Mechanic Revision - Pickpocketing

[HunterOG]

So I saw the mention of stats relating to pickpocket and it got me thinking; pickpocketing is one of the most enticing things in an RPG - it's always super appealing in concept and whenever I play a thief I power level that particular stat because I just cannot wait to get my hands on what everyone is carrying.

 

The downside is that the implementation is almost always the same - save in front of the mark, keep reloading until you are successful.  Essentially the only reason to bolster the stat is to save yourself the tedium of loading your game over and over again.  Couple this with the extremely stiff penalty of your target turning immediately hostile and you have, in my opinion, a broken mechanic.  It shatters suspension of disbelief and introduces a grinding exercise.  

 

So here are my questions:

1. Can the pickpocket ability be implemented into the interface in a different way than we are used to (the thief/steal cursor on the map)?  How can the mechanic be revised in order to make it more enjoyable, more flexible, and less upsetting to the flow of the game?

2. How can the mechanic be implemented in such a way that, ideally, a player never reloads, regardless of whether or not they fail?

3. Would there be any benefit to specialty pickpocketing?

 

=========================

 

1. Can the pickpocket ability be implemented into the interface in a different way than we are used to (the thief/steal cursor on the map)?  How can the mechanic be revised in order to make it more enjoyable, more flexible, and less upsetting to the flow of the game?

 

I'm of the opinion that the current style of pick pocketing in RPGs (the specialty cursor) doesn't really work.  For one, it relegates the target to a static object on the board, an impersonal dice roll.  Considering that the target of a pick pocket is supposed to be a living, breathing character, it seems a bit mechanical for something so intimate. 

 

Beyond that, the two executions we generally see are either 'NPC as chest', in which the mark has a specific object on him that can be lifted on a successful roll - or 'NPC as shopkeeper' in which we get a sort of inventory screen that has variable success rates for a whole mess of items (often everything that character is carrying).¹  I don't really buy either of these and I've found that neither of them are particularly rewarding afterward.  Pickpocketing should feel like a heist - you should feel like you got away with something, right?²  If you find something really rare or special, it should feel special.  Beyond that, there's an emotional element to theft that hasn't really been explored in games.  For example - just because you succeed in lifting someone's magic ring, who's to say that they won't suspect you after the fact?  What if you were steal something from an NPC that was super important?  What if the loss of that particular object was a death sentence?  These are just a couple examples, though I shouldn't get ahead of myself.

 

First, the steal: how to improve it?

 

My initial impulse is to include more circumstantial opportunities within the context of dialog trees.  If your pickpocket is high enough you can be granted the opportunity to snag something off a character during certain parts of a conversation.  This can also implement a sense of temporality to the activity - what if you were to steal someone's ring off their hand when they met you for the first time, or during a specific interaction?  This would make the opportunity a one-time affair, which actually solves a couple problems, namely:

 

A - Players won't know they missed an opportunity if their skill wasn't high enough, so they won't feel deprived of an experience.

B - Players would be encouraged to invest in the ability beforehand so that they are alerted to the opportunity when it arises.

C - By tying the act into the dialog tree, you can have the player fail while still acknowledging what they tried to do (eg. a handshake goes on a little bit too long, and the target feels uncomfortable from there on out).

 

It still doesn't solve the constant reloading issue (but we'll touch on that later).  Despite that, I think it could be a decent system to incorporate without impinging on a preexisting mechanic.

 

Another way to approach it would be to allow the pickpocket, upon reaching a certain level of proficiency, to sight and nab whatever he wants once he's achieved a high enough level (for example, when he walks near an NPC a little window pops up that shows him an item up for grabs).  No dice rolls, no reloading, no chance of turning the 'NPC circle' that hostile red.  He either can steal it or he can't, and he won't get the chance unless he can.  This may seem to destroy the sense of danger from the experience, but when was it ever really a risk for the player?  Who but the self-proclaimed masochist wouldn't reload after screwing up?  The risk factor in respect to the pickpocketing mechanic is a red herring - due to mitigation done by the player it could be said that it essentially does not exist.  But what could be introduced to replace it is the consequences of stealing from someone.  That, arguably, is where things would get really interesting - were he to steal something extremely dangerous or valuable.

 

A third angle could be the chance of guaranteed success, but items that are fixed based on your skill level.  Initially your pickpocket will only manage to pilfer loose change and pocket lint, but as you level up he'll be able to pilfer more valuable items.  So let's say the skill always works, but it only boosts from being used, and let's say you can only pickpocket an NPC once during the whole game.  Then the challenge becomes picking your targets at the right time, and choosing who to practice on, and who to save for later.³   Then the pickpocket is encouraged to sneak around more (so he can steal from enemies and raise the stat) and must also decide if a particular mark will be around later, when his skill is higher.  

 

Consequence

 

One thing that I think is completely under-explored is the consequence of theft.  In most games if players can get away with it at the time then it essentially didn't happen.  This is bogus and definitely rings false.  After burgling entire towns in some games, I start to really loath the citizens for being so stupid - how else could I have robbed everyone blind?  If the pickpocketing mechanic is made more intentional (as opposed to a dice roll) it offers up the chance for narrative engagement in the act.

 

The opportunities are endless and I don't really need to get into the specifics as I'm sure you can imagine your own scenarios.  Of the things which could be stolen: court documents, heirlooms, genie's lamps, royal seals, family crests, sacred relics, secret passwords, unlabeled tinctures, and on and on.  The theft of any one of these things could spiral out into the narrative and offer up something for the player to engage in.  The act of stealing it could have more importance than the thing itself (which is the basis for many great crime stories).  This, granted, is a ton of work for the designers.  But it would offer a certain facet of interest that I think games very much lack - namely the connection of items with the world the player occupies.  I think that more acknowledgement of what the player chooses to take would do a lot to make the game feel more real, and help to add facets to the concept of ownership in games.

 

Furthermore, it would add mystery to items in the game.  These days so many items are completely transparent.  As soon as you get them you know exactly what they are - "I give thing X to guy Y and get Z in exchange."  Remember the thrill of grabbing something and having no clue as to what it was for?  How much more exciting would that be if you stole it?

 

 

 

2. How can the mechanic be implemented in such a way that, ideally, a player never reloads, regardless of whether or not they fail?

 

I touched on this earlier.  My primary notion that I would want someone to take away is that idea of player risk - that when it comes to pickpocketing there is actually none - because they'll just reload to avoid the punishment.  How can a designer discourage this, or make the player not even want to?

 

One way is go straight Dark Souls - just lock down the action in the file and force the player to live with the fact that they didn't succeed.  If they fail, too late!  The game is saved and you can't go back.  This may seem punitive, but depending on the implementation of the mechanic it wouldn't have to be so bad: if the whole town didn't turn hostile and try and massacre them, for example.

 

Another would be to implement guaranteed success - sidestep the dice rolling for the particular mechanic and instead make it a proverbial door to open.⁴

 

Still another would be to downplay the punishments so that they are more subtle, and less obvious.  Maybe the NPC won't talk to them anymore, maybe they take a hit to reputation (which happens in many games already) - maybe something happens that the player can't actually see until later on down the road.  

