Forum


The JeremyS Manifesto - For Ryan and Others Who Care - The state of the new Kdice.
JeremyS wrote
at 6:52 AM, Sunday January 6, 2008 EST
Let me first start that I've enjoyed playing kdice, and I'm not bitter and angry towards anyone, espcially Ryan... I hope that what I have to say will be taken in the spirit of me offering advice as a gamesman rather than an angry "I quit" rant, because, quite frankly, I'm not angry with anyone, and I'm not sure I'm quitting. I am going to take a long and detailed look here at what I perceive to be serious flaws in the new system, but only because I want to see the game continue to improve.

Let me next say that I enjoyed playing this game, especially the game I experienced in December, when I broke into the "1k table" experience and got to meet and compete against the currently active elite players.

And, to be honest, I play games to experience them at the elite level. I hope that doesn't come off as arrogant... what I mean is that I don't judge a game so much by its first blush experience, or the game experience you get hanging out with newbies and bad players, but rather the continuing value it offers to long term players who are proficient.

As such, I spent awhile looking at the top end game of kdice last month, both getting to know the regular players in order to get a feel for how the community was balanced, and looking at what it took to finish at the top.

It was apparent to me fairly quickly that all the complaints about PGA at the 1k tables are whiny nonsesnse. As I was new to the game at the top level and had no allies to speak of, I was in a good position to judge what the truth of these rumours are. The truth is, if you play a fair and balanced game, act friendly, and explain your actions in a reasonably articulate fashion, you will be welcomed by the top tier. Maybe in the days before I started playing, this wasn't true, and there was lots of PGA and cheating, and the top tables got a well deserved bad reputation, but it wasn't true last month.

I also noticed that there was a pretty good balance in terms of rewarding skilled players.

However, under the new system, the balance appears to be gone, and that creates a few distinct problem that leads me to believe I am abandoning the game.

One quick definition: When I say old system, I'm talking about the system as it existed from October to December. I never played in the ELO days and they are irrelevant to this discussion.

There are two big problems under the current system, and believe it or not, I don't think the flagging system is one of them (although it has its quirks as well).

The first problem is that games are no longer a zero sum equation.

I'm not sure if they were precisely zero sum under the old system or not, I didn't look extremely closely at the formulas, but I believe they were close. Basically, the only thing that kept players points across the entire system from adding up to zero total was the fact that players couldn't get negative points. All the points that players had were created at the 0 tables, and gradually trickled up the ladder as players trying to play at new levels donated points to more experienced ones.

Under the new system, each game, at every level, generates more than zero points. In fact, I tracked the points at a 1k table, writing down exactly what everyone got, and it turns out that in each game, the sum total of points earned was EXACTLY 576 points each game, which averages out to just over +82 per player. At the 250 tables, it is exactly 1/4th the score, or 142 points per game added to the pool.

What difference does this make? It means that you can play BELOW AVERAGE at the 1k tables and actually still come out ahead.

Let's consider a player who isn't that great. Maybe at the 250 tables, he does exactly average, or about +20 points a game. Now, at the 1k tables, he does below average, and gets only half of what the total player average is per game. He still gains 40 points per game. So basically, this theoretical mediocre guy gains twice the points per game by doing poorly at 1k than he does by playing averagely at 250!

Before, sitting down at the 1k tables was a scary proposition. You stood to gain a lot, but you also stood to lose a lot. Now, because you're gaining points per game for mediocre play, it actually makes sense to sit at the highest possible table because even if you do somewhat below average, you're going to gain points faster than you would doing somewhat above average at a 200 table.

Our same theroetical average joe under the old system wouldn't have ever made it to the 1k tables in the first place, because by doing exactly average at the 200 tables, his score would remain approximately 200. If he did get to the 1k tables (because scores DO fluctuate), he would have gotten crushed quickly and sent back down to the 200 tables where one would presume he would stay at approximately the same score for the rest of the month.

This is not a problem with points inflation. If scores were simply 5x higher, then the 5k tables would be the new 1k tables, and 1k would be the new 200, and life would continue as it had been. However, the reality now is that the high points tables will not be filled by a small group of quality players, but rather a larger quantity of players who are merely dedicated/addicted. The elite experience of kdice is just going to simply vanish, and mediocrity will spread far and wide.

