Alternate dice rules

Page: 1 2
ID:265572
 
Apr 19 2006, 4:39 pm
I've posted this to my blog, but due to its relevance here I figured it couldn't hurt to bring up on the forums as well. The gist of it is, I'm thinking of revamping Incursion to just use a simple attack roll with a success roll method.

The plan as it stands now:

  • Attacker rolls d12, as many as they wish up to the maximum units they can risk. If the defender has only N units, then only Nd12 are rolled at a time to prevent ridiculously large offensive losses.
  • For standard attacks, 7 or higher is a win, 6 is a draw, 5 or lower is a loss.
  • For attacks against fortifications, 9 or higher is a win, 8 or lower a loss.

    The remaining problem with fleshing this out is the retreat rule. If you allow a retreat with resources, currently your highest defense die tells you how many resources you can move. Under this alternate system I'd have to think up something else. Right now I'm thinking that maybe an attack roll of 4-6 would be a retreat with as many resources (if playing the resource retreat rule), 1-3 an attacker loss.

    And of course, there's still a question of whether this is a good idea. I think it could shorten the game, simplify a lot of issues, and also make it a lot less Risk-like. Is there a serious down side?

    Lummox JR
Apr 20 2006, 10:02 am
Why d12? Wouldn't 2d6 give a more normal distribution (granted you lose the roll of 1)?

Also only an 8.3% chance of draws seems a bit low. Is that intentional.
Apr 20 2006, 11:32 am
In response to Jmurph
Jmurph wrote:
Why d12? Wouldn't 2d6 give a more normal distribution (granted you lose the roll of 1)?

But a normal distribution isn't really what I want. I'm looking for just simple probabilities, which this would screw with a bit. For example, the closest analog to 55% with 2d6 would be to count 7 and up as a win.

Another problem, much bigger, is that when you're rolling N dice, you can't exactly roll 2N dice and expect the result to be about (2d6)×N, because there's no way to pair up the dice. It's also just that many more dice to roll.

Also only an 8.3% chance of draws seems a bit low. Is that intentional.

The draw is really just a side effect of the odds I want to impose. There are no 9-sided or 11-sided dice. Going higher than 12 just to find a multiple of 9 would violate the KISS principle, and even then such dice are uncommon.

Lummox JR
Apr 20 2006, 11:35 am
In response to Lummox JR
Why can't you have a 9 or 11 sided dice if you want that?
Apr 20 2006, 11:52 am
In response to Jmurph
Jmurph wrote:
Why can't you have a 9 or 11 sided dice if you want that?

Well for one reason, I'd like to still give it a board game feel, as in if you could simply use pre-printed maps instead of randomly generated ones, you could play the full game in the real world.

Lummox JR
Apr 20 2006, 12:34 pm
I assume the reason you insist on dice is to keep the board-game feeling. I say, ditch it. Just use a plain probability, and play a sound based on success or defeat instead of the inane "dice roll" sfx. That way, you don't need to worry about analogs at all, and it separates the game further yet from Risk. The "number of dice" thing is bogus, too: I can't think of any reason you would want to roll fewer dice than you had to, unless you were retreating a massive number of resources from a well-Fortified but poorly-defended territory (even then, it's dubious strategy). Dice rolls make the game more boring, slower-paced, and the only positive effect is that it likens the game to the board-game-type experiences players are already familiar with. If you take away the classic d6 dice, you're already taking away the "normal dice" that most players use.
Apr 20 2006, 12:51 pm (Edited on Apr 20 2006, 12:56 pm)
In response to PirateHead
PirateHead wrote:
I assume the reason you insist on dice is to keep the board-game feeling. I say, ditch it. Just use a plain probability, and play a sound based on success or defeat instead of the inane "dice roll" sfx. That way, you don't need to worry about analogs at all, and it separates the game further yet from Risk.

I like the board game feel of that; I don't want to lose that.

Anyway, plain probability in what is essentially still a board game basically hurts player connection to it. Pulling numbers out of a black box and basing all player success or failure on that is a good way to kill interest.

