Use of Random Elements in a Game

[Barkeep49]

Over in Alan's "Help me pick a theme" thread there's started to be a discussion of random.org and its use in games. As I posted there I've been trying to move heavily away from Random.org, but I don't know that this is a "must" for a good game. I think a good game is about giving players meaningful choices. path's game obviously relied heavily on random.org and I don't think that was a bad thing. I do think, however, that the villagers ended up winning less because of the unique game play mechanics and more because of traditional WW strategy (using a seer to form a CoT, post analysis, etc). I think that this had less to do with the use of random outcomes and more to do with the fact that there was uncertainty over the random element. Basically all of the scans were taken as useless (except from true seers) because people didn't know the chances that they could trust the random outcome. I think we'd have seen a very different game if people knew what their scan accuracy was (though this would have also made the game FAR more mathy and put the wolves in a more difficult spot).

Others in that Alan's thread have also said they're trying to move away from random.org. Why? A lot of people have mentioned Apocalypse Awakens in one of the Golden Wolves threads. That game used a fair amount of random.org. What made that game successful despite (or perhaps because of) its random elements?

[Alan T]

Originally Posted by Barkeep49

Over in Alan's "Help me pick a theme" thread there's started to be a discussion of random.org and its use in games. As I posted there I've been trying to move heavily away from Random.org, but I don't know that this is a "must" for a good game. I think a good game is about giving players meaningful choices. path's game obviously relied heavily on random.org and I don't think that was a bad thing. I do think, however, that the villagers ended up winning less because of the unique game play mechanics and more because of traditional WW strategy (using a seer to form a CoT, post analysis, etc). I think that this had less to do with the use of random outcomes and more to do with the fact that there was uncertainty over the random element. Basically all of the scans were taken as useless (except from true seers) because people didn't know the chances that they could trust the random outcome. I think we'd have seen a very different game if people knew what their scan accuracy was (though this would have also made the game FAR more mathy and put the wolves in a more difficult spot).

Others in that Alan's thread have also said they're trying to move away from random.org. Why? A lot of people have mentioned Apocalypse Awakens in one of the Golden Wolves threads. That game used a fair amount of random.org. What made that game successful despite (or perhaps because of) its random elements?

My game "Haunting of Sorcerery Hill" I felt wasn't as good as it could have been because random.org just blew up in my face and some of the improbable results just caused issues with the story line. I don't think it was a bad game per say, but it wasn't as good as some of my other games (Clash of the Gods, Tombstone, Big City, etc).. I pretty much felt that was because of randomness being something I couldn't plan appropriately for as a GM.

[RendeR]

My fairy tale land game had a lot of random elements, but even so it required the players to make many choices regarding how to use the things they had and found.

I think a good game really comes from the GM more than anything giving the players the right tools to make the choices and move the game along.

[st.cronin]

Don't like it, when I run a game, I use random.org only for assigning roles.

[Passacaglia]

I think it depends on the game. The GM can best decide if it fits the theme of the game, or the design of it. In general, I don't like it, but on the other hand, I've used it in both my games (for handing out cards in Clue, and for deciding the outcome of games in the BCS). Of course, both my games sucked!

[st.cronin]

Path used it heavily in his last game, with the seers, but I thought having the fool and teacher's pet made the randomness less important.

[Barkeep49]

So people have been saying in general that they don't like it. But why? We've had successful games with a good deal of randomness (e.g. Hoops/Tyrith Marvel game. So it doesn't seem like having randomness precludes a good game. Yet many of us are expressing how we don't like to include it. Why? I have my reasons but I'd love to ehar others. And more importantly how can random elements be successfully used?

[st.cronin]

Originally Posted by Barkeep49

So people have been saying in general that they don't like it. But why? We've had successful games with a good deal of randomness (e.g. Hoops/Tyrith Marvel game. So it doesn't seem like having randomness precludes a good game. Yet many of us are expressing how we don't like to include it. Why? I have my reasons but I'd love to ehar others. And more importantly how can random elements be successfully used?

I didn't play in either of the Marvel games, except as a useless late game fill-in in one of them. I did enjoy path's game, but as I said above, the important roles were the NON-random ones. My objection to random results is it can render analysis pointless. If there's a reason for everything that happened, then analysis is meaningful.

[hoopsguy]

Just as a comment, I don't believe there was a good deal of randomness in my Marvel game. There was a lot of math, but very little randomness (Captain America is only person who had a random element in his powers). One huge random element was the selection of "The Twelve". But outside of this, not so much.

- People got to spend their energy points to try and deplete the health points of others.
- Roles were publicly known since players posted in the thread who they wanted to play (Schmidty = Hulk) so you could make reasonable assumptions what kind of health targets might have based on your own stats.
- Actions at night were clearly defined in terms of order of action. Heck, only one role was random (Apocalypse) as the head bad guy picked his team.

I'm not looking to threadjack the conversation here, but having now seen two references to the Marvel game as "random" I felt like that perception should be adjusted. In fact, looking back at the game the biggest change I would have made is to tweak the results of the random "Twelve" to have a little better mix of strategic vs non-strategic targets for the bad guys.

Back on point - I generally like to eliminate randomness from the games where possible. However, I don't think that is the most critical element to designing a great game as a moderator or enjoying a game as a player. If you are going to introduce randomness, then I like the idea of giving players choices.

Example - take this type of bodyguard role.
You are the bodyguard! Due to your combination of coolness and toughness you have the option of:
- guard every night with 50% chance of blocking and 0% chance of identifying foe on successful block
- guard every other night with 70% chance of blocking and 30% chance of identifying foe on successful block
- guard every three nights with a 90% chance of blocking an 50% chance of identifying foe

I think this is a lot more interesting than the standard bodyguard role because you are giving the player choices on how to use their power rather than "guard someone every night, random chance of stuff happening that you don't understand". Random isn't inherently bad, but people should understand their odds and be able to interpret them.