A lot of players object to the randomness of dice. Over the years there have been many, many attempts to limit or eliminate some or all of the random parts of ASL. These often walk into traps where by removing the total randomness they introduce some predictability which can be exploited in a sleazy way. For instance every so often someone will suggest that instead of using dice, use a card deck that has one "two", two "threes", three "fours" and so on. That way after every thirty-six draws all thirty-six possible DR combinations have been used and the results are "guaranteed fair". After thirty-six draws, shuffle the deck and start again. The problem with this mechanic is that after a few draws if you "count cards" you can start making predictions about what is likely to come next. If you have drawn thirty-five times and have not seen the "two", try for a crazy ordnance shot. Whoa! Critical hit! Who saw that coming? Is the "twelve" likely to come up soon? Dig some foxholes to draw it off. The person suggesting it then makes adjustments that make it harder to predict, but the changes either leave a sleaze gap or they approximate using dice.
The more fundamental problem is that I think reducing the randomness runs contrary to the premise of the game. One of the premises of the game (besides that there are certain participants that make a much larger contribution to the battle than their numbers might suggest, i.e. leaders) is that a good commander (aka player) has to be able to balance trying to cause bad things to his opponent, avoiding bad things his opponent tries to cause, and being able to adjust to bad things that happen just 'cause. Sometimes it's a three-way battle, with luck the winner.
There are certain places where the luck wins too easily. Directive Number Three lets both players set up and move a turn with a board between them, then remove the board and roll to see who moves first. I think that the winner of the first move has a great advantage (although I may be wrong, and it's a wonderful scenario anyway). Blazin' Chariots is a scenario where it seems as if when one player gets ahead early it's hard for his opponent to come back. So you have to be careful when designing things because early bad rolls can spoil the fun of the game if they take on too much import in the game. Overall in well-designed scenarios I think the luck factor is part of what makes the game thrilling.
JR