That's not to say that the best players won't be well represented in the top 25, and playing at the highest tables. I'm sure they will be, because they do far better than average AND play a good bit. But the days when the 1k tables were filled with a small cadre of great players, where there was close to a guarantee that you were playing against the best players in at least 5-6 of the spots on the table, are simply gone. With the pool of players available and willing to play at 1k (remember, our average joe is BETTER REWARDED for playing at 1k than 250), the talent level at the tables is simply diluted.

It gets worse as time goes on. Because average play gains points, that means that the average player will be at 1k fairly quickly, and the talent level at the 1k table will get lower and lower, which means that the average player will start to succeed more, accelerating the problem. I wouldn't be surprised to discover that there will be more 1k games being played than 250 games by the end of the month, and it's entirely possible that the 5k tables will be flooded!

Consider this, we are only four days into the month and already the entire leaderboard is over 5k. Last month, there were only a few 5k games and they were super high-stakes. By the middle of the month, the people at the bottom of the leaderboard will be somewhere in the range of 20k points, and a 5k loss won't even be a big deal.

This last part of my first point may come off sounding arrogant, and I apologize if it does, but it's not intended in that fashion. One of the biggest reasons that I enjoyed playing with the best players at the 1k table last month was because the amount of irrational stuff that happens with those players goes way down. At the 200 tables you constantly saw players do weird stuff, or get vindictive and screw someone even though it cost them any chance of winning as well. At the 1k tables, that element was missing, but not this month. I had someone earlier today with no chance of connecting and with his base on the other side of the board just randomly split me from an island and then restack huge in my middle... he apologized after it happened, but it ruined both of our games. I checked his profile and he is a career sub-1k player before this month (he has 2600 points right now). While in the long term bad play rewards other players, I'd prefer not to even play with people who make those decisions.

The second problem has to do with the new dominance system and the number of starting dice. I think the idea behind the new system - eliminate farming - is great. However, in practice it has created a big problem:

Basically, in the old system, dominance didn't start calculating until after your third move, and people almost always started with a five stack.

That meant that throwing your dice all over the board on turn one was a really bad idea, and people basically called it suiciding. Suiciding across the board usually meant that a couple of players would each take one or two territories from you in establishing their bases cautiously, and you were going to be in terrible shape.

Under the new system, the equation is completely different. First, because five stacks are far less frequent, someone who suicides and has an even-ish build (mostly 2s and 3s) may well wind up holding a large number of the territories he jumped on, because the other players simply can't punish him as effectively.

In addition, the risk/reward balance is totally different.

First, because dominance starts calculating immediately, someone who gets a lot of territories fast and then gets wiped out fast can actually wind up not losing all that much - I saw a 7th place turn 2 finish where the 7th place player wound up with POSITIVE points.

Second, because it's far more likely to actually hold a significant amount of the territory, the player who can suicide best on turn 1 wins an awful lot. I have seen quite a few games essentially decided on turn one, or come down to 2-3 players after turn 1.

Third, because there is no longer a zero sum equation on the board, even if you don't win, you're penalized a lot less for losing.

Last month, when you saw someone going 2v2 or 3v3 on turn one, it was the bold move of a person deseperate to get a consolidated position or a connect. Now, it happens constantly on the first turn simply because the risk of losing dice on turn one is laregly ameliorated by the fact that players are unlikely to be able to punish you for having few dice and the reward increased dominance brings.

What does it add up to? People have become super aggressive on turn one, and instead of games coming down to who can consolidate a well defended position early and then expand and negotiate, the games seem to come down to luck on the first turn rolls, and even the unlucky aren't too bad off thanks to every board being a huge bonus to points.

So, that's why thoughts about the new system. It's probably going to drive me into retirement, because the game has basically gotten a lot less fun... there's no longer any sense of stepping above the pack and playing with the big boys, because the big boys are playing at the same tables as everyone else now. I hope everyone takes my thoughts into consideration for the long term improvement of the game. I don't want people to think that I can't hack it with the new system (I was on the leaderboard the first few days but haven't played as much the past couple), and I don't want to people to think that I reject these changes just because they are changes... I am fine with change as long as it's for the better, I just don't think that that is the case here.

I am also not advocating returning to the December system. I think, however, that a "new-new system" that seeks to end farming but addresses the concerns I've raised here will go a long way towards making kdice a balanced and fun game for everyone.

Thank you for your time.

Replies 1 - 10 of 40 Next › Last »
JeremyS wrote
at 6:55 AM, Sunday January 6, 2008 EST
Typo correction - 250 tables add 144 points to the pool, not 142.
Danny_DCB wrote
at 7:14 AM, Sunday January 6, 2008 EST
If I was a troll I'd say TLDR!!!11oneone

Because I'm not I'll say it's too long and I didn't read it.

