That's me and my other pastime. I perform as John Lennon in several Beatles tribute acts.
Confident that the unit card was the solution to the elusive playabilty-realism conundrum, I plowed ahead. Dirk and I hammered out the basic game in a weekend, and for the next year and half I ran playtest after playtest making minor tweaks and incremental improvements with each revision. Pocket Armies was well received everywhere we ran a demo. Many were extremely impressed with the balance of playability and realism I had created, and that turned into steady pre-orders. The system was intuitive for them and "clicked" right away. With only a brief introduction to the core concepts they were off to the races.
But it wasn't universally praised like I had hoped. I could usually tell when I was losing a potential buyer. Some seemed to struggle with it no matter how straight forward I thought it was. Something was slightly amiss. Most sessions with veteran players played like speed ASL. We'd finish an entire scenario with a dozen units per side in under 90 minutes. But several sessions with newbies would take 3 hours and would fall apart before we even got halfway through. How could a game with an 8-page rulebook take sooo long? I had to figure out where the cognitive disconnect was...
I would query playtesters after every session, particularly if they seemed less than enthusiastic about the game. But nobody was able to pinpoint the issue. The problem was many people liked it, but they didn't love it. It's easy to explain to someone why you dislike a game, but much harder to explain why you only like a game but not love it. That's why we so often describe that great games have an "intangible" that separates it from the rest. Pocket Armies didn't have that "intangible".
Finally at a playtest at Origins, someone nailed it. They didn't articulate it precisely, but when I heard the comment in passing, I immediately realized it was the system's fatal flaw. The constant back and forth between the unit cards and game map, destroyed the visceral experience. People who played quickly and fluidly were very efficient at keeping the cards and units organized either in their mind or front of them. Those who couldn't, struggled even to figure out which unit or miniature corresponded to which card. The problem was compounded further if they couldn't visually identify WWII AFVs. As the number of units under their control increased, the problem grew exponentially. If they came strictly from the world of collectible miniatures gaming, the CRTs on the unit cards were overwhelming. Hardly the crossover game that would unite grogs and casual gamers as I had hoped...
On the excruciating drive home from Origins with my friend and co-designer Dirk Knemeyer, we brainstormed possible fixes. We talked about simplifying the cards, developing a better way of linking them to units on the map, but it was futile. I had invested hundreds of hours and I was certain the game was as good as the game mechanics allowed it to be. Pocket Armies suddenly felt like an abject failure and an evolutionary dead end. As basically an admission of defeat, I exclaimed "Unless we restrict it to a skirmish level game, the unit cards have got to go." Dirk replied (as is so typical of him) "DAMN it Tan. That's the answer!" I told him not to be ridiculous. He said "There has to be a way to move all those stats off the cards to somewhere else!" And that is Dirk's brilliance. He sees solutions when others see impossibility, even if he is unsure exactly how to get there. I'm a skeptic by nature but when he challenges me with an outlandish proclamation, somehow he inspires me to rise to the occasion. The dialog goes something like this:
DIRK: <Outrageous claim>.
DIRK: No it's not.
DIRK: NO IT'S NOT.
MIKE: Wait. <silence> Hold on. <silence> Maybe. <silence>. It might work...
What had eluded me the entire time was that the solution was not to process a massive list of dice modifiers in as little time as possible - that's the brute force approach. We have computers for that and it's just not practical for a board game. I had to "cheat" and find a way to get it done much more elegantly. Then it came to me - whenever a unit fires at another, no matter how massive the list of modifiers are for a game, only a handful actually apply. The vast majority are totally irrelevant. There had to be a mechanic that allowed a player to declare a target and roll dice without making a SINGLE calculation. Then concern himself with a VERY SHORT list of the modifiers that are only relevant for THAT situation. There was a mechanic but it did not involve dice at all...