Talk:Lucky Hochei

Why we should all love Rock, Paper, Scissors
Playing 25 times yielded the following results. For a grand total of 12 Fortunes (which would have cost 36 Tokens at the LFG) AND 8 free Sugary Blue Drinks        00:34, 17 February 2007 (CST)
 * 8 wins = 8 fortunes + 8 Sugary Blue Drinks
 * 4 ties = 4 fortunes
 * 13 loses = nothing
 * Problem is that in the time it took you to play 25 times, you could have easily gained more than 36 tokens by doing the quests on secondary characters, or playing beetle races. 83.159.9.78 06:20, 17 February 2007 (CST)
 * I was just thinking of maximizing the return on Tokens, for people who don't have many characters or who have done all the quests on their current characters and were unable/didn't want to make more.
 * If you suck at beetle races, you may not get many Tokens. The races to take time to Enter (sometimes not enough players) and time to finish.
 * Using the stats someone added to the article (Playing 99 consecutive rounds using /rock each round, yielded 33 Sugary Blue Drinks and 62 Lunar Fortunes.), it would have cost 62*3-99=87 MORE Tokens at the Lunar Fortune Giver
 * You would have received ZERO Sugary Blue Drinks (IDK their current value... but 100gp would seem like a reasonable minimum).
 * To get 87 Tokens Beetle Racing it would take 13 1st place wins OR 18 2nd place wins OR 29 3rd place wins OR some combination thereof...   [[Image:jkr.jpg]]     18:31, 17 February 2007 (CST)

Rock?
If you use the command /rock, you'll have the most chance to win/tie and get fortunes. Well, I noticed he uses alot of scissors/rock, so I used rock all the time and I got alot (soz, didn't count the total) of fortunes. anyone else noticed this? --Elder 04:12, 17 February 2007 (CST)


 * Not to be too much of a troll but you realize there's an equal chance to win with every option, right? --Pork soldier 07:08, 17 February 2007 (CST)


 * I hate to sound like one of those people who look for patterns in lottery numbers, but since computer-generated random numbers cannot be entirely guaranteed to be random in their distribution, I wanted to keep count of my RPS games. Below is his choices, FWIW, YMMV, ETC.--Ishmaeel 07:35, 17 February 2007 (CST)
 * Rock : 40 (27%)
 * Paper : 50 (35%)
 * Scissors : 55 (38%)


 * I didn't note it in my original post but I choose /rock each of the 25 times, so I wouldn't have to retype anything ;-) ...And as the saying goes, "There are lies, there are damn lies, and then there statistics"   [[Image:jkr.jpg]]     18:31, 17 February 2007 (CST)


 * Ishmaeel, you do realize that your confidence on a 145 number set is 12 points right? Your expected value is about 48 and all of your measured results are close enough to the range 42-54 to be statistically meaningless for a conclusion of favoritism. Give me those percentages on a 1000 point set then we'll start to talk. --Tometheus 11:56, 18 February 2007 (CST)

I found that if I chose whatever option he chose last, I had about 4/5 chance of winning. I only started doing this after I had 20~30 tickets left so I didn't keep a fully accurate count, but I only lost about 4~6 times after doing this. After I get 99 tickets I'll try it again and keep track.
 * Out of 99 tokens, my results are: Rock = 32, Paper = 26, and Scissors = 41.
 * And as a side note, I ended up with 62 fortune tickets, and 23 bottles of sugary drinks.

I think every district is different. I was in one where the guy almost never did Scissors, so everyone would always do paper to tie or win almost every game. - Entice789  (Talk | Contributions) 01:34, 18 February 2007 (CST)

Just finished 500 tickets with lucky hochei, pain in the ass, choosing rock everytime. Ended up with 346 fortunes, and 204 sugary blue drinks. That means I won 204 times, tied 142 times, and lost 154 times. Tha means I had about 40% chance of winning, about 28% chance of a tie, and 30% chance of losing. Given that 500 tickets still isn't all that accurate, one could say that all three numbers are close to 33%. Thus, I'd say that choosing rock everytime over anything else wouldn't give you a higher chance of winning. Doesn't mean you shouldn't though, as switching everytime doesn't net you more wins either. (assuming lucky hochei chooses on random) Shadowmist 18:42, 18 February 2007 (CST)

"Good ol' rock. Nohing beats rock." 132.203.83.38 09:50, 19 February 2007 (CST)

Switching based on last throw -- Gambler's Fallacy

 * To get the most yield for Lunar Fortunes, consider switching your options of rock, paper, or scissors based on what Lucky Hochei threw out the previous round.
 * If he had rock on the previous round, try to shoot for scissors for the coming round.
 * You will still recieve a Lunar Fortunes even if it is a tie and if you win, you will also get 1 Sugary Blue Drinks.
 * The chances of Lucky Hochei repeating his previous move are low.
 * Repeat this tactic and you should get more Lunar Fortune than just plain guessing every game.

