Randomizer Geekery

General GTG General Discussion
1

So, it's clear proof that I don't have enough things to keep me occupied, but I found myself wondering about how various randomizers (including people who use dice, as well as the various programs that have been cooked up) handle the exact process of random selection.  A very geeky question, and not an important one, but I'm idly curious. [1]

 

* When constructing a game, do you set a number of heroes, then randomly select that many all at once?  Or do you select one hero, then eliminate them from the list and select another hero, and repeat until you have the required number?  I'm not sure whether it makes any actual difference, but I felt like it was a somehow interesting question in the first place, whether you are simultaneously selecting 3-5 different items from a set of 18 (not counting promos; see below), or selecting one item from a set of 18, then another from a set of 17, then another from a set of 16, and possibly repeating this pattern up to twice more.

* When selecting heroes, do you randomly select a hero, then pick which version of the hero to use if there are more than one?  Or do you pick from a list of all hero character cards, whether or not they are unique?  The former method means you're three times as likely to play a unique hero (such as Expatriette or the Scholar) as you are any particular version of Wraith, Legacy or Bunker, while the latter means that Wraith, Legacy and Bunker show up three times as often as the unique heroes.  In either case, of course, heroes with exactly two versions, like Ra and Unity, are in-between; I figure that ideally, you'd somehow compromise between the two methods, but the only way I can think of to do that would be to make a very large randomizer list, in which a unique hero's name appears twelve times, a two-version hero has nine listings per version, and e.g. Legacy has eight listings as Finest, eight as Newest and eight as Greatest [2], giving Legacy overall exactly twice as much chance of appearing as The Scholar, but any one version of Legacy 2/3 as much chance - I don't know if it's possible to make the two ratios actually equal, and if so it would probably involve padding the list out to a much greater length.

* Is the determination of normal or advanced status handled together or separately for each villain?  In other words, is there one selection from a list of 40 (the 18 distinct villains, plus MB Blade and Cosmotron, and then advanced versions of each), or is there a selection from a list of 20, then a coin flip for advanced or regular?  (This is assuming you don't just always play on Advanced, as the Episode series of forum games do.)

 

[1]:  I do not use any of the randomizers and thus have no familiarity with them.  The one exception has been a single game in which I picked four villain dividers and randomly drew one of them to decide which of my new villains we'd face that day.  I forget exactly how I did it, but I believe it involved holding them under the table, mixing them up, and getting a player who couldn't see them to choose "1, 2, 3 or 4" based on the order I was holding them in.

[2] Speaking of which, if Grandpa Legacy is the "Greatest" and our version is merely "Fine", does he have any feelings of inadequacy toward his better-acclaimed predecessor?  Seems like he'd probably have something to talk about with a psychiatrist.  Government sponsored, of course; America wouldn't want its national champion having a nervous breakdown, after all.  Goodness knows what that could look like....

2

Fun fact, which I just discovered due to thinking about your questioning the randomisers - if you're selecting from a pool of all cards including promos, heroes with three copies will actually appear less than three times as often as those without. It would be three times as often if you could pull multiple copies of the same hero into the same game, but as it is the numbers work out so that you'll see the Scholar slightly more than any individual Legacy.

3

The randomizer I use most often selects the heroes one at a time, if it selects a hero that has a promo version, then it will randomly determine which verion it will use. It then "removes" that hero from the list and repeats the process all over again, until the predetermined number of heroes have been reached. They are then put into the order that they were randomly selected. Some of the players will call out which hero they want out of the random batch of heroes. Others will roll a die and assign numbers to remaining heroes.

4

Ronway's description sounds like how my randomizer works.  No hero should have any greater chance of appearing in the final list than any other, regardless of how many alt versions they have, and you will never see two alt version of the same hero in the results either.

5

We randomise out villain and environments with dice but choose heroes after those have been rolled - roll 1d20 (because there's no such thing as a d18) to choose the villain, rerolling if we get a 19 or a 20, or if we roll someone we've already fought that weekend, and then roll 1d12 for the environment, again rerolling if we get a repeat. If we've managed to play so many games that all the environments (or even all the villains) have been done, we "start again".

We did try randomising heroes for a while, making the same roll as for villains (but with less inclination to reroll unless we get someone we only just played), and rolling 1d2 if the hero was one with a promo (1d3 in the case of the Wraith, of course) to see which version would be played. But we kept losing a lot and were getting rather frustrated so went back to choosing heroes again. I play as one hero, while my partner plays as two - early on I did play two as well but I prefer only having one deck to concentrate on, plus I also seem to have a tendency to play those who end up with a whole load of cards out in which case I kind of need as much table space as I can get ;).

