How do I start adding numbers to my abstract idea for a game system

[valadil]

Hi.  I've had an idea for a game bouncing around in my head for a while and I want to get serious about writing it up and playtesting my ideas.

The basic idea is simple.  Take a character sheet not unlike world of darkness's.  You have a bunch of stats and a bunch of skills.  When you need to make a check you add these together and compare them to the difficulty target number.  No dice or other randomness.  You will however have a pool of will or fate points that can be used to increase your chance of success.  Where this gets interesting will be in opposed rolls.  Players will essentially bid for success.

The whole point of this system is for me to see how bidding works as a conflict resolution mechanic in place of dice. 

I've come up with a variety of ways to complicate the system.  Lists of available skills/subskills, will points for each stat, effects that can be bought with leftover will points if you go over, etc.  What I have not been able to do is come up with numbers.  I don't know if players should have 5 will points per day or 50, or even 50 per stat.  What's the range on stats and skills?  Even worse, how about advancement?  Can these points be used as experience (a la Deadlands) and if so, what are they worth?  All these things seem connected, but I can't find a starting point.

So that's where I'm stuck.  I'm tempted to pick something arbitrarily and fudge around that, but it seems like there should be a better method.  I've been googling around all day for relevant articles (much to my boss's chagrin) but all I can find is roleplay/GM theory or probability.  Any suggestions?

[Luke]

Hey man,
You're in a great place. And your gut is correct. Start by picking some numbers arbitrarily. Sit down with a friend and run through some really quick tests -- no roleplaying, just pure mechanics. Keep at it until you get something that just feels right.

Then work up some characters based on those numbers. Write 'em out, assign values. Invite friends over and try to roleplay with those characters. See how it goes. Take notes. Make changes after the session. Rinse and repeat until you've got a solid range of numbers.

-L

[lumpley]

Luke's right on.

If it were me looking for an arbitrary starting place, here's what I'd think about:

(1) 1 point should be the right size for 1 decision. At base, spending 1 point should change the simplest failure into the simplest success.

(2) The numbers we handle in our daily lives are well-adapted, generally, so for inspiration:
1, 5, 10, 25, 100
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, j, q, k
1:1, 2:1, 3:1, 4:1 = 1, 2-3, 4-6, 7-10
and always
1, 1, 2, 3, 5, 8, 13, 21...

-Vincent

[Paul Czege]

Hey,

For something like Will points, what I would do is think about how often I want players to be using them during a game session, and then I'd give that number of points. That is, I'd scale everything else (stats, attributes, whatever) so the player is spending a single Will point per usage. The reason I'd do this is because if using Will is supposed to be dramatically interesting or significant, then maybe I want to pace the game so the player is only doing it a couple of times a night. And yes, if players are rolling big die pools and spending five or ten Will points at a time, then maybe they only use Will once or twice a night. But they could also use a couple of Will points at a time, quite a few times in one session, and maybe that would undermine how dramatic or special I want the Will point usage to be. But if I scale everything so standard usage is a single Will point I organically prevent that.

Paul

[Callan S.]

Hi valadil,

It almost sounds like you've built 90% of a machine, then asked "Now, how does this thing work?". How did you build it without knowing what to plug in? Or am I asking bad questions?

[Wordman]

This may be "anticipating the result" a bit, but it seems to me that, because you are using both "opposed bidding" and "unopposed increases", there is a better starting point than just an arbitrary number.

Consider a hypothetical game where only opposed bidding resolution exists. In such a game, the "scale" of the numbers you use doesn't particularly matter, because presumably all players will be working on a the same scale. You could see this in action during the early days of the Hollywood Stock Exchange. On joining that site, your were given a million "dollars" to buy "stock" in movies and movie stars, with the result that a single share typically cost $10,000 dollars or so. Had they instead given every player $1000 to start, the share prices would have been around $10 dollars or so, but trading would have been exactly the same. The only real difference between the two is the resolution of the units. In this example, let's say a single dollar is the smallest unit you can use. In the $1000 buy-in case, you could only go from $10 to $11. In the million dollar buy-in situation, this would be like going from $10,000 to $11,000, but the latter situation allows more subtlety (e.g. going to $10,001 or $10,500 or whatever).

In other words, in a system of opposed bidding, the only thing that really matters is the ratio of the "quantum unit" of your "currency" to the total size of the pools available to the players, and this really only controls how nuanced the bidding can be. In such a game, power of a player relative to another could be manipulated by mucking with their relative currency pool sizes.

Now, consider a game that only uses "currency" to change unopposed tests. The most important thing in this case is how much spending a "quantum unit" effects the result. The ratio of the "quantum unit" of your "currency" to the total size of the pools available to the players matters here as well, but for totally different reasons: here it it controls "how often can tests be changed per game".

Now, you want to use both of these systems at once, and that actually helps you, because you basically have two different forces acting somewhat against each other to find the "ideal scale" for your currency.

What I'm trying (and probably failing) to get at here is that, because you are using both of these ideas, there is probably a distinctly optimal answer for the scale, rather than something arbitrary. You still might have to start just buy guessing, but the combination of the two gives you a pretty decent set of tools to figure out if a guess is too high or too low.

