Random Allocation, and similar rules, are the sort of thing that can really blindside you when playing 40k. The Wrath of Magnus. for example, has a Warpstorm result where you have to pick a random model with the Daemon rule, which might well be a single model from an army of 100 or more eligible candidates. Let's have a look at how to handle it.
To look first at Random Allocation itself, where we simply have to pick a random model in a unit to take a Wound, we find the 40k rulebook is not much help, simply telling us to pick the model 'at random'. So how can we do this?
Firstly, we can just let a dice sort it out. This seems simple, and sometimes is, but has its problems. In the ideal case where a unit is exactly six strong and standing in a line, the humble D6 rides to our rescue. If there are more or less than six models, though, it might help to have a bag of old-school RPG dice- D4, D10s, D12s or even D20s. Even then, you probably won't have the exact right dice, but most gamers will handle this by rolling, say, a D10 for a 9-man squad and just re-rolling a 10. There are a couple of pitfalls to beware of, though. Firstly, and this is what we might call rookie mistake #1, never roll multiple dice at once for this- say by rolling 2D6 to pick between models in a unit of 12. If I have to explain why, you've probably not been doing this very long but suffice it to say that whichever model got assigned '7' will be getting pretty nervous.
|Hint #2- he's in the middle of this.|
My preferred method with Random Allocation is to do just that- simply reach into the squad without looking too hard and grabbing the first model to hand. This has the advantage of speed, but obviously is too open to abuse to be used in tournaments or other situations where player trust might be a little lacking. Some rules, like Mob Rule, imply a 'fluffy' source for the damage, such as allocating wounds from Breaking 'Eads to the nearest Boyz to the Nob, or removing the runtiest Ladz in a Squabble, but again these can hardly be used in a tournament.
So, here's an alternative method, which comes to us from the world of computer programming. Basically, the Chop relies on splitting a set of data in two, and then picking the most desirable of the two remaining sets. So for example, to randomly pick a model in a squad you draw a line through the middle and flip a coin, or roll a D6 (1-3 left, 4-6 right). You then halve the selected models again, and roll again. Obviously this will often mean one 'half' is one model bigger than the other, but overall this should even out. Eventually you end up with a single lucky (or unlucky) winner, and since both players can clearly see what's going on, it's less likely that confusion will result.
The Chop principle can also be used in other random situations. For example, if we have to pick a single random model on the table, we can simply split the table down the middle and roll to pick one half or the other. Then we split that half, and roll again. This will start to zero in on a single location pretty fast, even in an Apocalypse game, and of course once a zone holds only a few models we can easily use a more traditional method to make the final selection- the Chop's principle advantage is how fast it handles large amounts of data, rather than in efficiency at the end.
We can even use the Chop on other random events, like Weapon Destroyed results, by zoning the tank's hull into port and starboard or fore and aft, though usually this can be handled with a single roll.
There are probably other ways of randomly selecting, but these are my thoughts on the matter. Let me know your preferred method in the comments, if you feel the urge...