account, im very open to questions and new ideas, and i'm big into making this a better game

I put in over 100 games on the test server just to help check if this new scoring thing worked, and I gave plenty of advice and helped whenever I could.

I don't need a #1 place to validate me or anything, I play the game cuz I love it and I like the community of players.

To answer one of your questions tho, when you discuss playing a lot of games. There is indeed a reward for playing a lot of games. But here's the key. You have to play a lot of games while MAINTAINING a high elo. (meaning 1900+) or 2000+ and finishing in the top 100 of all players.

That way, playing a bunch of games does help, if you are adding 0.01, 0.02, 0.10, 0.01, etc to your rank. But the suggestion that someone needs thousands of games is silly. Think of it this way. How many games do I have vs how high is my rank?


589 games to 34.11 rank (score)

Look at Wicked!

she has 116 games, 8.08

It didn't take her thousands of games to get on the top 25. She played 116 games, i'd say a lot of players here are addicted enough to post that many games up

wow, I just looked at your games played to compare, 1150? jeez! hehe And I thought I played too much :)


Point is, you dont need to play thousands of games to get on the top 25.

ah my bad, I forgot u did the phoenix project. nevermind

But my point was simply this. Playing a lot is useless. Playing up to high elo, however, is completely beneficial, when it comes to climbing up spots.

Account, here's one of the proposals I made to Ryan, and let me know if you think it is any good.


I suggested we do monthly competitions. So every month, reset the rank to 0. At the end of the month, take a screenshot of the top 25. Then create a link (on the top left) and label it Past Winners by Month. And then have a link u can click on. April's top 25 list. May's top 25 list. Etc.

It would be good on a couple of reasons. For players that make the top25 they get to see their past accomplishments if they want. It creates a sort of history of kdice so that we can see how players have done in the long run. It always is good in that every month a score reset would allow everyone to have a shot at the top 25 again, as well.


Last point - Account, I think the new scoring thing is very fair. However, after a while it does get to be out of reach for newcomers if the rank values arnt reset every so often, and I was on of the biggest proponents of a score wipe, so take it easy on me will ya? heh

...wasn't playing up to high elo already beneficial...?

Yeah, if you were to reset both the rank and the ELO, I agree that there would need to be some sort of waiting period before rank points began to be counted. Maybe after a week, maybe when 1600 or 1700 tables open, I'm not sure... but some time for a hierarchy to establish itself would be good.

Another thought just crossed my mind after rereading account3x's post. There is a certain bias towards players who play much more often. I'm just looking at the top 25, and I see Wicked! and riser both have pretty phenomenal win percentages, but just over 100 games played. Then just above each of them you have hatty and Canarioz who have both played over 400 games.

Now, hatty and Canarioz are both excellent players, but it seems fairly clear that if Wicked and riser and had played 400 games each, they would certainly be above hatty and Canarioz. The system is rewarding somewhat too heavily for those extra games, I think. Maybe this is a decent solution, maybe it overcomplicates things, but off the top of my head I thought it might make the rankings a little more accurate (and a little easier for new players to climb ranks) if you modified the final ranking based on the number of games played. I'm not suggesting something like rank/games, because then it favors players who had a lucky streak in a small number of games.

I think though, that dividing the rank points by, say, the square root of games played gives a more accurate ranking. This way, those who do very well in less games such as riser and Wicked! will fare somewhat better, and at the same time it won't penalize those who play lots of games too heavily, but it should keep them from gaining an insurmountable lead.

Of course, since it seems a lot of people are having trouble grasping the way the scoring works now, adding one more level of complexity might cause more trouble than it's worth, but hey, it's an idea.

I can’t help but have to post one more time:

“My only concern is that whenever you set the rank AND elo back to 0 and 1500, is that whoever plays the first game and wins gets 1 rank added to their score; That's not a real display of talent, so much as it is a display of 'I played and finished a game in first before anyone else finished a game'�

if the argument about players who cannot achieve high elo even if they play 14000 games stands, then you are contradicting yourself. Fine, the very first player will get 1st rank, but if that player doesn’t have the skills, he/she will eventually be outstripped by the great ones like yourself. So I really don’t see the concern here.

if you think this is a concern, you are in a way agreeing with me that once some player achieved a high elo, the later comers will have a hard time to catch up.

Just like gpokr, after the reset, say the very first player won over $200,000 with his/her initial $1500. it is of course hard for the others who start later and with $1500 to catch up. But, if the very first player has no skills, he/she will eventually lose the $200,000 and the late comers, if they have skills, with eventually win and pile up to have more than $200,000. make sense? They are both games about skills and luck so I don’t think the comparison is off base. I know someone will argue that gpokr don’t use elo, but all the arguments here is centered around SKILLED PLAYERS WILL BE ABLE TO MOVE UP AND BE #1. and in that aspect, I don’t see how elo would make kdice different.

