I post this separate from the thread where we were discussing EF because this is more of a practical treatment of it.

Basically the effect-first GM technique is about introducing events without creating an independent NPC entity for it.

For example the GM might rule that a hand comes up out of the magic pool to drag a character down, without having stats or having made up a "magic pool" in terms of magic power and so on.

Usually the problem comes when players try to interact with an event the GM hasn't yet made up stats for, because it takes a lot of effort for the GM to reverse-engineer the effect into a mechanics based entity.

What simplifies matters is when reverse-engineering effects into in-mechanics entities is easy.

Here's an example of such a system. It's really simple but it's only to server as an example illustrating how a easily reverse-engineered system would look like. It also incidentally involves some maths which makes it unsuitable for actual play, but this could probably be fixed with some clever mechanics.

The system

The characters all have 1 (one) single stat: Power. This stat runs from 1 and up. Maybe beginning characters start at 10 or so.

Whenever something threatens the characters or they want to overcome some difficulty, they roll against the Threat of the task or the opposing force. Threat works just like Power, it also runs from 1 and up. There is no difference except that for the characters it's called Power and for the GM's stuff it's called Threat.

Overcoming a Threat

What happens when "rolling against a threat" is that the player picks up a D6, rolls it and adds his/her Power to it. The GM does the same. The highest wins.

Pooling Power

The players can work together letting their characters together work against a single threat. The total character power is given by summing the square of all the characters' Power and taking the square root of that. Like this:
Actual Power=Sqrt(Power1*Power1+Power2*Power2+...)

The same rule goes for the GM pooling threats and you still only roll 1 die on the player side and one die on the GM side.

Example:
Bobbo and Rok wants to take out the Golden Legged Crane perched on the mountain-top. Bobbo has Power 15 and Rok has Power 17. The Golden Legged Crane has Power 25.
Together Rok and Bobbo has a power of sqrt(15*15+17*17)=22.67...

Rok's player (or Bobbo's - they have to decide who gets to roll) rolls 1D6 and adds that to 23. They roll 4 for a total of 27.
The GM rolls 1D6 and adds 25 with a roll of 4 too, for a total of 29.

Too bad for Rok and Bobbo, they failed to take out the Crane.

Very bare-bones. But you see the point. The GM could easily create any effect and assign a Threat rating. No need to define everything about the Golden Legged Crane and so on.

Cool use of the Pythagorean Theorem.

And I think you have a system that would work in general terms.

The only concern I have is that it seems totally dissociated from the rest of the system. What happens if one player is doing combat, and another is doing magic? Who goes first (and gets to kill the baddies)? If the player doing combat wounds the opponent, does that make it easier to dispatch with magic? Or not?

IOW, in having two systems, now, you have to determine how they interface. This is why I have all along been suggesting that you use something that incorporates the rules from your main system, because then "interface" becomes obvious.

Mike

I'm sorry if I wasn't clear. This was just an example of a system in which EF wouldn't be a problem. It has a lot of other problem on the player side in terms of detail.

What I wanted to do was to present something which could be considered a purely EF facilitating game-mechanic as a starting point to compare with conventional game mechanics to discuss how one might use the benefits of both.

From the player's point of view the mechanics is probably unsatisfactory, as pointed out, since there is very little to give any differentiation between the characters.

However I still hope something useful is captured in the example.

Note how it's extremely simple for the GM to weigh the odds. The GM can also calculate if the characters can win if they cooperate, if they can do it single-handedly or maybe not at all. In any case the GM has total control over the situation.

However, to me it's not immediately obvious how to combine the simple reverse-engineering abilities of a system such as the above and a comparatively detailed one.

Mike wrote: IOW, in having two systems, now, you have to determine how they interface. This is why I have all along been suggesting that you use something that incorporates the rules from your main system, because then "interface" becomes obvious.

I'm not sure I understand how this really supports your idea of effect first. The only difference I see is that the math is simpler, but in almost any game you could set the difficulty level to an impossible level and tell the person its just impossible.

I mean, let's look at the basic issue here. What you're talking about is having mechanics to model what the GM has in mind accurately.

Let's take the classic, accidental kill of the major villian in the first 5 minutes of the game. This is a result of it even being possible to occur. If you set the odds at a astoundingly low 1% chance of the PC's killing the viliian, there's still a 1% chance, period.

Regardless of the fact that most of the time, doing this will give you the results you want, eventually your luck will run out and the 1% will come up. Is this a matter of the GM not knowing the odds? No.

What this is about, is the GM desires a certain result, yet wants to give the illusion of fairness to the players. As in your example above, the players may not be happy that the magic hand can drag them into the lake without giving them a chance to roll. But to be honest, if what you're looking for is for the players to get dragged into the lake no matter what, why are you even giving them the dice?

What you seem to be focusing on, is a math based rule to back up GM fiat. If you're going to do that, you might as well make GM fiat a rule and install something like a reward for players everytime the GM cheats, such as handing out plot points, Deadlands chips, whatever, everytime the players rights get stepped on by fiat.