The "number of dice" thing is bogus, too: I can't think of any reason you would want to roll fewer dice than you had to, unless you were retreating a massive number of resources from a well-Fortified but poorly-defended territory (even then, it's dubious strategy). Dice rolls make the game more boring, slower-paced, and the only positive effect is that it likens the game to the board-game-type experiences players are already familiar with. If you take away the classic d6 dice, you're already taking away the "normal dice" that most players use.

True, but there doesn't seem to be a whole fleet of alternatives to using the Risk dice rules, and I'd like to lose those.

Lummox JR
Apr 20 2006, 7:42 pm
I've been playing around with different dice or probability rules lately, myself. My big epiphany was that the simplifications that pen and paper roleplaying uses with conventional d20 rules are not necessary in computer games. I'm not the most mathematically or algorithmically inclined kind of person, and so what I invariably end up doing whenever I need to add a new element of chance to a project is ceaseless experimentation. So, based on the elements of random chance, I don't see how your simplified system would present any major flaws. Going by percentages it looks really good.

One thing that I don't particularly like about this d12 system is that it takes away the defender rolling dice. One thing I’ve enjoyed using on the defensive in Risk - so actually, it might be a good thing that you're making this a difference, to separate Incursion further from Risk - is the option to slow combat down (again, since you're trying to speed things up, this might just be more of an argument in favor of making the switch). By drawing things out as a defender by only rolling one die at a time when things are close, you can sometimes get the attacker to give up. So that’s one element that may be taken away from the defender.

I admit that I haven't played that much Incursion (because its such a long game, ironically), so I don't really know if that kind of nitty gritty psychological stuff is really noticed during a game. But still, it could be considered a down side.
Apr 20 2006, 7:47 pm
In response to Lummox JR

True, but there doesn't seem to be a whole fleet of alternatives to using the Risk dice rules, and I'd like to lose those.

Have you thought about changing what dice rolls mean, entirely? You could subtract the attacker's roll from the defneder's roll, and then take that many pieces off, or something even further from traditional dice rules.
Apr 21 2006, 5:15 am
In response to Lummox JR
Hrm just an idea for thought, but what if more units/troops/whatever gave more rolls and you keep the best? So my 3 troops v. your 5, you get 5 rolls keep highes, I get 3. Maybe you reduce down, so I only get 1 roll and you get 3?

Or how about a factor system. IE if roll exceeds but less than doubles opponent rolls, something happens, if it doubles but does not triples, something better happens, if it triples or more, something yet better happens, etc.?

Just some ideas.
Apr 21 2006, 8:10 am
In response to Lummox JR
Lummox JR wrote:
Jmurph wrote:
Why can't you have a 9 or 11 sided dice if you want that?

Well for one reason, I'd like to still give it a board game feel...

there are 9-sided dice available in nearly any hobby/game store - AD&D has used them for many years. d4, d6, d8, d9, d10, d12, d20, and d100 are all fairly common die now-a-days.
Apr 21 2006, 9:10 am
In response to digitalmouse
I've played ADnD for years and I've never seen, or used, or heard of, or heard reference to a d9 die. I've bought dozens upon doznes of dice, sometime in packages of 15 or 20 at a time. In such a dice set, wouldn't all the common dice be included? Sure you're not a little off base with the d9 bit?

Rolling non-standard values really isn't too hard. For example, to roll a d3 you can roll a d6 and have 4, 5, and 6 be 1, 2 and 3 respectively. To roll a d9, just roll a d10 and reroll an die that comes up 10. No need to make special die for those things.
Apr 21 2006, 9:50 am
In response to PirateHead
actually there are examples of d9 on the net (even a paper one listed here: http://elothtes.pbwiki.com/ ELotH:%20TES%20-%20Pen%20and%20Paper%20RPG%20Dice)

..and d9 is mentioned here in several forms as well: http://www.dicecollector.com/diceinfo_how_many_shapes.html


but for some reason i was confuzzling d9 with d10. i blame elation this time.