For example, in your system, when a player is bidding, the "rational" question to ask is: "is adding one more quantum unit to this bid worth the effect that quantum unit could have altering an unopposed roll later?" When answering that question starts to get difficult for a player to decide, you're getting close to the "sweet spot" for your currency scale.

[valadil]

Luke,

Encouragement = good.  Thank you for providing it :-)

Lumpley,

Good point about 1 point having its own value.  I wouldn't want to get to the point where players are bidding/chipping in increments of 5 at a time. 

Paul Czege,

I hadn't thought of will in per session terms, but in game time terms.  Resting would recharge a certain amount of all will pools and a meal would let you recharge a stat.  The amount recharged would depend on the stat itself, as would the cap for each pool.  This leaves me wondering how many encounters/challenges players will face between meals and/or rest periods.  Maybe per session is better?

Callan S.,

Well, I think I've built all the individual components, but gluing them together in a consistent manner is what's eluding me.  It's look more and more like instead of tracking down some ideal numbers, I'll have to pick some arbitrarily, playtest with those, and then make adjustments till it all fits together right.

Wordman,

Took a couple read throughs, but I got there.  Or at least I got to a destination that's interesting to me whether or not it's the one you meant to lead me to :-P.  Basically, if the scale is too low bidding will have a dramatic effect, but be too scarce to happen often.  If the system goes too high bidding will always happen and the base values will be meaningless.  I need each unit to be worthwhile and for bidding/spending to always be considered, if not always done.

--

As I already mentioned, there's Will points for each stat (plus some general will that will probably double as XP points).  Those recharge with sleep or food.  At the moment I'm thinking stats will vary from 1-8 (I'd considered as low as 4 and high as 10 previously).  Each pool will cap at 3-5 times the stat, so up to 24-40 points (40 is sounding high, so I'll probably go back down to 3).  Skills will have the same range and they'll also directly limit how much can be spent on any one bid.  So at best you could bid high to double a skill.  Guess it's time to start making a demo character sheet and write up some challenges.

[Wordman]

Or at least I got to a destination that's interesting to me whether or not it's the one you meant to lead me to :-P.

That's good enough for me!

[Paul Czege]

Hi,

What you want to quantify in per session terms is the stuff that's dramatically significant. Perhaps the spending of Will is dramatically significant. But perhaps instead it's the act of refreshing the Will pools that's dramatically significant. Scale the game so the stuff that's dramatically significant is paced to occur with a frequency that satisfies your aesthetics.

(For an example of refreshes that are dramatically significant, see the Dying Earth RPG.)

Also, be careful about having your Will points do double duty as XP. Do you want players to spend them, or do you want them hoarded for advances? It's not easy to create such finely balanced mechanics that one or the other usage isn't more advantageous. And if so, it's not fun to be the GM trying to use in game situation or adversity, or social means to force or encourage players to spend their points the other way, and it's not fun to be the player either.

Paul

[Egonblaidd]

One reason why I'd opt for a per-session refreshment rather than a per-game-day refreshment of Will points is that it is much easier to approximate the number of resolutions there will be per session, while the number of resolutions per game day could vary wildly from day to day.  You might spend three days on the road traveling, then find yourself in a convolution of fights and diplomatic exchanges on one particular day.

It seems like you'd also want one Will point to be significant.  There are two ways you can do this: decrease the number of Will points allotted to each player, or increase the number of resolutions per session.  Think about it like this: if you have 50 resolutions per session and the players can manipulate things so that 30 of these are auto-successes, then you might want to have at least 20 Will points per character.  If you have 100 resolutions per session, and players can get auto-successes in 60 of these, then you'll want at least 40 Will points.  There are three questions you'll have to answer in order to get the numbers right: How many resolutions are there going to be per session, how many of those will be auto-successes, and how many Will points on average will a player have to spend to get a success?  You might also consider how often you want players to fail.  Anyway, to answer those three questions you'll just have to do some playtesting.  You can change the number of resolutions per session by requiring resolutions for more or less actions, which may not be a bad idea if you feel that one aspect of the game is over-represented in resolutions while another is under-represented.  One last question is, how long is a session?  4 hours?  2 hours? Half an hour?  That will certainly affect things, and the length of a regular session should be communicated in the rules, as well as ways to modify the rules for longer or shorter session, if appropriate (such as changing the number of Will points per session).

Something that might be more advisable than having players save Will points for advancement would be to have them SPEND Will points for advancement.  You mentioned that there would be Will points for each attribute.  Perhaps that attribute would increase if, by the end of the session, you had spent two thirds (or even all) of that stat's Will points.  Just food for thought.  This would encourage players to spend Will points instead of hoarding them, but they would still want to save them for use later in the session, so they wouldn't want to blow them at every opportunity.

Anyway, it sounds like an interesting system, though it seems that some of the suspense is taken out of it when you know when you will succeed and when you will fail.  However, it will also make players think harder before doing something, since they (might) have to spend Will points in order to complete it, when they might want those Will points later.

[valadil]

Hmmm.  I think I'll start playtesting single encounters for now and worry about refreshes and advancing later.  I'm wondering if it may make sense to make refreshes always be GM discretion and leave it at that.