[Poll] Points Inflation - Guess how much

[Blitzaholic]

EXAMPLE: If you vote 51 to 75, that means you think the average score on CC is 1051-1075.

it is probably around 1100 or 1200 range for average, but its not that big of a deal if you ask me, it certainly isnt helping me out where at the end of 2008 I was 4307 score and now several months later i am hovering around a 2400 to 2500 score, I dropped about 2000 points and most were lost to those with scores in the 700 to 1300 score range, mostly 1 vs 1;s in tourney games, even If I won 4 of 5, I still was losing points early on, not as much when my score dropped, but, I dont play for points, at least I havent been in the last 6 months to a year. I am a medal lover of late. I am not as concerned with this as I am in freestlye speed with players who have clickable maps and others tools that most on the site do not have, (especially new players to the site, and new being even players on for 6 months of less) it is the easiest exploit to gain points. I am ok with speed, I am ok with freestyle (well not really, but I will be nice) but I am not ok with freestyle speed, that combined is a major manipulation of scores, but I am trying to accept this. it is a process.


all should play on an even and fair playing field and this is not happening with many freestyle games and freestyle speed games.

[owenshooter]

all should play on an even and fair playing field and this is not happening with many freestyle games and freestyle speed games.

to me this is the most obvious reason being avoided, as freestyle seems like a cheat fest where points are easily gained through the use of add ons that many people in the CC community know nothing about, since less than 3% ever visit the forums. not to mention you have guys running scripts that take their turns, etc... i don't like freestyle, i avoid freestyle at all costs, and i think that the freestyle setting in particular, belongs in this conversation. even a mediocre freestyle player has a more significant chance of trouncing a far better player unaccustomed to the setting and that does not run clickable maps, etc... nice point blitz, glad to see someone else thinks this... freestyle is the big pink elephant in the corner that nobody is discussing in this topic...-0

[Bruceswar]

all should play on an even and fair playing field and this is not happening with many freestyle games and freestyle speed games.

to me this is the most obvious reason being avoided, as freestyle seems like a cheat fest where points are easily gained through the use of add ons that many people in the CC community know nothing about, since less than 3% ever visit the forums. not to mention you have guys running scripts that take their turns, etc... i don't like freestyle, i avoid freestyle at all costs, and i think that the freestyle setting in particular, belongs in this conversation. even a mediocre freestyle player has a more significant chance of trouncing a far better player unaccustomed to the setting and that does not run clickable maps, etc... nice point blitz, glad to see someone else thinks this... freestyle is the big pink elephant in the corner that nobody is discussing in this topic...-0

I play speed freestyle. Honestly when an 8 man speed fs esc game comes up... 5 or 6 of the 8 players use clickable maps. It is not as big of a problem as people make it out to be. It is the use of it that is more a problem. People do not understand how to properly use it. Speed kills. Just like driving fast you have to know how to handle the car. Manimal was a perfect example of how speed was not winning him points. He was at a time the fastest player on CC, yet could never get past major. Even if you take away Clickable maps the best players will still have an advantage in speed. A bigger issue if you ask me is 1 vs 1 games of any type. More farming goes on there than any other game type. Look at what max did, or for that matter anybody now who claims... I am not playing ?'s I am playing other ranks.. AKA cooks and the like. At either rate Freestyle is not the problem.

[Timminz]

A bigger issue if you ask me is 1 vs 1 games of any type. More farming goes on there than any other game type. Look at what max did, or for that matter anybody now who claims... I am not playing ?'s I am playing other ranks.. AKA cooks and the like. At either rate Freestyle is not the problem.

Sequential 1v1 vs. a low rank, is a LIKELY win. Freestyle vs a low rank, is all but GUARANTEED.

[owenshooter]

A bigger issue if you ask me is 1 vs 1 games of any type. More farming goes on there than any other game type. Look at what max did, or for that matter anybody now who claims... I am not playing ?'s I am playing other ranks.. AKA cooks and the like. At either rate Freestyle is not the problem.