 

Depending on the style of the game, the pickpocket could be the only one who takes the rap, rather than the whole party.  His poor reputation could be mitigated by player action using the other characters.

 

 

 

3. Would there be any benefit to specialty pickpocketing?

 

This is more of a sidebar, but still fits in with the idea of revising the mechanic.  Pickpocketing is a general skill, at the moment, but who's to say that we couldn't have specific kinds of pickpockets?  Depending on the type, the player could have different opportunities presented to them. 

 

A cutpurse, for example, could always succeed at filching coins from characters.  A sleight-of-hand artist could specialize in taking rings/watches off characters right in front of them.  Pickpockets could have proficiency in stealing items relating to their other skill sets (weapons, amulets, etc) or their race (short guys stealing items off the bottom half of the mark) or other traits that would allow them to thrive in a specific environment.

 

What if there were pickpocket teams?  Criminals of all types generally work in groups as it is, how much more interesting could things be if crooks worked in pairs or even larger groups?  Going too far could make the mechanic clumsy, but lets say you had one character with high charisma chat up an NPC while a pickpocket raids them for stuff.  What if you had a high-minded cleric speaking with an NPC, and had the option of your rogue stealing something during the conversation?  The upshot would be an item, but it could damage the relationship between the two party members.

 

 

 

Conclusion

 

This is all to say that I think the mechanic could be redefined to great benefit.  I think that right now it occupies the dice-rolling side of CRPG design - the part that that deals with chance and probability.  The numbers part of the game works great when it comes to battles and gameplay elements that aren't so binary, and offer the players a chance to correct their mistakes, mitigate a bad roll, or the possibility of moderate success/failure (as opposed to complete failure) - but I don't know if that mode really works so well for pickpocketing.  It might be better implemented as the opportunity for player discovery - a reward system that doesn't operate on a strict success/failure dynamic, rather a door to be opened.

 

If you made it all the way though, thanks!  I welcome any thoughts or input.

 

Cheers

Owen

 

 

===================================================

 

 

 

^1. I neglected to mention the gold you can lift off of peasants and other generics in CRPGs - I don't really consider this a full implementation of the mechanic, more of a bone thrown to the person who invests heavily in the skill.

 

^2. Actually, my personal favorite implementation of theft in a game so far is Assassin's creed.  Even though you're just stealing token amounts of spare change from NPCs, the physicality and slow intention of the action is extremely rewarding.  The fact that it was easy didn't bother me in the least.

 

^3. You may notice that most of these suggestions are based on the principle of guaranteed success.  That hasn't slipped by.  I think that failure in theft is a very compelling notion, however it doesn't really tie in with top-down style games for one reason in particular - the goal of a pickpocket who is caught is escape.  Fleeing in CRPGs almost never makes sense and usually feels stupid.  You're just watching a bunch of little dudes chase each other across the map in a slow mockery of pursuit.  To incorporate an escape mechanic into the genre in any meaningful way would require a ton more work and I'm not even certain of the returns, to be honest.

 

^4.  I can totally see the opposition to pickpocketing being 'too easy' - but I think that if there were going to be risk involved without constant quicksaving there needs to be some opportunity for player remedy, or at least the notion of partial success/failure.  Right now the mechanic works like a coin flip, and unless the player can somehow circumvent the punishment they'll just reload their save and break the system.

[Prometheus]

My initial impulse is to include more circumstantial opportunities within the context of dialog trees.  If your pickpocket is high enough you can be granted the opportunity to snag something off a character during certain parts of a conversation.  This can also implement a sense of temporality to the activity - what if you were to steal someone's ring off their hand when they met you for the first time, or during a specific interaction?  This would make the opportunity a one-time affair,..

If you steal an item in conversation wouldn't the person suspect you for stealing the item?

 

A third angle could be the chance of guaranteed success, but items that are fixed based on your skill level.  Initially your pickpocket will only manage to pilfer loose change and pocket lint, but as you level up he'll be able to pilfer more valuable items.  So let's say the skill always works, but it only boosts from being used, and let's say you can only pickpocket an NPC once during the whole game.  Then the challenge becomes picking your targets at the right time, and choosing who to practice on, and who to save for later.³   Then the pickpocket is encouraged to sneak around more (so he can steal from enemies and raise the stat) and must also decide if a particular mark will be around later, when his skill is higher.

I don't like learning by doing skills. First If you make pickpocketing learning by doing all skills should be learning by doing. Second I would need to pickpocket each enemy to raise my skill if I want to get the good items.

 

One thing that I think is completely under-explored is the consequence of theft.  In most games if players can get away with it at the time then it essentially didn't happen.  This is bogus and definitely rings false.  After burgling entire towns in some games, I start to really loath the citizens for being so stupid - how else could I have robbed everyone blind?  If the pickpocketing mechanic is made more intentional (as opposed to a dice roll) it offers up the chance for narrative engagement in the act.

One Thing they could do to make pickpocketing more interesting is to make trusting modifiers to pickpocketing. If somebody likes you(you did a quest for him/ the faction reputation is high, etc.) it would be easier to steal from him. But if you stole from a lot of people or he doesn't like you it will be more difficult to steal from him.

 

I think the best way to prevent a lot of reloading from pickpocketing (and other skills) is to make them not random. If you make an auto saving system(like you suggested) a lot of people will complain.

Edited August 3, 2013 by Prometheus

[Sacred_Path]

There is only one credible way of implementing pickpocketing, and that is via the stealth mechanic.

 

In reality, most pickpockets operate in pairs; one chats up the victim while the other rifles through parts of the clothing preferrably without body contact to the victim (i.e. jacket pockets, backpacks). In those cases where a pickpocket operates alone they will almost always try to touch you beforehand to divert your attention; but speaking for myself, when a stranger comes up to me and touches me I instinctively move away, and I don't know how people can so gullible as to just stand there.

 

Therefore, since our CRPG pickpocket will almost certainly work alone, and victims don't tend to wear modern ziplocked backpacks, the only way to steal would seem to be for the victim to be oblivious to the fact that you're even there.

 

How the pickpocketing then is handled interface wise - wether you should just get a random item or a "shop" inventory opens up, whether there should be opposing rolls and wether your stealth skill should give a bonus to your roll - I'm neutral about that.

 

Consequence

 

 

 

One thing that I think is completely under-explored is the consequence of theft. In most games if players can get away with it at the time then it essentially didn't happen. This is bogus and definitely rings false. After burgling entire towns in some games, I start to really loath the citizens for being so stupid - how else could I have robbed everyone blind? If the pickpocketing mechanic is made more intentional (as opposed to a dice roll) it offers up the chance for narrative engagement in the act.

