An avenue I was going down was doing a Silverton style Express dice game.  The idea was to take different combinations of dice of various length and allow players to pencil and paper their way down through undefined mountains to build mines and bring back raw materials to the cities.  The notion of paying for things with dice, having mines and junctions be certain combinations and using straights for engines in the game (ending in boxcars - or 6's) with lower elevation track being the lower dice rolled working well.  The idea was to have eight dice, no rerolls and to just wing it.

I still may end up doing it.  It needs a lot of refinement as it is exceedingly luck dependent and tactical with very little strategizing along the way.  Much more refinement.

Anyhow, I thought back on my other games I've worked on with their layers of cards, etc.  I picked up another set of dice and begin rolling them as I have been, almost out of nervousness, for the past couple days.  It struck me that dice are like cards in sets of six at the very least.  Unlike cards, they show up whenever.  You can get a bunch of 1's, but there's only one 9 of clubs.  That's a lot of randomness.  I like a little, particularly for solo games, but dice are so unmanageable.

I then tinkered with the idea for a game to use two sets of standard dice... two of them red and three of them white.  The red dice would be used to indicate a random event and would be a look up.  The other dice could be used in combinations of 1, 2, 3 and 5 dice which would allow for multiple combinations.  In other words, like a "hand" - almost like how Pocket Civ has different values for each era.  It's almost like you could generate multiple hands with special bonuses off of each set of dice with any number of actions.  Players could take turns picking dice out to fit combinations they like.  1's and 6's would be valuable, but 3's and 4's would do heavy work.  Scaling up, unique combinations such as all 1's and 6's could generate huge events if prerequisite choices had been made before.

The hard part, would be charting, so you have to pick patterns that people know.  Margin of Error uses a couple charts at different places and some easy to memorize and look up ones for other effects.  Other games, particularly war games, use charts and bonuses.  I typically hate charts (yeah, I know... Bindle Rails is essentially a chart filling exercise, but still...).

So does this lead to a Silverton game?  I don't think so.  Not so much.  I see a Silverton Express game needing to be much simpler.  There's got to be a way to do it.

Themes that tend to lead to this involve games of building - which I love whether they are board games, computer games or even programming.  Civilization games (and yes, Roll Through The Ages) have been done like this, but there are other themes to explore.  For example, The Game of Life: Card Game would actually probably be a good candidate for this style of game.  I'd love to do a city building dice game - and if anyone knows of one that builds things kinda like SimCity using dice and a modern theme, let me know.  Heck, even Bindle Rails Express could probably be done to be a hell of a lot more interactive than it is now - and with simpler bits.

These ideas will be flushed out in the next week as I decide on how I want to go about it and whether it ultimately holds up to being "fun."  I have high hopes.  I'm sure other designers have been here and could point me to success stories and published examples.  It's an avenue that I never really explored because I never really used to like dice.  Maybe I've been using them wrong all these years.