Sequential 1v1 vs. a low rank, is a LIKELY win. Freestyle vs a low rank, is all but GUARANTEED.

hey now, timmy... settle down.. bruce said freestyle is not the problem, so it's not the problem... ahem... again, this should be a very large portion of the conversation, as it is a very real problem that is tied to this discussion...-0

[Artimis]

Code out the ability to side step the time delay and immediately start again in casual freestyle games by running out the clock and you'll fix the biggest problem with Freestyle games. I keep posting this topic URL(Eliminate Freestyle Double Turns) in the Freestyle City Mogul games I'm playing trying to get players to voice their opinion on it, but so far of all the players I've smashed over the head with this cheap tactic only one bothered to post in that thread. Getting players to engage is proving a real frustration.

Integrate the Clickable Maps addon into the site so that it is available to ALL players and you'll level the playing field, fixing the second biggest problem with freestyle games. You then have the choice of using it or sticking to the drop down menus at your discretion. I don't think Yeti would object to this and would likely welcome the extra programmer input from the site admins on CM.

[Bruceswar]

A bigger issue if you ask me is 1 vs 1 games of any type. More farming goes on there than any other game type. Look at what max did, or for that matter anybody now who claims... I am not playing ?'s I am playing other ranks.. AKA cooks and the like. At either rate Freestyle is not the problem.

Sequential 1v1 vs. a low rank, is a LIKELY win. Freestyle vs a low rank, is all but GUARANTEED.

Here we are back to 1 vs 1 again.... and I said... any type...meaning freestyle as well. BTW you can lose FS or SEQ 1 vs 1 vs any rank.

[e_i_pi]

I like how the thread goes off topic the moment the results are posted. First we have admins... sorry... a volunteer and an admin... talking about how the site doesn't need to change, then how much it has changed, now we have the ongoing feud over freestyle.

I might make a Stop Double Turns Freestyle thread, and try to steer it towards solving the points inflation problem...

[Bruceswar]

The problem will not be solved until they get serious about it. That might happen in 2020?

[Timminz]

A bigger issue if you ask me is 1 vs 1 games of any type. More farming goes on there than any other game type. Look at what max did, or for that matter anybody now who claims... I am not playing ?'s I am playing other ranks.. AKA cooks and the like. At either rate Freestyle is not the problem.

Sequential 1v1 vs. a low rank, is a LIKELY win. Freestyle vs a low rank, is all but GUARANTEED.

Here we are back to 1 vs 1 again.... and I said... any type...meaning freestyle as well. BTW you can lose FS or SEQ 1 vs 1 vs any rank.

If you re-read what I wrote, you will see that we agree on most points here. The only thing I would say, and that you would seem to disagree with, is that freestyle makes it FAR more likely for the experienced player to take advantage of an inexperienced one than sequential does.

To make an example, I have just map ranked a player who is known by some, as a pioneer of 1v1 farming. Note that this particular player has thousands of games played in both these categories.

sequential = 60%
freestyle = 92%

I challenge anyone to find a player with a 92% win rate in sequential 1v1. It's just not possible.

[Timminz]

I like how the thread goes off topic the moment the results are posted.

Right, you are.

Sometimes it's a little too easy to respond to a post, without thinking about what the thread topic is.

[lancehoch]

So, how many extra points are on the scoreboard? 4,141,896 give or take a few thousand depending on how you count New Recruits? Over the lifespan of CC (1,234 days as of May 18th) that works out to be 3,356.5 points per day. If a New Recruit drops to 975 before they leave, on average, then this means that about 135 New Recruits are joining per day and then leaving. I still think that this number is a little high.

[Timminz]

So, how many extra points are on the scoreboard? 4,141,896 give or take a few thousand depending on how you count New Recruits? Over the lifespan of CC (1,234 days as of May 18th) that works out to be 3,356.5 points per day. If a New Recruit drops to 975 before they leave, on average, then this means that about 135 New Recruits are joining per day and then leaving. I still think that this number is a little high.

I think the bold part is where your estimate is off. I would guess that the average new recruit losses at least twice that much.

[e_i_pi]

So, how many extra points are on the scoreboard? 4,141,896 give or take a few thousand depending on how you count New Recruits? Over the lifespan of CC (1,234 days as of May 18th) that works out to be 3,356.5 points per day. If a New Recruit drops to 975 before they leave, on average, then this means that about 135 New Recruits are joining per day and then leaving. I still think that this number is a little high.