 

 

 

The opportunities are endless and I don't really need to get into the specifics as I'm sure you can imagine your own scenarios. Of the things which could be stolen: court documents, heirlooms, genie's lamps, royal seals, family crests, sacred relics, secret passwords, unlabeled tinctures, and on and on. The theft of any one of these things could spiral out into the narrative and offer up something for the player to engage in. The act of stealing it could have more importance than the thing itself (which is the basis for many great crime stories). This, granted, is a ton of work for the designers. But it would offer a certain facet of interest that I think games very much lack - namely the connection of items with the world the player occupies. I think that more acknowledgement of what the player chooses to take would do a lot to make the game feel more real, and help to add facets to the concept of ownership in games.

 

 

 

Furthermore, it would add mystery to items in the game. These days so many items are completely transparent. As soon as you get them you know exactly what they are - "I give thing X to guy Y and get Z in exchange." Remember the thrill of grabbing something and having no clue as to what it was for? How much more exciting would that be if you stole it?

Consequence is the most important thing to get working. What you suggest could easily, in a game with just global reputation like in IE games, be simulated via a drop in reputation. Personally, I don't really see the need for that - it's one more thing that needs to be taken into account during balancing, and as you said consequences of failure are usually rather harsh anyway. Every new attempt to steal can mean disaster. I'm also not of the opinion that save-scumming needs to be prevented - it's one of those things that players can easily regulate themselves (as long as you don't make "no reload" stealing nigh impossible because of very low odds).

[Michael_Galt]

I really like the majority of that.  But they have already said that you won't level based on your use of a skill, so that is a moot point.  I think that the restriction of 1 single opportunity is silly.  Maybe there are conditions, like, how many times you have stolen from them (meaning, regardless of your skill, they have inherently become more wary of being stolen from, and might even suspect you, thus, making it harder to steal from them further), WHAT you have stolen from them (You steal 10 gp occassionally, not too important to them, if they are a noble- maybe they don't even notice.  But you steal their signet ring, and thereafter, they are EXTREMELY wary of thieves), HOW you steal from them (you initiate it during conversation, they will automatically suspect you after, you do it by stealth, they have no idea you were connected, you don't use stealth, but do steal from them, they have a slight idea it might have been you). 

 

I think that system would be far better.  Maybe you can't learn about certain really high value items without initiating conversation or having someone else tell you about them (like a thieves' guild, a friend/enemy, themselves, etc).  For example, you wouldn't know about their highly powerful charm they always keep in their breastpocket unless you somehow found out about it.  You could steal their ring, or from their purse, because, as a thief, you will naturally be looking there.  But you aren't likely to be feeling them up so you can locate items in less common places on their body or in their clothes....

[Walsingham]

I almost never pickpocket, because it's such a save'n'reload fest. I'd rather not do it at all.

 

I can understand leaving it in as is. But to get me fired up I'd want to be able to think tactically and improve or decrement my odds sensibly.

 

FWIIW I believe it would make much MUCH more sense to follow someone and wait for them to put stuff down than try and lift it off them. Or to con it off them.

[HunterOG]

Thanks Prometheus.  In response to your questions:

 

If you steal an item in conversation wouldn't the person suspect you for stealing the item?

 

Sometimtes.  Let's say you're playing and you enter into a conversation with an NPC.  This is the first time you've spoken, so he reaches out to shake your hand.  If your pickpocket is high enough, this gives you an opportunity to slip the ring off his finger during the handshake.  In this instance he might not necessarily notice what is happening - or maybe he would eventually notice, but only if you dawdled and talked to him for too long.  When I envision the mechanic working this way, I tend to imagine in the context of a classic IE game, where dialog is often interspersed with descriptions of characters' looks or actions.  In a game such as Fallout 3, where acting is visual, this sort of mechanic might not make as much sense.

 

 

I don't like learning by doing skills. First If you make pickpocketing learning by doing all skills should be learning by doing. Second I would need to pickpocket each enemy to raise my skill if I want to get the good items.

 

Fair point.  It can be tedious to repeat the same skill ad-naseum just to boost it up (I will never, EVER craft another iron dagger for as long as I live).  However if the mechanic were to auto-fire when it was switched on, or be woven into the interface fluidly enough that it can be performed very easily, this might not be so tiresome.

 

 

One Thing they could do to make pickpocketing more interesting is to make trusting modifiers to pickpocketing. If somebody likes you(you did a quest for him/ the faction reputation is high, etc.) it would be easier to steal from him. But if you stole from a lot of people or he doesn't like you it will be more difficult to steal from him.

 

For Sure - I agree 100% here.

 

 

I think the best way to prevent a lot of reloading from pickpocketing (and other skills) is to make them not random. If you make an auto saving system(like you suggested) a lot of people will complain.

 

That would almost definitely be the case (especially in this style of game) - but I figured it couldn't hurt to throw it out there.  As someone who doesn't have to move product, it can be fun to entertain more draconian design methods.

[HunterOG]

Thanks Sacred_Path; I tried to hit each of your talking points:

 

There is only one credible way of implementing pickpocketing, and that is via the stealth mechanic.

In reality, most pickpockets operate in pairs; one chats up the victim while the other rifles through parts of the clothing preferrably without body contact to the victim (i.e. jacket pockets, backpacks). In those cases where a pickpocket operates alone they will almost always try to touch you beforehand to divert your attention; but speaking for myself, when a stranger comes up to me and touches me I instinctively move away, and I don't know how people can so gullible as to just stand there.

Therefore, since our CRPG pickpocket will almost certainly work alone, and victims don't tend to wear modern ziplocked backpacks, the only way to steal would seem to be for the victim to be oblivious to the fact that you're even there.

How the pickpocketing then is handled interface wise - wether you should just get a random item or a "shop" inventory opens up, whether there should be opposing rolls and wether your stealth skill should give a bonus to your roll - I'm neutral about that.

 

The use of the stealth mechanic crossed my mind, and I'm unsure as to how I would prefer it to be used.  On one hand, the ability implies a level of stealth, so to 'activate stealth mode' would in many ways just be one extra button to click (at least in terms of how stealth is used in IE style games to this point) before doing the same old thing.  This style of game, being the abstraction that it is, always verges on the ridiculous.  Whenever my party swarms some stationary farmer like a horde of bees it crosses my mind that any sane human being would probably move aside or freak out.  That's part of the reason why I think it might make more sense to incorporate it into character conversation/interaction, it might sit more comfortably in that interface, rather than the field view.

 

I talked about pairs of pickpockets a little bit further down, and I do think that it could be implemented into a game if the designers felt that it had merit.  I think it could be great fun if it were to go in that direction.  Also, a pickpocket may work alone in some respects, but generally is a member of a party.  Who's to say they wouldn't take part in the execution of the crime?

 

 

Consequence is the most important thing to get working. What you suggest could easily, in a game with just global reputation like in IE games, be simulated via a drop in reputation. Personally, I don't really see the need for that - it's one more thing that needs to be taken into account during balancing, and as you said consequences of failure are usually rather harsh anyway. Every new attempt to steal can mean disaster. I'm also not of the opinion that save-scumming needs to be prevented - it's one of those things that players can easily regulate themselves (as long as you don't make "no reload" stealing nigh impossible because of very low odds).

