You should read the posted article - it's only a page long. For those who don't wish to do so, the upshot of it is the idea that war games really need some form of randomization, but that dice are possibly not the best answer. The author provides an alternate form of randomization: a deck (or more) of playing cards, with each player having the same number of standard decks (each card to be drawn in place of a die roll). The idea behind this being that neither player gains an advantage, as each has the same spread of random numbers. Over the course of the entire game, the actual numbers drawn will be the same by each player, therefore no player gains advantage from being a lucky roller.
A commenter on that post pointed out - correctly, IMHO - that this is only of limited benefit: players who are good at counting cards would easily be at an advantage.
However, there's another flaw: the author's premise is predicated on the assumption that each player will roll an equal number of dice (and thus, using the alternate system, draw the same number of cards) during the course of the game. While I'm not a "grognard" by any means, I have played a wide variety of war games, and I've never played one wherein each player rolls the same number of dice during the course of the game. Based on my experience, this solution would not actually solve anything; the unlucky player, using fewer cards, would still be likely to find himself drawing more low cards than his opponent. Or vice versa: his opponent, using fewer cards, would find himself drawing more high cards. Either way, the solution fails.
I do agree with the author's point of view regarding randomization, however. A war game without chance is, to me, a flat-out mental exercise, like chess or Chinese checkers - not a war game. While I find these games fun, I do not seek the same sort of entertainment from war games.
However, being one of those Unfortunates for whom all forms of randomization seem to do nothing more than provide fickle Fate an opportunity to make life interesting (in the Chinese curse sense of the word) I can definitely see the benefits of mitigating the role randomization plays in a game.
Case in point: I recently played a pair of games that served as a prime example of how randomization can be handled badly; namely, Ganesha Games' Fear & Faith (F&F) and Song of Blades & Heroes (SoBH).
Before fans of these games feel the need to leap to their defense, let me clearly state that I'm also a fan. I love their simple yet highly flexible nature. It's not my goal to denigrate these games, but merely to point out what - in my case - happens to be a flaw in the design of their core combat system.
My fiancee and I started the day with a game of F&F. After I lost abysmally, and was left feeling thoroughly dissatisfied as a result, we tried a game of SoBH. All started well, and for a couple of turns I had a good advantage, thanks to some decent tactical use of my models. However, a few turns in, my most powerful unit engaged in melee one of my fiancee's slightly less-powerful units. In one turn, my unit - the strongest and most costly unit of my entire force - went from being fresh to the fight to being dead, simply because I rolled a 1 and my opponent rolled a 6.
It was all down hill from there; the game ended as badly for me as the previous F&F session had.
Again, I don't hate the games. But these two sessions made me wonder: why had I lost so badly?
Don't get me wrong. I'm no stranger to losing games. (As readers of this blog probably already know. Hello! Look at the title!) And I rarely blame a game system for my failure to achieve victory. I admittedly take actions to make the game more enjoyable instead of making sound tactical decisions. I'm more concerned about playing an interesting game than I am with winning. I expect that to cost me, and it usually does.
As it turns out, Ganesha's house system works fine when the players' rolls average out. With a little bit of tactics to stack up modifiers, models have a fair chance of defeating enemies of equal Combat score values. It's fairly sound, and I've played games of Mutants & Death Ray Guns in the past that went fairly well.
However, the Ganesha system breaks down when one player is a lucky roller and that player's opponent is a Victim of Venomous FateTM. And that scenario perfectly describes every game between my fiancee and myself. Invariably (well, except when she's not feeling well or is tired), my fiancee bends the dice-rolling bell curve toward the high end. I, on the other hand, bend that curve toward the low end on a regular basis. (I know what you skeptics are thinking, but rest assured that it has nothing to do with the dice themselves. Regardless of which dice are used, the outcome's always the same.) In most games, this is an annoyance, but not a deal breaker.
Not in the case of the Ganesha house system, though. In that system, low rollers have little chance of defeating high rollers, much less removing them from the table. On the other hand, lucky rollers have an excellent chance of not only defeating an enemy, but in removing that enemy from the table. The minor problem of the combat system's win-to-loss ratio is exacerbated by the fact that, with a single roll, even the most powerful model can go from healthy to out of action. In the case of lucky versus unlucky, the unlucky player is going to be fighting an uphill battle to just survive the game, much less win it. That's exactly what happened to my demon, Bugly - despite a power attack (which should have put his enemy at a severe disadvantage) he was utterly decimated because of a single roll of the dice.
Honestly, after thinking about it and looking at the numbers, it seems like my best tactic when facing my fiancee in a Ganesha Games system is to stay out of hand-to-hand combat entirely. Of course, that doesn't help much, because even if I have ranged weapons and she does not (as in the F&F game we played) she's got almost a 1 in 3 chance of not being affected by my attacks. It is (and was, as a practical matter) just a matter of time before I'm forced to engage in melee - and then it's all over.
So, what does this all mean for the larger issue of randomization?
Well, I'm not saying that using dice to add a random aspect to a system doesn't work. But that random aspect has to be mitigated. The system, IMO, should never put a player in the situation I found myself in with Bugly. A unit that tough should never live or die on the merit of a single die roll, unless the system allows for stacking of modifiers based on tactical decisions (much greater stacking than is provided for by Ganesha's system). And even then it's questionable if a single die roll is a good idea...
Ideally, a system that uses randomizers should allow players to roll multiple dice (or draw multiple cards) per combat roll. Necromunda and Malifaux spring to mind as good examples of this. Also, although it's nice and simple to roll everything (win/loss, damage, and effect) into a single combat roll, it's probably not such a good idea to do so. Allowing one roll to determine the fate of a unit - especially a powerful one - doesn't allow a player a chance to recover from the effects of an unlucky roll of the dice. A simple, one-roll system too easily leaves players at the mercy of their dice, and makes tactical decision making take a back seat to the random element. It's also a surefire recipe for leaving a player feeling cheated at game's end.
Mind you, this is all opinion, expressed from my point of view as a player whose dice hate him with an unbridled passion. Your mileage may, of course, vary.
. . . . .