Can you make an excerpt that would be readable for such a lazy guy as me?
leekstep wrote
at 8:40 AM, Sunday January 6, 2008 EST
This is the most thoughtful and informative and important post I have read in a long time!

I agree that these changes should be corrected as soon as possible.

1) Zero Sum
I didnt recognize the problems of positive sum games rewarding mediocrity until reading your post. In addition to this extremely valid and disturbing fact, points escalation has the effect of diminishing the weight of career stats.

I know Ryan introduced extra points to make the game easier for n00bs who were frustrated at always having zero points. A correction for this problem might be having small net positive points awarded at 0 and 10 tables, with the higher tables returning to zero net system. PLEASE IMPLEMENT CORRECTIONS IMMEDIATELY!

2- Starting Dice
Jeremy has described the problem very well. The game is now all about getting a lucky start. My research is only anecdotal, but I believe skilled players who typically used to win nearly 20% of their games are now winning less games.

I know Ryan made a formula change to accomodate different maps sizes and configurations. The fact is that the new game blows, the game is entirely determined by being the one guy starting with a 5 stack or going early and getting lucky with 3v3 connections. The new game totally blows!

The corrections should not be too difficult. Give everyone one additional die to start with, or return to the old starting forumla. PLEASE IMPLEMENT CORRECTIONS IMMEDIATELY!!

3- Dom calculations and flagging.
I understand what Jeremy is saying. I havent met anyone who fully understands the nuances ot the new dominance and flagging rules, but I like them. No farming is cool! And although it stinks not being able to flag before you bleed your dom away if you arent in last place, I guess it is up to the user to learn new strategies to maxumize points.

Keep the new scoring and flagging system!

In conclusion, it should be apparent to everyone by this point that 1) inflated net sum games, and 2) not enough starting dice are horrible horrible problems and should be corrected immediately.
leekstep wrote
at 9:21 AM, Sunday January 6, 2008 EST
I just realized that if the net points are over +500 at the 1k tables, then it will be +2500 at the 5k tables and +5000 at the 10k tables.

I thought I was brilliant when I predicted a few hours after the month started that it would take 80,000-120,000 pts to win the month.

Now I realize I was way off, it will probably take 500,000 pts to win the month.
Wicked! wrote
at 10:48 AM, Sunday January 6, 2008 EST
Bravo.
I did notice something fishy with the 0 point balance as well, but never calculated it.

I don't think you should quit, and I know you said you aren't certain, but still.

In November I asked Ryan if there were going to be 5k tables and he said that there weren't going to be any....
Do you see any 5k tables?
There ought to be.
JDizzle787 wrote
at 11:13 AM, Sunday January 6, 2008 EST
Very nice and elaborate post there Jeremy. Its nice that you can calmly speak out on this stuff.



In my opinion, it's really difficult to get anywhere in the first round. I don't suicide though, since it's never worked for me. I nearly threw my monitor out the window yesterday for the shitty starts I was receiving. I was thinking and planning my rolls, and crucial, progressive moves never won, and I never had a secure grouping of dice.

I think it's an awesome idea by Ryan to start everyone off more evenly, since no one is left disadvantaged, but it's not working this way. Too little dice for the amount of action going on. With no 5 stacks in the first round, it's very hard to get any where or connect at all.

leekstep wrote
at 11:24 AM, Sunday January 6, 2008 EST
Ryan thanks for the fixes! We need that extra +1 dice to start.

You are the greatest inventor of fun in our generation!
JeremyS wrote
at 12:02 PM, Sunday January 6, 2008 EST
Where did Ryan say he was adding +1 dice per table?
JDizzle787 wrote
at 12:07 PM, Sunday January 6, 2008 EST
It appears that during my first few games today, I noticed how much easier it was to start a game, due to what seemed like more dice. Dunno if it's confirmed or just things evening out.
Wicked! wrote
at 12:13 PM, Sunday January 6, 2008 EST
Ryan was lurking around at a 1k table and Leek suggested it. Ryan without missing a beat said that he'd add them next update. :D

♥
KDice - Multiplayer Dice War
KDice is a multiplayer strategy online game played in monthly competitions. It's like Risk. The goal is to win every territory on the map.
CREATED BY RYAN © 2006 - 2026
GAMES
G GPokr
Texas Holdem Poker
K KDice
Online Strategy
X XSketch
Online Pictionary