Once an idea has passed basic examination and made it onto the list it is usually allowed to settle for a few weeks at least but most for far longer. Every few months I select one of the ideas on the list and start working it up into a basic prototype. Step one of this process for me is always roughing out the general rules on paper. This first version of the rules always needs re-writing before the prototype is finished but it has a few purposes. Firstly it allows me to see if the idea fragment deserves to be built into a full game, some turn out to be bad ideas and get dumped at this point and some are only a section of a game not yet ready to be fully realised. Secondly, at this point I can judge if a game will be best served by being a set of rules for miniatures, a print and play or a full game release project. Those that need to be and are ready to be made into a full game are then made into a prototype.
For me personally a prototype is a working tool and it is constructed from bits of paper and card written on with pen and stickers. This is for two reasons, firstly I often don't know what the game will be at this point and I don't wish to commit to a design before I know for certain. Secondly, those cards will be written and re-written over time so designing and printing a high quality version at this point would be a waste of time and money.
The process of creating a prototype is a highly involved one and for me forms the first stage of a sort of pre-alpha testing. Even with a relativity simple game I am not capable of holding every card, rule and interaction in my head at once so usually, having written the rules first when writing the cards I will find a contradiction with a rule or a redefinition to be required or I will need to introduce a new mechanic or economy with its accompanying rules. As such most full project prototypes will go through three or four total re-writes where written elements get tested against each other just in my head, but it usually catches a lot of issues. The prototype then goes on a shelf to settle again.
The next stage is to select a prototype for alpha and beta playtesting. Playtesting of any form is both vital and extremely resource intensive so the selection is extremely important. I choose a handful of projects that I think have a quality and spark and begin by judging their components and estimating their final manufacturing price and therefore the price I'll need to ask from my target audience. £10 seems a reasonable price for a first time designer, for that I can manage about 80 cards and a few punch boards.
My next major criteria is based on how difficult a game will be to playtest. Playtesting is not only important but it requires you to find a group of splendid, patient, intelligent people, ideally strangers if you can find them who are willing to work for free. As such a game that is quick to explain and has the lowest possible minimum player count is significantly easier to develop. I have what may be an excellent game sitting on my shelf purely because it has a three person minimum player count.
The final stage of selection for development does tend to depend somewhat on fortune. In the case of SSO it had everything I needed, low unit cost, broad player count, solid theme and both co-op and VS play styles but the answer of how to work its rank token based "guilt" system, oxygen based turn limit and narrative structure all became clear just as I was looking for a first launch design. Similarly, for Moonflight it had a reasonable unit price, could be expanded well and had a good, clear, strong concept. Despite that I was contemplating pitching it around to publishers at UKGE when I realised I felt too strongly about it to not have it as my second major project.
So, prototype selected and created, its time to head into innital testing.