I agree it may not be a real display of talent INITALLY. But in the long run, over the course of the month, I truly believe those really great players will make it to the top.

I believe the TRUE top players will be at the top even when we reset EVERYTHING. How about test it in a sandbox?

the more I read you post, the more I feel I am right about people being afraid to lose their high rank

“Point is, you don’t need to play thousands of games to get on the top 25.�

But you do have to play hundreds of games to be in the top. Things get harder when the top players keep on playing and pegged their elo, score, rank, whatever.

Playing a lot is not useless. I have a 0.10 when some 1600+ players have only 0.00 or 0.01

by the way, I also played hundred of games in the sandbox. I would have played thousands but I actually have a decent job.

Also, I am not exactly in the phoenix contest. In a way, I was there to prove that you can lose and lose and have a low score but regardless, the more your play, the higher the rank you are. Look at my score and rank - 655th(0.10) , 1428(12082nd) games played 1150. I am a living proof that the system is somewhat flawed. I am not suggesting everyone go out and play thousands of games to gain 0.0001, but you know what I am saying, right?

“Account, here's one of the proposals I made to Ryan, and let me know if you think it is any good.�

i am not asking everyone to agree with me. No matter what happens and what ryan decided to do, I’d still come and play. I am just here to voice my opinion and really no one has to listen. I know I tend to be somewhat of a devil’s advocate sometimes (or all the time). I may sound like a pain but I really enjoy coming here. anyway,

I hope that everyone who cares come and reply to this post, so that we have a more all-rounded opinions from everyone, not just the top players.

Just gonna reply on one part. I do feel it is a legitimate concern that whoever plays 1 game and ends first (after a complete reset) gets 1/1 = 1 added to their rank.

Here's a scenario.

A player wins 3 firsts in a row, finishes 3rd, then gets another 2 and a 1st.

If that player was the first to finish playing that game before any other game was completed, he has 1 rank (highest elo). Then again then again. A small luck burst of 5 or 6 games could equate to 5 or 6 rank score.

Looking at the top 25 now, 6 rank points would have someone in 21st place or so. And that is now, at the END of a month. So yes, I do feel it is a legitimate concern.

With so many players, assuming that someone will win 5 or 6 games with a hot streak after a reset is not unheard of - and if that player ultimately has poor elo and isn't that great, but had that one good run, I do feel is it not fair to other great players who climb to 1700, 1800, 1900, 2000 elo later on or even higher, but cannot overcome what that one person got with a short burst to climb into 1690 or so elo, then back down to 1500-1600 where they hover normally.

As for the more games played part, yes, you do add 0.01 or so values every so often. But again... someone can climb to 2000+ elo and win one game, come in at the 10th best elo, and get 0.10 in just one game. So really, it's all about rating (elo) when it comes to rank. Games played might boost the value on the right of the decimal a little bit 0.00

but it wont really help to move the number on the left of the decimal, if u catch my drift

“but cannot overcome what that one person got with a short burst to climb into 1690 or so elo, then back down to 1500-1600 where they hover normally.�

Applying this logic to the current situation – you can see why I say it’s hard for people like me to crack to the top 25. In fact, what you said totally validate my point. Thank you.

As for the number of games played – it does play a role in your elo, rank, score (I am still confused as to which is what. How about we call them Apples, Oranges, and Banana?). So it’s not ALL about rating, even the # of games played perhaps only play and small and insignificant role.

It's very sad that after 3 weeks so many people don't understand the ELO rating vs score (earned points) difference.

A very clear explanations can be found at the following links. *Warning* these links are from the unofficial wiki.
<a href="http://kdice.wikispaces.com/Ranking+System">Rank from score</a>
<a href="http://kdice.wikispaces.com/Scoring+System">ELO rating</a>

Note: the news page shows top 5 players rank_from_ELO_rating, while the top players page shows top 25 rank_from_score.

Regarding the reset. I think the ELO rating should be preserved. Why force decent players to go to lower tables? I'm not worried about the initial values, e.g first ELO gets 1, second .5 etc. In 3 weeks people earned a lot of points, so that initial advantage was not as big as people thought at the time.

Yes with the new rank from score, players that play more have an advantage. However, if you play many games at at low ELO you don't really get that many points, e.g. accountx3: after 1150 games earned 0.10 score points.

"""*Warning* these links are from the unofficial wiki."""

This webpage is just as "official"/"unofficial" as every Website on this internet..