 

 

Yeah, I never liked the 'reputation' style numeric myself - I find it to be too broad a rubric that is applied to very specific player actions.  At least when they're global.  For whatever reason I took a lot of satisfaction in my reputation when it came to particular locations, like in the Fallout games.  By locking my behavior to geography it definitely gave it more meaning than just being generically 'righteous' or 'villainous.'  It also made me feel 'known' in places I'd been and 'unknown' in new locales.  

 

In terms of consequences, I'm imagining scenarios that are scripted directly into the game and are specific to a particular theft - the altering of character behavior or the environment.  This could be in conjunction with a more general reputation system that encompassed the 'mathy' parts of the game - reaction rolls, shop prices, etc.

[Fearabbit]

My preferred system is always that which is very simple and intuitive while feeling kind of realistic. Some of your ideas are excellent in that regard, but others seem a bit confusing and/or frustrating. For example, if a merchant suddenly didn't want to sell me stuff because I robbed him once, that's very vague and I'm not likely to remember it. (What, I'm supposed to keep track of who I rob?)

 

In any case, my thoughts on the matter. I fear they won't be as nicely structured as yours.

 

1) Pickpocketing can be done in several different ways. One is the sleight-of-hand method where the victim doesn't realize he's being robbed. Another is the more typical variant for crowded places like bazaars and so on: You simply grab the item (the purse dangling from the belt, for example) and run away as fast as you can. I always hated it that pickpocketing in games is this very binary "either you succeed and nobody notices a thing, or you fail, don't get the item and get arrested" system. If I get caught stealing the item, the fun is only just beginning!

 

Idea: Items can be in different places of the body and they vary in weight. Depending on where they're kept and how heavy/big (i.e. obviously missing when removed) they are, there is no chance of stealing them unnoticed.

Example: You rob a passerby. A menu appears that tells you he has three items - an important letter in his coat pocket, a purse on his belt and his valuable rapier in his sheath. The slot with the letter is green, indicating that with your current pickpocketing skill, you can take it without him noticing. The purse is in a yellow slot, i.e. you can get it, but you'll have to run away (it's only secured by a small strap that can be cut through). The rapier is in a red slot - it's heavy and you can't easily access it.

There's a Thieves Guild quest to get the letter, and you could finish that up right now. But once he notices the robbery, the NPC will be more alert, raising the pickpocket skill you need for the other items so much that your only option is an open robbery in a dark alley, which will make you a known and hunted criminal (I just had that idea and I quite like it!).

You decide that the rapier isn't that important, but you do want to get the purse. That means you'll get the letter, then cut the purse off and run away as fast as you can, while the NPC calls for the guards.

 

2) The example above applies to crowded areas. If you are in a room alone with someone, or if a person has already seen your face because you were talking to them, you won't be able to run away (unless you want the guards to look for you). Basically, if you know a person, then you'll have to distract them to turn the "yellow slots" into "green slots". This is where the dialogue pickpocketing system comes into play.

One thing to consider: Generally in RPGs, NPCs only talk to the main character (which is a bummer because it makes no sense that my dumb half-ogre barbarian would do the talking - and fail at it - when there's a cunning rogue right next to him BUT ANYWAY). My main character may or may not be the rogue in the party, so the option to pickpocket someone during a conversation should appear independently of my main character's skills.

 

3) Consequences - what should they be? In my opinion, we don't want a generic system here. No reputation drop (doesn't make sense if you don't get caught) and no "everyone suddenly turns hostile". Consequences shouldn't be tied to failure, they should be tied to the way you achieved your goal, and to the quest that set you this goal. Basically, if a quest can be completed by pickpocketing something, then there should be some cool consequences. (In my example above, you could find out later that the letter contained state secrets that are now in the hands of a rival nation, or maybe some NPC wrongfully ends up in jail.)

Another possibility is to make the generic items you can get via sleight-of-hand (the ones in the "green slots") more rare as you commit more and more crimes in one city. People become more aware that there are pickpockets about, and the green slots turn into yellow slots, forcing you to make a run for it more often or put more points into pickpocketing.

 

But wait, what happens when you run away, anyway? What happens if you get caught?

- You take a "green slot" item: Nothing happens except for the consequences of stealing that item (see above).

- You take a "yellow slot" item: The NPC cries for help, guards chase after you. You have to leave the scene of the crime and go into hiding for about a day. If you don't succeed in running away, the guards will take the stolen item and put you in jail.

- You want to take a "red slot" item: You'll have to rob or kill the person. Robbery means the person sees your face, so killing them will be safer... but of course that means a murder investigation sooner or later.

 

The consequences of robberies and murders should be handled by a general crime system, and I'd very much welcome at least a rudimentary murder investigation somewhere in that... but that's not really the point of this thread, so let's forget about this for now. The real question is, do these possible failures for pickpocketing (ending up in jail and not getting the item) still encourage the player to reload?

Well, yes. Of course they do. That's why they are so far removed from the crime - the idea is that generally, the player succeeds in stealing the item and getting away with it. In the case of running away, there's a general "go into hiding" penalty that can't be avoided and thus doesn't encourage reloading. Getting caught when running away is like dying in combat - you seriously screwed up and yes, reloading is totally cool in that case. Unless you want to roleplay getting caught, or maybe there's even a quest that requires you to get caught.

Nobody gets punished for reloading after getting caught, but there is a way to keep playing if you do not do that, and it's more fun than having to kill every NPC in the town because you got failed at pickpocketing 2 gp and now everyone is hostile.

 

And of course with a crime system that makes you go to jail instead of simply killing you, there's the possibility of adding a whole lot more that makes "failing" actually fun. Escape from prison, building up a good reputation with criminals the more you get caught (street cred, so to say)... sky's the limit. Basically if there's always one more bit of content that results directly from you "screwing up", then the game keeps you interested in your own story, and you won't feel the need of reloading.

[Walsingham]

So, hold on. If I can winnow your last post a bit, fearabbit...

 

1.Don't be noticed. [test, using: ambient distractions, plus players distraction, plus skill, less any observers]

2. If you are noticed, don't hang about [player intervention on warning]

3. If you do hang about, don't get grabbed [test? player intervention? Combination?]

4. If you do get grabbed, don't get caught [unclear, depends on tactical combat engine, probably]

5. If you get caught, don't go to jail [bluff or fight?]

6. If you go to jail, don't do time [escape use case]

7. If you do time, don't do hard time [within jail use case; alliances, guards etc]

8. If you do hard time, don't die [fighting, alliances, psychological resilience]

Edited August 4, 2013 by Walsingham

[Fearabbit]

So, hold on. If I can winnow your last post a bit, fearabbit...

 

1.Don't be noticed. [test, using: ambient distractions, plus players distraction, plus skill, less any observers]

2. If you are noticed, don't hang about [player intervention on warning]

3. If you do hang about, don't get grabbed [test? player intervention? Combination?]

4. If you do get grabbed, don't get caught [unclear, depends on tactical combat engine, probably]

5. If you get caught, don't go to jail [bluff or fight?]

6. If you go to jail, don't do time [escape use case]

7. If you do time, don't do hard time [within jail use case; alliances, guards etc]

8. If you do hard time, don't die [fighting, alliances, psychological resilience]

 

I actually think that's a bit more complicated than what I proposed. But yeah, I know my post up there was all over the place.

