I don't begin serious work on a game until I have a clear picture in my mind of what I want it to be. Then I keep working on it until it resembles that picture; that's how I know it's done. This means however that there is often a lengthy period between concluding primary research and beginning to actually put the thing together while I wait for my brain to make connections and bring the thing into focus. Sometimes it's weeks, sometimes months, sometimes years.
Sometimes it's a slow, gradual, and subtle process, this idea here tying into that one there, until one day it dawns on me that the thing is ready, and maybe has been ready for quite some time. Other times, nothing seems to come together. I have ideas, but they're all frustratingly distinct and separate from one another, and I can't seem to tie one thing to another. In these cases it's often that there is a piece that is missing, some wonderful thing that, once revealed, connects everything else, makes everything click. Like that rug the guy had, that tied the room together.
I don't know what that thing is, of course - if I did, it'd be easy enough to solve the problem - but I can often feel its absence rather acutely. I had this problem with Westphalia. I had fragments of ideas that I just couldn't find a way to flesh out. This thing was cool, and that thing was neat, but I couldn't figure out what these cool, neat things had to do with each other. I had a bunch of ideas but I didn't have the idea, the thing that made it a game. It occurred to me that the thing I was missing was an answer to the core problem of any negotiation game, which is, how do you get people to negotiate with each other? "If I give you x, you're going to win, so I'm not giving you x."
Then I played Cole Wehrle's John Company, and I rather fell in love with the promise cubes. These are effectively IOUs you can give to other players in exchange for something now, and you can demand their return through certain specified actions benefiting the player who has one over you. Having another player's cube is good for me because (a) if the game ends with it in my possession, it damages their position, and (b) the threat of that loss in victory points is sufficient that you'll be compelled to do something I want to get it back. This sort of fungible commodity goes a long way toward incentivizing semi-cooperative behaviors, toward making folks amenable to negotiating.
Hey, maybe I can steal this, I thought, and with Cole's blessing I plopped a modified version of it into my mental framework for Westphalia. This "leverage cube" worked somewhat differently - it could be taken by force as a result of military operations - but it was the missing piece. Suddenly all the various commodities were tied together by the leverage mechanism, and players holding leverage had ways to compel intransigent players to give an inch or two. Very quickly the picture of the game came into sharp focus.
Once these things snapped into view, other aspects of the game gathered at the periphery, such as the roles of the individual factions and, most importantly, the idea that multiple factions could share in the victory. Gaining victory would require some measure of cooperation and trade; it was more-or-less mathematically impossible for any one player to achieve victory without help from somebody. Of course that brought up the question: well, what if all the players just work together so that everyone meets their victory conditions? And the immediate answer of course is that if that happened, then only one of those players would win via a scoring round - that could only happen if five other players were willing to work their butts off achieving their own victory conditions only to let someone else walk away with it, and so just as it was in their self-interest to help one another, it was also in their self-interest to cut someone out of the deal. I took it one step further by making the victory conditions for two players - Austria and France - each dependent on achieving victory for two others - Bavaria and Spain for Austria, and Sweden and the Dutch Republic for France.
Here's the thing, though: those leverage cubes? They didn't work! I didn't know that at first. Certainly something wasn't working, something was off, but I couldn't quite place my finger on it. What happened is that during the initial tests Spain's position was far too weak, and after bouncing some ideas off of Mary, I introduced a new commodity, Prestige, of which Spain had quite a lot. Historically it was Spain's prestige and reputation that enabled it to continue getting the loans that hid its insolvency, so this commodity had the bonus of reducing the rate at which players accumulated Debt. Now, when players won a battle, they could demand either a Leverage or a Prestige from their opponent. Every player went for Prestige; no one bothered with the Leverage.
And that's when I realized that the Leverage cubes were completely extraneous and needed to be junked. Their primary purpose of course had been to facilitate negotiations. But the game's web of interconnected victory conditions, and the ability of some (but not all) players to share the win filled that niche. That was the real center, the actual thing that had been missing, and after jettisoning the distraction that was the Leverage cubes, I was able to realign my little cardboard solar system so that everything orbited around that instead.
Of course, I don't know if I would have come up with that aspect of the game if not for the Leverage cubes. Though it would ultimately be discarded, that mechanism was a way in. Which is all to say that while I don't start working on a game until I think I know what it is, sometimes it turns out to be something else entirely.