defending and little injuries and big injuries and falling down and (maybe) getting up again and...

If people did want to abstract combat into a really simple resolution mechanic, a Contest probably would be a good starting point. But GURPS players, like many others, want their detail in combat.

Okay, but: Even if we do play out combat blow by blow, why not use Contests for individual exchanges? My take is this:

The proposed Contest means that if I make my TH roll by a lot, you'll have to make a defense roll by a lot to avoid the hit. In other words, it means that I threw a particularly fast (or otherwise tricky) blow that's difficult to stop. Which may initially sound like a reasonable assumption...

...but in fact, there are many "special effects" I could attempt with an attack. To pick three common ones, I could try to hit a small target (like an eye), or I could try to make two attacks, or I could try to make an attack so fast that it overwhelms your defenses (as above).

The oddity with the Contest described above is that it just assumes I tried the third of those options. Yet why should it assume that? Why not instead assume that if I made my roll by a lot, then I automatically aimed for your eye (whether I actually want to or not)? Or why not assume that I automatically attempted two attacks (even though I intended no such thing)? Why just decree that I automatically attempted the third option?

I like the GURPS method better: If I want to go for your eye, I need to say so in advance and take the penalty (hit location). If I want to throw two blows, I need to declare that and take the penalty (Rapid Strike). And if I want to try blowing past your defenses with speed, I need to declare that and take the penalty (Deceptive Attack). I like how this places all my options on unified footing, and how it doesn't make any weird decisions for me about what I'm attempting with my action.

My humble addition to all that others have said –