:p
Apr 21 2006, 10:16 am
In response to digitalmouse
That "how many shapes" page is really neat. It would be neat to run a custom dice company that could make dice to fit any number range (11 to 26, 7 to 23 odds, etc) for unconventional gamers.
Apr 22 2006, 8:40 am
In response to Jmurph
Jmurph wrote:
Hrm just an idea for thought, but what if more units/troops/whatever gave more rolls and you keep the best? So my 3 troops v. your 5, you get 5 rolls keep highes, I get 3. Maybe you reduce down, so I only get 1 roll and you get 3?

I thought of a system like that early on, but it's infeasible. If you go by comparing highest dice, the attacker advantage becomes overwhelming, even though in the Risk system, part of the advantage is also in the ability to absorb losses. So this adds in overkill.

Plus even with large rolls vs. large rolls you have a problem: An inordinate number of the rolls will be tied. If ties are not broken, you have way too many draws for all those dice thrown. If they break to the attacker, the attacker's all but invincible; if they break to the defender, you have perpetual stalemate.

Or how about a factor system. IE if roll exceeds but less than doubles opponent rolls, something happens, if it doubles but does not triples, something better happens, if it triples or more, something yet better happens, etc.?

Not sure I follow, but I don't think it'd be workable.

Lummox JR
Apr 22 2006, 8:49 am
In response to Gathin
Gathin wrote:
I've been playing around with different dice or probability rules lately, myself. My big epiphany was that the simplifications that pen and paper roleplaying uses with conventional d20 rules are not necessary in computer games. I'm not the most mathematically or algorithmically inclined kind of person, and so what I invariably end up doing whenever I need to add a new element of chance to a project is ceaseless experimentation. So, based on the elements of random chance, I don't see how your simplified system would present any major flaws. Going by percentages it looks really good.

The important thing about simplification, though, is that it adds an element of good design: the KISS principle. Players prefer games where the rules and (to some extent) the odds are easily understood.

One thing that I don't particularly like about this d12 system is that it takes away the defender rolling dice. One thing I’ve enjoyed using on the defensive in Risk - so actually, it might be a good thing that you're making this a difference, to separate Incursion further from Risk - is the option to slow combat down (again, since you're trying to speed things up, this might just be more of an argument in favor of making the switch). By drawing things out as a defender by only rolling one die at a time when things are close, you can sometimes get the attacker to give up. So that’s one element that may be taken away from the defender.

Yes, I have considered that as a down side. Rolling defense gives you a stake in the territory beyond just controlling your borders.

I admit that I haven't played that much Incursion (because its such a long game, ironically), so I don't really know if that kind of nitty gritty psychological stuff is really noticed during a game. But still, it could be considered a down side.

Agreed. Then again, it's been used to great effect in other strategy games. What's more, many that use the success roll mechanic do so because of expediency. It makes the game go faster.

Lummox JR
Apr 24 2006, 5:16 pm
Aaiko had a good idea that puts a twist on my d12 concept and allows for the psychological factor of no defender roll to go away: Make the defender do all the rolling. The idea is, the attacker says "I'm throwing N dice against you" and the defender rolls to stop it. So basically, my probabilities would be reversed:

Defender rolls 8-12: Attacker loses 1
Defender rolls 1-6: Defender loses 1
Defender rolls 5-12 (w/fort): Attacker loses 1
Defender rolls 1-4 (w/fort): Defender loses 1
-- loses 1 fort if 2+ dice rolled and 4 was the highest roll

The retreat rule is still a puzzler. However this effectively ends the psychological problem. The roll is still automatic, requiring no defender intervention except retreat decisions, but it's their dice making the roll instead of the attacker's.

Any thoughts on this modified system?

Lummox JR
May 8 2006, 6:57 pm
This could really use some more discussion, since I've hit a snag and need to talk out mechanics some more. Here's what I've got so far:

  • Attacker chooses up to 6 units, or at most as many as they can lose. The defender rolls that many 12-sided dice. There may be more dice than defending units.
  • Each defender die over 7 kills an attacking unit; under 7 kills a defending unit. 7 is a draw.
  • Fortifications add +2 to each die.

