I’m spinning this off of the main thread about tracing the origin of the term “d66” because it’s not strictly germane to the topic – none of these examples actually use the term “d66” to describe their dice-rolling methods – but I’m going to post it anyway as a matter of general interest: following a conversation with Tumblr user @notclevr, it appears that before tabletop wargames (and, nearly concurrently, tabletop RPGs) got their hands on the mechanic, the principal (though by no means exclusive) users of the old “roll a six-sided die twice, reading one die as the ‘tens’ place and the other die as the 'ones’ place” trick may have been tabletop American baseball simulators.
The most notable example of the type – and the only well-known example still in publication today – is J Richard Seitz’ APBA Baseball, first published in either 1950 or 1951 (accounts vary). In this game, a d66 roll is cross-referenced with a card representing the active player and a “board” representing the current situation on the field:
For example, with Carlton Fisk at bat, a d66 roll of 31 would yield a result of “8”. Assuming for the sake of argument that the situation on the field is a runner on first and a grade C pitcher, consulting the “Runner on First Base” board, this corresponds to an outcome of “SINGLE—line drive to left; runner to third”.
(This example is, strictly speaking, incorrect, as Carlton Fisk didn’t have his major league debut until 1969 and I’m using the wrong lookup tables for any year in which he played, but you get the idea!)
Interestingly, APBA Baseball is not the first game to use this setup. It’s heavily derived from Clifford Van Beek’s National Pastime, a game whose patent was registered in 1925, though it wasn’t actually published until 1930. Even at a glance, the similarities are substantial:
Indeed, though National Pastime’s lookup tables are much simpler than APBA Baseball’s, where they overlap they’re often word for word identical. It’s generally accepted that Seitz plagiarised National Pastime without credit when creating APBA Baseball (ironically, given his own famously combative stance toward alleged imitators!), though he was within his rights to do so, as National Pastime had fallen into the public domain by the time APBA Baseball was published.
We can go back even further, though. As far as I’ve been able to determine, the earliest known tabletop baseball simulator to use d66 lookup tables for resolving plays is Edward K McGill’s Our National Ball Game, first published in 1886:
A copy of the game’s 1887 US patent application can be downloaded here. This one uses an unusual 21-entry variant of the standard d66 lookup table in which the order of the rolled digits is insignificant, with doubles being half as likely as non-doubles rolls; it’s unclear whether McGill was aware of this when he laid out the table. Unlike later incarnations of the genre, there are no individual player statistics, with all at-bats being resolved via the same table.