Let me try to sum up the most important points...

 

1. Decide whether to play it safe or take the risk.

2. If you play it safe, three options: Wait till you're better, use (quest-related*) distractions or only take the easy stuff.

(This is basically all that's related to "pickpocketing" as the crime, everything below should be part of the usual crime system.)

3. If you take the risk, prepare to run away and hide. [basic combat with the guards, only you try to run away from it]

4. If you run, don't let the guards knock you out or corner you in such a way that you can't run anymore.**

5. If you can hide, lay low for a while. [Not sure yet what "getting away" and "laying low" mean gameplay-wise]

6. If they do corner you, they'll put you in jail.

7. Once in jail I guess there can be several ways of making it interesting - a trial you can weasel yourself out of, escaping, whatever.

 

*) By that I mean that distractions shouldn't be items you can buy or something you can generally do, because that would defeat the purpose of having many layers to the system - but many pickpocketing attempts should be part of quests and these quests can have special conditions, allowing you to use dialogue pickpocketing or prepared traps etc.

**) I think fighters have a class feature that allows them to keep enemies from running away? So something like that could work against you. Also when running away, the guards might be able to attack you a couple of times, so if you're unlucky you might get knocked out. In the latter case, you don't die but also wind up in jail.

 

Also, stop me if this is starting to get too complex but I thought that maybe the pickpocketing menu could work like this:

- An NPC has clothing, right? So what if this clothing has pockets, straps and so on where items might be placed.

- Depending on where they are placed, your pickpocket skill determines whether you can see the item or not. No matter your skill, you'll always see a purse dangling from the belt, for example, but you might not see a watch in the coat pocket.

- The slots also have a predetermined difficulty - coat pockets are easy, but pockets on tight trousers are difficult, and so on.

- So the each pocket has a "difficulty" rating and a "visibility" rating.

- So you see a menu with several slots. You see their difficulty and depending on your skill, you see whether something's inside.

- There may be multiple items inside a pocket.

- You can empty out all pockets that are marked as "can be pickpocket without alarm" for your skill, but you can choose only one from those that "can be pickpocket but alarm the NPC", whether you there's an item in it or not, and whether you can see that item or not. (If you can see what's inside, and there's nothing in there, the pocket may be made impossible to pickpocket if only to avoid accidents.)

- Additionally, all clothing has a difficulty rating, which is VERY high and never goes lower than "will alert the NPC". So a skilled pickpocket could grab the whole cloak from an NPC, but will have to run away afterwards. (This may be complexity overkill but I imagine it could be fun.)

 

I think that sounds more complicated on paper than it would actually be when you play it, and the system itself isn't much more complex to script than a normal inventory. Would like to hear what you guys think about it.

[Mr. Magniloquent]

In reality, most pickpockets operate in pairs; one chats up the victim while the other rifles through parts of the clothing preferrably without body contact to the victim (i.e. jacket pockets, backpacks). In those cases where a pickpocket operates alone they will almost always try to touch you beforehand to divert your attention; but speaking for myself, when a stranger comes up to me and touches me I instinctively move away, and I don't know how people can so gullible as to just stand there.

 

Hehe. A buddy of mine used to do this in NWN. We'd join a hardcore PnP rules, permanent death, role-playing server where money was scarce and talk was cheap. Both of us would be halfling rogues maximized with dexterity with pickpocketing focus feats. I'd find a mark, then drag him to a tavern with some ad-hoc story about needing help with a quest. Once seated and thouroughly engrossed, my stealthy friend would clean the mark out.

 

Few things were more satisfying than having a rumor come around about two mysterious--and beguilingly well equipped adventurers in town. There was one server where our luck ran out though. I was able to talk my way out of it, at great monetary cost *wink*, but my friend was hung like the dirty theif that he was.

 

Good times.

[HunterOG]

I really like the majority of that.  But they have already said that you won't level based on your use of a skill, so that is a moot point.  I think that the restriction of 1 single opportunity is silly.  Maybe there are conditions, like, how many times you have stolen from them (meaning, regardless of your skill, they have inherently become more wary of being stolen from, and might even suspect you, thus, making it harder to steal from them further), WHAT you have stolen from them (You steal 10 gp occassionally, not too important to them, if they are a noble- maybe they don't even notice.  But you steal their signet ring, and thereafter, they are EXTREMELY wary of thieves), HOW you steal from them (you initiate it during conversation, they will automatically suspect you after, you do it by stealth, they have no idea you were connected, you don't use stealth, but do steal from them, they have a slight idea it might have been you). 

 

I think that system would be far better.  Maybe you can't learn about certain really high value items without initiating conversation or having someone else tell you about them (like a thieves' guild, a friend/enemy, themselves, etc).  For example, you wouldn't know about their highly powerful charm they always keep in their breastpocket unless you somehow found out about it.  You could steal their ring, or from their purse, because, as a thief, you will naturally be looking there.  But you aren't likely to be feeling them up so you can locate items in less common places on their body or in their clothes....

 

RE: 'skill level by use' - sorry about that - I'm jumping in kind of late here and haven't had the chance to catch up on everything that's been referenced thus far.  I'm mostly interested in exploring the implementation of new ideas, so if I throw something out there that's already been nixed, I'm not trying to be difficult, I swear!  Thanks for the heads up though

 

The more I've been reading over talk on the boards, the more I find myself wondering about player motivation.  It's interesting to think that we all love these games, but that there's been a dearth of them recently, and that almost all of us enjoy them in retrospect - the reasons we loved them in the past might not be what drives us to play a game today.  It's really interesting seeing the variety of preferences that folks who've been excited about this sort of game tend to have.

 

Perhaps certain kinds of theft have different manners of execution?  Stealing a coin purse could be fairly impersonal and done through stealth, while stealing a ring or amulet would have to be done in close quarters, during a conversation.  Depending on the variety of NPC behaviors, maybe certain items can only be stolen when an NPC is engaged in a specific activity (this kind of Walsingham brought up - following someone and taking their stuff when they set it down).

 

The idea of a precondition for pickpocketing certain items was something I had thought about as well!  I think that could be a rewarding gameplay element, and also get people more engaged in dialog with NPCs.  If someone mentions an item of value that a character is holding, then you would know to look for it.  I think this happened in BG occasionally, but it was never a requirement - rather a clue as to who you might want to steal from.  It could really broaden the dynamic without having to completely reinvent its functional elements.

 

As I think about it more, I find that in many cases these sorts of mechanics can't help but come up against the manner of abstraction that an IE-style game engages in.  In most instances, your decision making as a player is limited to a few specific areas: the field and command cursors, dialog trees, item/inventory management, and combat mechanics (which are usually embedded in the field interface).  One could either fold a new mechanic into the existing interface, or frame out a new interface for its use.  In the above post I'm mostly concerned with remaining under the comfortable, pre-existing umbrella and how that relationship could be improved, but it does lead me to wonder if stealth abilities mightn't get their very own interface of some kind.  I'll think about this some more, but if anyone has any interesting examples I'd welcome them.

