|
Some statistics about the raffle: - 258 people entered the raffle This raffle isn’t entirely comparable to the last one because last time the grand prize was a pre-made custom rather than a breeding, which I assume would be more attractive to users who may not yet have gotten their feet wet RPing. Still, I think it’s still helpful to look at the numbers. So comparing to last time: - 180 entered the raffle (all of them bought at least one ticket) While this raffle was somewhat less effective as a gold-sink, I really enjoyed the additional participation (+42%) that the free ticket brought. I felt it worked very well with the higher ticket price, combining both the “everyone should be allowed to play” with the “grand prizes are big prizes so tickets should be expensive” feedback we’ve been getting. **So, I think the experiment was a success, and generally feel that this is a good model for future raffles**... but then again, I didn’t really participate so I don’t know what it felt like. What do you all think?
Posted 09/07/14, edited 09/07/14
|
|
|
I liked the free ticket, because it did get more people participating and lessened the impact to nuggets for people who don’t have nuggets to throw away (especially since this one coincided with September OotS). I think it would be a good idea to keep a mix of expensive vs inexpensive raffles just to make sure everyone is getting a chance to buy tickets if they want them. (edit: Skye brought up a good point though of: the most coveted raffle only items should not force people to buy tickets at 1k+ a pop every time. I only bought 5 tickets + the freebie because it was all I could afford, and I only bought that much because I wanted the breeding and glowing changingshroom. But it seems like the only players who won the big ticket items all had capped tickets at 13.) Though one trend I’ve noticed that has me bummed out and confused is that in the past couple raffles it seems like most of the winning tickets have been 400+, with for example, only one “13” ticket winning and the rest being 250+. It feels like that’s true of past raffles as well (though it’s only a feeling, I can’t remember the actual winning numbers from last time, the time before, etc). It kind of makes it feel like it’s not worth buying tickets until after that mark because RNG is favoring higher numbers. It just doesn’t seem random, even though I know it is (probability and statistics say it is possible to get a cluster of random numbers). You would normally expect a wide range of numbers across the spectrum instead of clusters of 400-1000. It’s not really a concern that can be remedied, but I thought I’d put that out there in case it’s something to do with the settings you use on the random number generator or the generator itself being skewed. Edit: How are the numbers drawn? I know you say you ignore numbers belonging to people who already won something, but do you key in that person’s numbers to have the generator ignore those numbers? Or do you keep drawing from 1-1000 and just ignore numbers from people who have won if they come up? If you somehow tell the generator to ignore certain numbers, that might skew the odds. But I don’t know.
Posted 09/07/14, edited 09/07/14
|
|
|
I DO like the “Everyone being allowed to play” mechanic, that was super nice, and some people who only had the free ticket even won! And that’s pretty awesome to me… I also LOVE smaller ticket caps! However in contrast, personally, I didn’t like the price of more tickets. Yes, the grand prizes are big prizes, but look at it this way… In real life, I COULD go to Vegas and pop a quarter I found on the street into a slot machine and might win a million bucks. Same thing applies to buying a 1$ scratch ticket at the grocery store. The whole notion that “big prizes should be expensive” is like saying only millionaires should be allowed to win the lottery. Personally I don’t think we should have to break the bank (and sacrifice other wants for some users) just to buy the delusion of winning. Because lotteries and raffles based on chance are just that, based on -chance-, it shouldn’t matter if you’re “Rich” or “Poor”. Even on here, and double especially since as we gain more player numbers, the chance of winning them is going to drop more and more and more, already we’re at 258 people, of which only 20 won, and MOST of those prizes -weren’t- “big prizes”, only two things are -only- obtainable through a raffle at the moment (and are the only reasons most people buy tickets -I- imagine. It’s why _I_ bought more than the free one.)... So realistically drop that down to TWO in 258. That percentage/odds of winning that are… Depressing. Super freaking depressing. (Which is the other reason I didn’t buy many tickets this raffle, especially with trying to save up for other things.) That’s less than 10% chance of winning… Between 7-8%, actually… Which means on average I have better odds at wining 100$ from a scratch-ticket. (Those advertize a range of 1 in 4 to 1 in 10 for prizes below 1k, and I’ve won 25-100$ off those several times in my life.) Especially looking at the RNG on top of it all, which is rather prone to ignoring whole huge blocks of ticket numbers… What I mean by that is, sure it works out to “One in 13 got a prize”, but If you group players up in groups of 13 in order of tickets purchased… The numbers drawn have winners all grouped up weirdly, so it still doesn’t feel at all that ‘equal’. (Example ticket groups from 14-253, where as to there’s a LOT of them in the 500+ range. 14 of the 20 prizes were 500+, which is weird to me if there were only 1141 tickets, statistically shouldn’t those be more spread out? ... But I digress, that tangented a bit off my topic because RNG is a constant enemy of mine. lol.) This is getting very TLDR for a lotttt of players so to simplify my logic here. “We’re already just buying a delusion!” We’re buying tickets -just- to sit here for a few days dreaming of winning. And personally, I feel that the 500 price, maybe with a lower ticket cap, is probably better for that, since it won’t bankrupt so many players who are already struggling to -just- afford monthly coats, and sometimes only manage every few months.