Removing this from the article. This is what is known as the Gambler's fallacy. "The chances of Lucky Hochei repeating his previous move are low." For a random event, in any given throw the chances of him throwing a repeat are exactly 1:3. That's the exact same odds as if I say "The chances of him throwing scissors are low" and just keep throwing paper. It seems silly that ANet would, rather than using a random number generator, actually code in something that says "if X has been thrown, we'll reduce the odds of getting X again." I've seen him throw something 5 times in a row, which by the above reasoning would be next to impossible. However, each throw is random, so you have 1/3 chance of paper, 1/3 chance of rock and 1/3 chance of scissors. EACH throw. The past has nothing to do with it.


 * The gambler's fallacy is a logical fallacy involving the mistaken belief that past events will affect future events when dealing with random activities, such as many gambling games. It can encompass any of the following misconceptions:
 * A random event is less likely to occur because it recently happened.
 * Gambler's fallacy

--Tometheus 10:41, 18 February 2007 (CST)

It wouldn't be entirely silly for A-Net to use something else than uniform distribution, it would add an element of fun to the player try to guess some other semi-random algorithm. Of course, with uniform distribution being the obvious choice, we'd need strong evidence to suggest that different is actually in effect. --Per Abrahamsen 15:45, 18 February 2007 (CST)

He threw 9 papers in a row once. Lots of people weren't happy cause they just kept using rock. LoL 132.203.83.38 09:50, 19 February 2007 (CST)

I'm not so sure it's completely random. I've gotten pretty long winning/tying streaks by basing my next move on his words from the last throw. Is it possible it's not completely random?

More stats stuff
No, you don't have a 100% chance to win an item over extended play. The chance increases but will never reach 100%. Saying "you will receive X items" for 99 games is also false since the expected values or averages might be those numbers but you might actually receive squat. --Fyren 09:28, 19 February 2007 (CST)


 * The information needs to be in there. Run some statistical analysis of your own.  I've theorized and played over 1500 games myself, it comes to be a 1:1 ratio trade-off of tokens to an item.  If this is a truely random system, .3333(bar) of the time will result in 0 items, .3333(bar) of the time will result in 1 items, and .3333(bar) of the time will result in 2 items.  Taking the average of the items you can win over the total possibility will give a 1:1 ratio.  I understand what you are saying, and that is why I put the extended play in there.  If you dont like how it is in there, put it in there in a form that you like and give me credit for it.  --Glassman324 09:55, 19 February 2007 (CST)


 * If you recognize it's an average then you also recognize that what you actually said is false. Don't try to simplify things if it means what you say is incorrect.  --Fyren 09:58, 19 February 2007 (CST)


 * Over time, a statistical average flattens out to become the real average. Nothing that I am saying is incorrect.  Run it through a statistical analysis program, get a white board and do the math, try it in game.  It all comes out to be 1:1 ratio.  --Glassman324 010:09, 19 February 2007 (CST)


 * are you talking about that because you should get 0 items 1/3 of the time, 1 item 1/3 of the time an 2 items (fortune and drink) for the final 1/3 you came to a result that gives you a 1:1 ratio of items to plays over time. because that that is correct if you look at it like that. --Lemming64  10:10, 19 February 2007 (CST)


 * Thank you. --Glassman324 10:13, 19 February 2007 (CST)


 * You said you have "a 100% chance" of getting an item. You don't.  Your chance of getting an item will never reach 100% no matter how many games you play.  You said "99 games will result in 99 items."  They don't.  Saying the average will be the actual outcome is incorrect.  You're mixing up reality and statistics.  You just said that over time, the statistical average tends to the real average; that's backwards.  You're not stating the average as an average, you're stating it as truth.  --Fyren 10:19, 19 February 2007 (CST)


 * Then put the information in there as you see fit. One of the idea behind a wiki is free knowledge, and you keep stripping people of free knowledge when you completely delete my posts (instead of changing them).  If you would be so kind to put the posts up there, with the proper statistical lingo, that would be great.  I'm done with this discussion.  Sincerely, --Glassman324 10:26, 19 February 2007 (CST)


 * The whole thing seems a little pointless really, with the information already there anyone who cares can probably work it out for themselves. The 100% thing could be worded in a way that doesn't imply that you WILL win an item every time you play, but again probably best to just not bother. --Lemming64  10:27, 19 February 2007 (CST)