[HunterOG]

save-scumming

 

Thank you for introducing me to this expression, by the way

[J. Trudel]

I actually thought abnout this mechanic in cRPG a couples of time before and I don't hink any cRPG I have tried got it right yet. I don't even think my solution is good but here it goes. 

 

1st part : Identifying the target. 

 

Opening a menu when you pickpocket is lame, most items a character carries are imposible to pick pocket. What I suggest is a different approach. From afar you can ''take a peek'' at what you could steal from the target and select it. For exemple, you sit in a bar and you ''take a peek'' at different NPC. You notice that one of them carries a big purse. You select it for your target. What is cool about it is the fact that sometimes you don't know anything about what you'll end up with. This part is also without consequences. 

 

2nd part : The ''pick'' 

 

Now you know what you want. It's just a matter of geting it. You make an attempt (diffenrent method could be applied here) , and either fail or succeed via an opposed roll agaisnt your target appropriate ''defense'' on an equal Pick-pocket vs defense you get this probabilities (based on the P:E hit roll mechanics)

1-5 % : Grab the item but get noticed by the NPC

6-50 % : Failed to grab the item (You thought it was too risky to get caught)

51-95 % : Suceed, but don't stay around for too long or the target may get suspicious.

96-100 % : You weren't seen at all. The target may actually think it was another NPC

 

3rd part : Getting away with it. 

 

With this system, most of the fails doesn't give any penality. However if you fail and the target notice you they shouldn't turn hostile. They may actually start a conversation and ask for what you took back. This person disposition toward you would then be badder but you may have the option to be forgotten for what you did. Could actually be a nice side quest. This way, sometime a failed pick pocket attempt might end up as a reward, this could be an interesting way to encourage the player not to reload, cause you never know what good can come out of a bad action.

[JFSOCC]

My preferred system is always that which is very simple and intuitive while feeling kind of realistic. Some of your ideas are excellent in that regard, but others seem a bit confusing and/or frustrating. For example, if a merchant suddenly didn't want to sell me stuff because I robbed him once, that's very vague and I'm not likely to remember it. (What, I'm supposed to keep track of who I rob?)

 

In any case, my thoughts on the matter. I fear they won't be as nicely structured as yours.

 

1) Pickpocketing can be done in several different ways. One is the sleight-of-hand method where the victim doesn't realize he's being robbed. Another is the more typical variant for crowded places like bazaars and so on: You simply grab the item (the purse dangling from the belt, for example) and run away as fast as you can. I always hated it that pickpocketing in games is this very binary "either you succeed and nobody notices a thing, or you fail, don't get the item and get arrested" system. If I get caught stealing the item, the fun is only just beginning!

 

Idea: Items can be in different places of the body and they vary in weight. Depending on where they're kept and how heavy/big (i.e. obviously missing when removed) they are, there is no chance of stealing them unnoticed.

Example: You rob a passerby. A menu appears that tells you he has three items - an important letter in his coat pocket, a purse on his belt and his valuable rapier in his sheath. The slot with the letter is green, indicating that with your current pickpocketing skill, you can take it without him noticing. The purse is in a yellow slot, i.e. you can get it, but you'll have to run away (it's only secured by a small strap that can be cut through). The rapier is in a red slot - it's heavy and you can't easily access it.

There's a Thieves Guild quest to get the letter, and you could finish that up right now. But once he notices the robbery, the NPC will be more alert, raising the pickpocket skill you need for the other items so much that your only option is an open robbery in a dark alley, which will make you a known and hunted criminal (I just had that idea and I quite like it!).

You decide that the rapier isn't that important, but you do want to get the purse. That means you'll get the letter, then cut the purse off and run away as fast as you can, while the NPC calls for the guards.

 

2) The example above applies to crowded areas. If you are in a room alone with someone, or if a person has already seen your face because you were talking to them, you won't be able to run away (unless you want the guards to look for you). Basically, if you know a person, then you'll have to distract them to turn the "yellow slots" into "green slots". This is where the dialogue pickpocketing system comes into play.

One thing to consider: Generally in RPGs, NPCs only talk to the main character (which is a bummer because it makes no sense that my dumb half-ogre barbarian would do the talking - and fail at it - when there's a cunning rogue right next to him BUT ANYWAY). My main character may or may not be the rogue in the party, so the option to pickpocket someone during a conversation should appear independently of my main character's skills.

 

3) Consequences - what should they be? In my opinion, we don't want a generic system here. No reputation drop (doesn't make sense if you don't get caught) and no "everyone suddenly turns hostile". Consequences shouldn't be tied to failure, they should be tied to the way you achieved your goal, and to the quest that set you this goal. Basically, if a quest can be completed by pickpocketing something, then there should be some cool consequences. (In my example above, you could find out later that the letter contained state secrets that are now in the hands of a rival nation, or maybe some NPC wrongfully ends up in jail.)I like the gist of this idea, it's got some similarities with other ideas. I do think that the ability to see items should be limited. Perhaps if you had a high perception skill. or if you knew the item was there before hand.

 

For instance, you can see the purse, but you can't know what's in it. a ring on a finger is not hidden. A medallion or necklace might or might not be hidden. whatever is in someone's inner shirt pocket is invisble.

 

Another possibility is to make the generic items you can get via sleight-of-hand (the ones in the "green slots") more rare as you commit more and more crimes in one city. People become more aware that there are pickpockets about, and the green slots turn into yellow slots, forcing you to make a run for it more often or put more points into pickpocketing.

 

But wait, what happens when you run away, anyway? What happens if you get caught?

- You take a "green slot" item: Nothing happens except for the consequences of stealing that item (see above).

- You take a "yellow slot" item: The NPC cries for help, guards chase after you. You have to leave the scene of the crime and go into hiding for about a day. If you don't succeed in running away, the guards will take the stolen item and put you in jail.

- You want to take a "red slot" item: You'll have to rob or kill the person. Robbery means the person sees your face, so killing them will be safer... but of course that means a murder investigation sooner or later.

 

The consequences of robberies and murders should be handled by a general crime system, and I'd very much welcome at least a rudimentary murder investigation somewhere in that... but that's not really the point of this thread, so let's forget about this for now. The real question is, do these possible failures for pickpocketing (ending up in jail and not getting the item) still encourage the player to reload?

Well, yes. Of course they do. That's why they are so far removed from the crime - the idea is that generally, the player succeeds in stealing the item and getting away with it. In the case of running away, there's a general "go into hiding" penalty that can't be avoided and thus doesn't encourage reloading. Getting caught when running away is like dying in combat - you seriously screwed up and yes, reloading is totally cool in that case. Unless you want to roleplay getting caught, or maybe there's even a quest that requires you to get caught.