Posted 09/07/14, edited 09/07/14
|
|
|
I know the whole point is to gold sink, but holy crap I had a hard time earning enough nuggets to max. I’m not sure if it was the timing or what, but it was a stuggle and without an anon donation I got I wouldn’t have made it in time 8’D Also, I almost have a small sense of regret dropping that much, which I haven’t felt before, oops.
Posted 09/07/14
|
|
|
While I do acknowledge that a gold-sink is a good idea, as someone who rarely has gold to sink, I really liked the format with the free ticket. My RP partner and I both had a shot - a limited one, yeah, but still a chance - and then I bought one more ticket and that was that. We didn’t expect to win but there also wasn’t that ‘man I should have bought at least one’ regret going on.
Posted 09/07/14
|
|
|
I pretty much feel the same as Skye on this one :> ty for saying all the words me poor tired brain can’t muster, haha. Gotta reiterate especially though that however raffles are operated, the one free ticket is much appreciated. Even if it is a little chance, it is better than none, and I am happy to have had one in this raffle. I suppose it would feel rather unnecessary if the entry fee was much lower, though. I would just drop some money on one ticket.
Posted 09/07/14
|
|
|
Because the “RNG” stuff has a definite answer and the mechanics things are a discussion, I’ll address the RNG comments here and save the mechanics feedback for later. I think most of the concern stems from misconceptions about how random distributions work: —- *“But it seems like the only players who won the big ticket items all had capped tickets at 13”* It may seem that way when you’re looking at a lot of 13s on the winner list, but it isn’t true. In this raffle, 4th place winning a changingshroom was won by a user who bought 3 out of 12 tickets. In the last raffle, the grand prize was won by someone who bought 10 out of 25 tickets. That said, people who buy more tickets do tend to win more often and do tend to win better prizes, because for any given ticket they are more likely to win than players who buy fewer tickets, which results in them being more likely to be drawn sooner as well. —- *“How are the numbers drawn?”* All raffles on Mycena Cave thus far have used the same algorithm. The drawing works like this: - select a ticket at random from a uniform distribution over the ticket pool There are no “parameters” or “ignored numbers” because any parameter would by definition change the odds of the function. —- *“only one “13” ticket winning and the rest being 250+”* The next ticket was 254, which would put reasonable grouping or cluster sizes at around 100. As for why only one <200 ticket got selected: well first of all, with a sample size of 20, you really can’t make any kind of statistically significant observations over a random distribution of 1141 elements — but ignoring that for a moment — any given ticket had a roughly 8% chance of being falling into any 100-range group. So while it is true that over a large number of drawings one would expect around three tickets in the 0-200 range, over a single drawing having only one is not unusual. —- *“You would normally expect a wide range of numbers across the spectrum”* and *“looking at the RNG on top of it all, which is rather prone to ignoring whole huge blocks of ticket numbers”* You should expect the opposite of a wide range of numbers across the spectrum: clumping and uneven drawings are exactly how random distributions behave with a small sample-to-size ratio (ours was ~0.0175, i.e. tiny). Evenly distributed results are actually very non-random, and getting a wide range of numbers across the spectrum is very unlikely. This effect gets more pronounced the fewer tickets are drawn as compared to the number of tickets in the pot. If the clumps tended to be the same between many raffles, that would be somewhat more surprising. However, as can be seen by past raffle results in the announcement forum, this has not been the case. *Edit: this is a [really cool explanation](http://www.wired.com/2012/12/what-does-randomness-look-like/) on the topic*
Posted 09/07/14, edited 09/12/14
|
|