6

I personally use a tiny magic donkey. I show him one of every hero card and he taps his little hoof once for yes and twice for no. Sometimes he goes a little crazy and taps out what sounds like the drum solo to In A Gadda Da Vida. That's the last time I show him the Engine of War card.

7

I generally don't randomize heroes, but if I need to I roll my trusty d24. (24 is either a reroll or a wild card, depending on my mood.) Environments use a d14, and villains use a d20 (which does not include the vengeance villains, but I only do those voluntarily anyway.) The villain d20 includes promos, since those are so drastically different, but hero promos are considered the same, so if you roll wraith you get to pick.

8

We have friends who just shuffle the character cards. It works really well, actually. 

I'll point out to them that the alternates create an increased chance for those heros to show up… :confused:

9

Sentinels Sidekick can randomize the whole game or individual slots. Either way, promo cards are ignored in the initial choice. If a card with promos is chosen, then another random draw is made to determine which of those to use. It works exactly like Ronway said =)

You can also tell Sidekick what expansions & promo cards you have, so it won't draw cards you can't play. And you can tell it to allow duplicate heroes.

FWIW Envisioner: The two methods you describe in your first paragraph are exactly the same. It's a simple random sample without replacement.

 

10

We play that if you roll a hero with a promo card you pick the one to use.  We don't like unbalanced teams either, so we will roll a villain and environment and usually 2 heroes, then if there is a glaring weakness we pick heroes to compensate.  Also that way we can avoid TLT or Dark Visionary most of the time.

11

Yeah, but I don't have an i-device. :confused: Greatly looking forward to an android version! :slightly_smiling_face:

12

Patience =)

13

I didn't really think they were different, I just wondered which one was being used.

14

I really like the random. Sure, sometimes you get a team that isn't ideal (or even downright sucks) for a particular villain-environment combo but working at a disadvantage can be a lot of fun.

15

If it comes up, I use Spiff's. Mostly the people I play with pick and choose who they will use. And then I choose villain and environment unless someone else has something specific in mind of who and where.

16

When he said both methods are the same, he meant that exactly, to the point that you can't really differentiate which method is actually used.  More to the point, the former method is technically impossible.  Even attempting to select many at once would actually be the same as selecting one at a time, eliminating it from the list, the only difference being how much time occurs between selecting one and selecting the next. (In other words, an individual quantum of time is so small that the likelyhood of one being selected at the same time as another is so small as to be practically impossible).  To be fair, you could be wanting to know how many people randomly select their heroes knowing what the first selections of heroes are before the last selections of heroes are reveiled, though this would have little change in the actually results.

Now that I think about it, there is a way for these to differ, not in which heroes are selected, but how they are assigned.  Maybe this is what you refered to, maybe it's not, but it's something I find would be interesting.  While selecting heroes one at a time, removing from the list as they are selected, the reveiled hero is assigned a player prior to the selection of the next hero.  This means that the players have to choose who gets that hero not knowing what heroes might come next; they have to choose between taking something good now, or holding out for something better at the risk of being stuck with something worse.

17

Or they find all heroes awesome, thus nothing is being risked at all!

18

I've gotten to the point where I agree with this.  I played Nightmist a couple more times where I didn't focus on her being a damage dealer and it went much better.  She was the last hero I didn't like to play as.

Bunker I've grown quite fond of.  You just have to get over that desire for him to have all his cards out and dominate the game. and find his real utility.

19

Since I currently only have 14 heroes, 10 villains, and 8 environments I just use a d20.  One roll for villain; one roll for environment (reroll 17-20).

For heroes I roll the d20 and 01-14 is the standard hero.  If I roll 15 or higher I roll again and if 01-12 I take the promo from the stack of those instead.  Less chance of getting a promo version of a hero, but there is a chance they get into the solo game I am setting up.

20

@ the Jay Mann:  One way I can envision (see what I did there?) a difference is if you used a deck of sleeved character cards, thusly unable to tell their identities from the back.  You could shuffle the deck and deal out five face-down cards, then simultaneously reveal them all, or you could reveal one at a time until you had five (you'd also have the option of shuffling more times in between individual reveals).  The end result is the same, but you'd see either one hero at a time or five of them at once, so they'd feel different.