Nobody gets punished for reloading after getting caught, but there is a way to keep playing if you do not do that, and it's more fun than having to kill every NPC in the town because you got failed at pickpocketing 2 gp and now everyone is hostile.

 

And of course with a crime system that makes you go to jail instead of simply killing you, there's the possibility of adding a whole lot more that makes "failing" actually fun. Escape from prison, building up a good reputation with criminals the more you get caught (street cred, so to say)... sky's the limit. Basically if there's always one more bit of content that results directly from you "screwing up", then the game keeps you interested in your own story, and you won't feel the need of reloading.

I'd rather you didn't exactly know what you were stealing and that the difficulty (red/yellow/purple, to avoid colour blind problems)

is determined not by item, but by item slot. So you know there is a red item slotor a yellow slot or you're stealing from. If you don't first investigate your victim, you'll have to steal blindly.

 

Perhaps an interesting scouting/investigation/research skill working in conjunction with pickpocket would make the skill more dynamic.

the position from which you steal should matter too. If I walk up to someone and take something from his or her sightline, that should be easier to spot. (duh)

I also think that rather than dealing with absolutes (red is always get caught, etc) it should be a likelihood, one which decreases with player skill.

I proposed somewhere else that in cases of a pickpocket getting noticed, it might be nice if there is a grace period in which you can flee the scene before it gets noticed. If you can flee on time, you're off the hook.

Some targets may notice some thefts faster than others.

I'd like to make it so that while possible, unnoticed theft is only possible if you are extremely highly skilled, or focus on only the easiest pickings(So,you'd pick the green slots only.)

 

you would not know what it contains unless it's visible. so, purse slot shows the purse, but not what's in it, pocket slot is hidden, only if another source tells you what's in there will you know. You can try to steal it, but there may be nothing in there, or it might be irrelevant.

wrist slot would show someone's bracelet, unless it's covered with armour.

 

I would also avoid the "just looking" trope by starting a timer the moment you try to pickpocket someone, each theft also takes time of the timer. (based on skill level check vs someone's perception check)

a well planned theft can be executed with bonuses. a partner distracting the victim, a public distraction (smokebomb, a fight nearby), an unconscious victim has no defence, inebriation or intoxication (make someone drunk, or slip a roofie in their drink. lure then into a opium den and see the smoke get to them)

some items may give bonuses too. (specialized equipment for a specific slot, specialised training for a specific slot)

 

And if my actions become bolder, I'd love to see some reactivity in the world for it. More guards in places that have seen a lot of burglary or theft, some npc banter. Warrants with the wrong name on it, reward posters for information, etc.

[Michael_Galt]

I am really hoping that the developers read this thread, as there are a lot of good ideas here.  It is seriously an important element, which is often done poorly.  I understand there would be tons of scripting and writing necessary to incorporate some of these ideas, but I would personally rather have an excellent "thieving system" than a 15 level dungeon...  15?!?!?   Really?  Considering they didn't even make 15 level castles, it just seems ridiculous.  I remember Watcher's Tower from BG 2, and even that seemed a bit overboard... and most levels were rather small.  I can personally say, that chances are, I won't go beyond a few levels of this dungeon, because I don't like senseless grinding and unless it is incorporated into the main storyline, won't be motivated to go there... And I hope it isn't, because I don't want Temple of Elemental Evil all over again.  I just can't imagine my character thinking, "Well, a mass civil war/inception of great destruction/my death/__ looms on the near horizon, but I'm going to spend the next several months trying to clear out this dungeon, because there are ancient treasures here..."

 

Back on topic, I completely agree that failing should not automatically result in EVERY NPC trying to attack you (including random children), that there should be a degree of ambiguity about it (not a menu of items, as your character likely does not have x-ray vision, though maybe a Cipher thief might not be a bad idea...), that there should be different ways of initiating stealing from an individual, and that it should get more difficult over time, based on how many times that individual has been stolen from/or the value, or the how many times you have stolen from people in that area in general (and/or the cumulative value).  For instance, if you have been stealing from everyone's coinpurses, maybe there are more guards in the streets.  If you steal from merchants, maybe there are more guards inside shops.  If you steal from both, more guards everywhere, and general paranoia and rewards out to catch the thieves responsible- maybe it creates a storyline in and of itself, because you have stolen so much.  That would be interesting.

 

So I am going to randomly write their names, in the event that they actually keep track of threads/posts that mention them... Chris Avellone, Josh Sawyer, Tim Cain, Feargus Urquhart, Chris Parker

Edited August 6, 2013 by Michael_Galt

[anubite]

I think any attempt to curtail save scumming / reloading is just going to fail. If the mechanic is so fundamentally wrong that it forces a reload then preventing said reload means players just won't take the risk.

 

Pickpocketing in most games, even if it ends up not being something you spam reload for, is imbalanced. You either get way too good unique loot by doing it, or you get garbage trash and it's not fun/woth the effort.

 

I could see pick pocketing as a context-sensitive skill, where you can only pickpocket certain story-related NPCs under story-related events. But then, the skill is rather restricted and limited.

 

Auto-success won't work, we'll either have rewards that just aren't worth pressing the button for, or players will reload until they get a specific item they wanted to pick pocket, or, we'll be giving players with pickpocket way too much easy money.

 

If pickpocket must be implemented, it should ideally only work against enemies that can fight back. You shoudl be forced to pickpocket in combat, where there is risk attached (you're wasting turns stealing).

[Kjaamor]

Just as a gentle reminder on the root matter of the topic, pickpocketing is not necessarily something that all players find 'enticing'. Indeed, when I saw the title for this thread, my first thought was 'Ah, that'll be a topic about getting rid of that pointless skill, I'll go and offer my support for that'. I was very surprised to find out it was going in completely the opposite direction.

 

Frankly, and this could be considered pretty poor form, I've done an awful lot of tldnr in this thread. The only reason I'm here is just to remind devs that while pickpocketing could evidently use a serious overhaul, there's a very good chance many of us would continue to ignore it. In any game where my hand hasn't been forced, I have never put a single point into that skill, and despite the passion of the thread's other contributors I feel I am hardly alone.

 

So that's not so say the ideas within are bad, just that should they prove costly to implement they might not be worth it.

Edited August 6, 2013 by Kjaamor

[moridin84]

In the new XCOM game they have a mechanic that stops "save scumming" without saving the game each time.  

 

Save a game just before taking a shot at an alien and miss. No matter how many times you reload the game that shot will always miss. 

http://xcom.wikia.com/wiki/Random_vs_Seed_(XCOM:_Enemy_Unknown)

Edited August 6, 2013 by moridin84

[rjshae]

The OP makes some good points. Rather than a success/fail, I'd like to see something more akin to a recursive partial success:

1. You receive a pop-up showing a set of items you can nab.
   • It only shows those items that you can access based upon your skill rank versus the opponent's save rank, with modifiers for the item placement. I.e. body armor should get a massive save bonus; items hanging from the belt get no modifier.
   • You only get a general impression about what's available to pinch, so you only see things like armor, weapon, garb, trinket, &c. Items can be better identified once you exit the dialogue.
   • You can get synergy bonuses when another member of the party has a comparable pick-pocket skill.
2. Each time you take an item from the pop-up, the list is regenerated with the opponent gaining a cumulative bonus to their save. This repeatedly shortens the list of remaining items, and at some point there is nothing left to take.