I mean, if you're going to make an illusionist game, that's truly illusionist, you will have two sets of rules, the ones the players think are in effect, and the ones the GM actually runs.

Chris

Bankuei wrote: I'm not sure I understand how this really supports your idea of effect first. The only difference I see is that the math is simpler, but in almost any game you could set the difficulty level to an impossible level and tell the person its just impossible.

I mean, let's look at the basic issue here. What you're talking about is having mechanics to model what the GM has in mind accurately.

Let's take the classic, accidental kill of the major villian in the first 5 minutes of the game. This is a result of it even being possible to occur. If you set the odds at a astoundingly low 1% chance of the PC's killing the viliian, there's still a 1% chance, period.

This is just an example (oh, yeah, there it is, in the text above, huh)? Um, yes, this is just like every other system that exists that facilitates this sort of thing. Given it's simplicity, it's very much like HW, in fact (in that game, instead of Power, you have some ability score like Affinity for Death, or Grimoire of The Bright Force).

I'm not sure why we need such an example when many exist already. Why did you bother posting it?

Mike

Because I still don't have HW Mike! :) So I couldn't point at HW and say: hey here is something that facilitates EF! Because I don't have HW to point at. If HW facilitates it, great. It's just that I don't have any references to good systems so I made up something barebones on the spot, and to be honest, also to get myself a little aligned with these new mechanics.

It was brought up in the earlier thread that "there might not be a way to support EF" so I just wanted to present something that did, just to have something to point to, but I'm repeating myself.

Bankuei: I don't get how you interpretate that the Golden Legged Crane is a major villain that shouldn't be killed. If the Threat given to it is the one mentioned it's obvious that the GM thinks both options (it being defeated or not) are acceptable.

John: That idea is actually pretty interesting :) Good thinking.

Pale Fire wrote: John: That idea is actually pretty interesting :) Good thinking.

PF, sorry to yoink a quote from your other thread, but I really want to make sure I'm understanding you correctly here:

EF is about creating the desired effect first. In terms of opposition maybe I want the players to have a medium challenge defeating the badguys. Or maybe the lock should be a low challenge to the them. Or the chasm should be impossible to jump.

However, setting this difficulty level can be very hard in many if not all games, at least until plenty of experience of the system "in play" is acquired. That frequently leads to under- and overestimations of the player's odds.

Bankuei wrote: So the idea with these mechanics is to create a means for the GM to be able to define the possible range of results without the system violating them, correct?

Say I'm playing the fight with GLC, and I want to kill him, I want him dead. Do you, as the GM, tell me beforehand, that the best result I can get is to defeat him? Or do I just have to deal with the results?

So, in other words, Fortune in the Middle is OK for magic, but not for combat.

Mike

Ok, I see where you're coming from, although it seems to make more sense to focus on establishing a concrete set of rules for the risk/result factor.

I know you haven't gotten a copy of Hero Wars but you can download their quickplay rules for free, which cover pretty much everything short of magic. That system does a good job of balancing a shifting risk/result factor through the use of a gambling mechanic.

In this case, the point of making the mechanics mathmatically simple for the GM has some value, but on the other hand, making the possible range of results transparent seems to hold more value.

I don't think the major issue of EF has to do so much with the difficulty levels, as much as the open range of results. Again, coming back to combat(one of the most fudged things), if there's a 1% chance that the PC's will be killed, its not that the % is bad, it's the fact that the result is even possible. The PC's being the ones who will always be in the limelight and always in the conflict will be subjected to that 1% each combat which will eventually result in their loss. If, instead, we know that capture is the worst thing, then it makes more sense to adjust the range of results rather than simply the probabilities.

Chris

I feel your conclusions are right on, but that doesn't really get us closer to a solution I fear.

In HW we have a unifying mechanic which kind of guarantees that no outcome can suddenly mess up the balance it creates.

With a little more detailed system, like your average BRP, you suddenly have to account for possible detailed advantages coming from particular details in the situation.

To take a concrete example: With a lucky hit the monster pushes you off the cliff and you die most untimely.

There are no way to completely avoid those things, but some basic predictability is a good start.

I'm thinking of a hypothetical system in which you could rank any ability into an objective (HW-like) value. For example if you have Ygg you could multiply hitpoints, weapon-damage, toughness, and the square of your weapon skill rate your overall effectiveness in combat (very roughly)

To further simplify, one could rate partial abilities as Low, Medium, High, Very High and so on, and have simple addition rules how they effect the total rating.

For example I look at my Low hitpoints, High weapon damage, Medium toughness, High Weapon skill and see that according to the table my Combat Rating is "High"

That's assuming it's mathematically possible to create tables like that without too big errors (which in turn depends on how they interrelate with each other)

IF we had a working system like that, the system would work like a Champions-light.

To use a champions analogy:

The characters are 300 point, 250 point and 350 points respectively. A total of 900 points. If the GM sends out 5 100 point worth of badguys it's obvious that they characters have little trouble.

The GM can also think: I want this place to be a 500 point challenge, meaning I can make powers for 500 points of worth, because that's how bad this forest is.

It doesn't seem too easy though.