    Now the problem with this is, while it fixes the odds about where I'd like them for a large skirmish, they're the very same odds as a small skirmish. A large skirmish however includes the very real possibility of overkill, the attacker taking too many losses against a single unit for example, whereas a small one doles those out one roll at a time and if lucky can win without incurring extra losses. In other words, the new mechanic encourages attackers to send in just 1 unit at a time except during large-scale battles.

    Limiting the number of dice thrown to the number of defensive units solves this problem, but not by a lot. There is still no advantage, as there is in Risk, to rolling more dice then.

    What I'd really like would be to modify the system based on attacker success odds of around 40-45% for 1-unit skirmishes (less if outnumbered), and around 55% for 6 units. Risk achieves this by virtue of its highest die comparison, which means in a nutshell that more dice = higher rolls--not to mention it also limits the defender to just 2 dice, whereas an attacker can roll 3.

    I considered a kind of compromise between this d12 system and Risk, by having a dual roll (attacker rolls Nd12, defender Nd12) comparing high dice and giving ties to the attacker. But again smaller combat situations are messed up, and large-scale combat could suffer too, since I think this actually drives up the success odds for an attack. There's also the question of how to fill in for dice the defender doesn't have to roll. And what do you do if the defender has only 1 unit? The odds say that unless you restrict the rolls to just 1 defender die (in which case attacker odds go way up), it's still better then to risk 1 unit at a time rather than throw in 2 and potentially take an unnecessary loss.

    Here's a site with a table of Risk odds: http://www.plainsboro.com/~lemke/risk/
    If you cut the "both win one" percentage in half and add it to "attacker wins both", for cases where 2 losses are possible, that gives you the approximate attacker success rate. I'd like to achieve similar results. Notice the attacker only has the advantage when rolling more dice than the defender, but that 3-on-2 is the skirmish of choice for large warring countries.

    Any thoughts on this?

    Lummox JR
May 9 2006, 9:16 am
In response to Lummox JR
Last night after making the last post, I played around with a new idea. I can dip the odds lower depending on the defender's strength.

So here's my new idea. The attacker runs a skirmish of S units (up to 6), which is how many they will risk. The defender rolls that many dice, plus one more if they have more units than are in the skirmish. If they roll the extra die, they throw out the lowest roll.

The upshot of this is that the odds of a small skirmish succeeding against a well-defended position rather suck. Large-scale or evenly matched confrontaitons tip to the attacker at the regular 6/11 odds, but in any case where the skirmish is outnumbered, the defender has the advantage. In fact by running a number of small skirmishes, your odds go down vs. risking the same number all at once. This is exactly the effect I was looking for.

This will, however, slightly screw with fortifications. I believe the solution will be for the increase per die to be +1, or else to simply eliminate the draw and reduce winning odds per die to a coin flip. If I go the latter route, though, the odds of a fortification falling to 1 die are 1 in 4, so I'd have to require at least a 2-die roll to kill one (like I do now), but that's still too high. So I'll probably stick to the +1 advantage. Either way, I think I'll still require 2 dice rolled to kill a fortification.

Lummox JR
May 9 2006, 12:38 pm
In response to Lummox JR
Dear, troubled Lummox JR,

I still think you're banging your head on the cement by sticking to dice rolls and psychology and stuff.

Let me try to put myself into your mind set: I want to be able to finely manipulate the odds and perhaps even adjudicate the odds depending on situational conditions, but I want the players to feel the psychological effects associated with rolling dice in order to determine their success or failure and the excitement of high rolls versus the horror of low rolls, and the adrenaline rush when a mediocre roll is met by an even lower roll, etc.

Call me a cheating bastard, but here's where my thinking goes from there: instead of actually coming up with a dice system that compensates all the intricacies of what I want to do, I come up with a pure-probability engine that can handle all the different situations and odds that I want to use, then I come up with a system that fakes dice rolls based on the already-determined results. That way, I get the best of both worlds: I can pull the finest strings and make adjudication where needed, yet give the players the feeling that their dice-rolling matters.

They aren't real dice, and the player's can't see them fall. They'll never know. You can even delete this post so that nobody will ever catch wind. You're not above a little finagling, are you? =D
Page: 1 2