[rjshae]

In the new XCOM game they have a mechanic that stops "save scumming" without saving the game each time.  

 

Save a game just before taking a shot at an alien and miss. No matter how many times you reload the game that shot will always miss. 

http://xcom.wikia.com/wiki/Random_vs_Seed_(XCOM:_Enemy_Unknown)

 

That's easy to implement by pre-generating a key for each possible action and Xor-ing that with a key generated for the game; the modified key is used to generate the random result, so that you always get the same outcome.

[Michael_Galt]

Just as a gentle reminder on the root matter of the topic, pickpocketing is not necessarily something that all players find 'enticing'. Indeed, when I saw the title for this thread, my first thought was 'Ah, that'll be a topic about getting rid of that pointless skill, I'll go and offer my support for that'. I was very surprised to find out it was going in completely the opposite direction.

 

Frankly, and this could be considered pretty poor form, I've done an awful lot of tldnr in this thread. The only reason I'm here is just to remind devs that while pickpocketing could evidently use a serious overhaul, there's a very good chance many of us would continue to ignore it. In any game where my hand hasn't been forced, I have never put a single point into that skill, and despite the passion of the thread's other contributors I feel I am hardly alone.

 

So that's not so say the ideas within are bad, just that should they prove costly to implement they might not be worth it.

I don't know that the "majority opinion" is, but I sincerely doubt that the majority of players don't use thieves, especially to steal when it is possible and worthwhile.  The 3 base classes were fighter, thief, and mage.  In my experience, people tend to have their preferences, but tend to play some variation of each of those for a playthrough.  My least favorite is a "rogue", because they generally aren't that great in combat.  That being said, I very much enjoyed my assassin in BG2, which had boots of speed and infravision, that would run around slaughtering people one by one, by hiding (using potions of invisibility as well) and then luring them off or attacking right in their midst.  Even when I had my paladin character, I justified my thief's stealing as a necessary evil, because I was not able to secure enough funds to be properly outfitted to fight evil, which was the higher priority.  I could save more people being better outfitted and made large donations to the various churches (mine especially, of course).  In Arcanum and Fallout, I stole like a boss, because it was hard to come by lots of things any other way.  In DAO, thievery was useless, so I stopped playing my thief character. 

[HunterOG]

 

The OP makes some good points. Rather than a success/fail, I'd like to see something more akin to a recursive partial success:

1. You receive a pop-up showing a set of items you can nab.
   • It only shows those items that you can access based upon your skill rank versus the opponent's save rank, with modifiers for the item placement. I.e. body armor should get a massive save bonus; items hanging from the belt get no modifier.
   • You only get a general impression about what's available to pinch, so you only see things like armor, weapon, garb, trinket, &c. Items can be better identified once you exit the dialogue.
   • You can get synergy bonuses when another member of the party has a comparable pick-pocket skill.
2. Each time you take an item from the pop-up, the list is regenerated with the opponent gaining a cumulative bonus to their save. This repeatedly shortens the list of remaining items, and at some point there is nothing left to take.

 

 

This is an excellent and elegant distillation of my original post.  If the term 'recursive partial success' had been in my lexicon I might have managed to cook a real meal that evening.  The execution works more or less as I'd enjoy seeing it as well.  I threw out a couple iterations or approaches, but this one would work for me.

 

The long and short of it is that I think the binary 'success/fail' behavior of a lot of non-combat skills (pickpocketing in this instance) in these games are some of the weaker design aspects.  Aesthetically and thematically, I love the abilities.  Mechanically speaking I'm essentially flipping a coin several thousand times which seems woefully simplistic, especially when compared to such a dynamic and sophisticated combat structure.  

[Kjaamor]

I don't know that the "majority opinion" is, but I sincerely doubt that the majority of players don't use thieves, especially to steal when it is possible and worthwhile.  The 3 base classes were fighter, thief, and mage.  In my experience, people tend to have their preferences, but tend to play some variation of each of those for a playthrough.  My least favorite is a "rogue", because they generally aren't that great in combat.  That being said, I very much enjoyed my assassin in BG2, which had boots of speed and infravision, that would run around slaughtering people one by one, by hiding (using potions of invisibility as well) and then luring them off or attacking right in their midst.  Even when I had my paladin character, I justified my thief's stealing as a necessary evil, because I was not able to secure enough funds to be properly outfitted to fight evil, which was the higher priority.  I could save more people being better outfitted and made large donations to the various churches (mine especially, of course).  In Arcanum and Fallout, I stole like a boss, because it was hard to come by lots of things any other way.  In DAO, thievery was useless, so I stopped playing my thief character. 

 

I agree that the "majority opinion" is unclear, and without that I'm just talking about personal experience. However, between myself and other offline friends, I've noticed that it is not uncommon for people to roll what effectively boils down to an archer/lockpicker/trap-disarmer class. Personally, I have seldom used stealth with my rogues/thieves, and when railroaded into single-classing have more often used them in mage-killer roles with a bow.

 

Generally, stealth is a better-handled mechanic than pickpocketing, of course, which is frequently a discipline requiring a high degree of time and effort for relatively little reward compared to returning bandit scalps.

 

Again, I don't necessarily disagree with anything posted in support of the OP, and I agree that pickpocketing needs an overhaul - but I feel that full removal should be an option if P:E would otherwise leave this sorry skill (and the wealth mechanics) in their present state.

[anubite]

Fundamentally, failing a pickpocket is not fun. You probably get some kind of (likely very harsh) social punishment, plus you also don't get any loot.

A pickpocket is fun only if a success nets you something worth the risk. If the risk is small, the reward must be small. If the reward is small, who cares? If the reward is large, the risk must be large. If the risk must be large, we must prevent save scumming. If the risk is large, then this means a large percentage of people who specialize in pickpocket will fail to get the item they desire from the target. The end result is most people will be paralyzed in fear at the idea of using it. Allowing them to save scum is the only way the average user will consider using it, but save scumming is gaming the game. It's kind of a stupid mechanic.

 

It's a fundamentally broken system in a cRPG. In a pen and paper game, I guess it's fine.

 

Stealth on the other hand really has no place in an isometric cRPG. Every cRPG I can think that offered it... I never once was able to use it. Namely because it's only really effective if your entire party can stealth. If they all can't stealth, then it can only be used to initiate encounters. Maybe I could see it being used to allow thieves to access certain levels or rooms that are guarded by monsters, in order to spring a trap on them or something, but that seems like an awfully niche function of the skill. Theives generally aren't good at initiating fights, since they get focus fire'd...

 

In a pen and paper game, I can see it working becaues the way the game is played is totally different. But stealth in a cRPG is just a hassle that's hard to design around.

Edited August 8, 2013 by anubite