I think the bold part is where your estimate is off. I would guess that the average new recruit losses at least twice that much.

Okay, I was being conservative when I quoted 975. I simply took the scores of every New Recruit currently on the board, and divided by the number of New Recruits. So this isn't counting all the games they're in and are going to lose by deadbeating / poor play / bad luck (also doesn't count the one's they'll win). New Recruits on the board are on average 975. I can't get stats for those that have left and never come back, without putting intense load on the server.

[sully800]

I've said it before and I'll say it again: There's a simple solution to point inflation that will also dramatically reduce the impact of noob farming.

Change the score formula to loser/winner*20 - 1, but subtract the same amount of points from the losers.

This means that if 6 people are ranked 1000 and they play a game together, each player loses 20 points but the winner gains only 95 (instead of the current 100)

In effect you subtract 1 point from circulation for every opponent that is defeated (please note that I haven't checked if 1 point is appropriate, we could use fractions of a point to make the system more accurate). This could effectively, stop or even reverse inflation depending on how many points are subtracted.

The reason it also protects us from farming: If you beat five of your peers you would get 95 points instead of 100. A reduction of only 5%

If you beat five people ranked well below your score you would be impacted much more. If a 4000 point player beats 5 people each at 800 points he would normally gain 20 points and under the new system would gain 15 points. A reduction of 25%.

This all means that people would be encouraged to play others near their own rank, because the rewards for playing lower ranks would be much smaller relatively, so gaining points from lower ranks would become much harder.

[AndyDufresne]

Interesting idea Sully---what discussion can we get about the actual value of the subtraction?


--Andy

[sully800]

Well we can figure out the value based on the data e_i_pi collected and posted in this very thread.

What we currently have is the total extra points that have been added into the system. What we need is the total number of defeats that have taken place on CC since 2006. Divide the extra points by the number of defeats and you have the points that should be subtracted per defeat.

Simplistic estimate: Let's say each game has an average of 4 players. That would mean each game has 3 defeats. 3 defeats * 4118643 games = 12355929 total defeats.

(1183.13-1000)*22617 active members = 4141851.21 extra points on the scoreboard.

extra points / total defeats ~ 0.333 extra points per defeat.

So really you should subtract 1/3 of a point for each defeat to maintain the current point level. Or you could subtract 1 point or more per defeat until the average user was back to 1000 points and then switch to the 1/3 of a point system.

A more accurate estimate of the total defeats could be be made if it is deemed worthwhile, but I think the current numbers are a good "schematic" type of estimate.

[Fireside Poet]

...and how does terminator games get into this equation where it is possible to gain points and lose the game?

[john9blue]

...and how does terminator games get into this equation where it is possible to gain points and lose the game?

Just subtract one point for each elimination that takes place in a game. The game log would read:

A eliminated B from the game
B loses x points
A gains x-1 points

There ya go.

[Kotaro]

Oh damn. A 1 point loss. How in the world will anyone n00b farming ever recuperate from that.

[e_i_pi]

Sully, your solution is good, but it doesn't quite hit the nail on the head. I applaud you for it though, because it does kill two flu-bearing animals with one palm-sized hard thing. There's a neater solution though:

Set up a CRON process that runs once per week or month. The process totals the points on the scoreboard, totals the players, and the above/below 1000 points gets added/subtracted from every current players score.

It's a simple solution, requires zero maintenance, and would take maybe 30 minutes to code. It is automated, accurate, and unbiased.

[sully800]

Oh damn. A 1 point loss. How in the world will anyone n00b farming ever recuperate from that.

Noob farmers typically gain 1-5 points from their opponents and lose ~100.

If you take every 5 points win and make it a 4 point win they suddenly need to win 125% as often to maintain their previous rank. That wouldn't be possible so ranks would drop.

This would affect you much more when you play people who are drastically below your current point level.

@FP - Terminator games are easy because the points are already calculated for each defeat. No extra work required! Team games are also covered, because you would take the total scores, determine the ratio and then subtract one point, just as with singles.

[john9blue]

Sully indicated that 1 point might be "overkill", but the lower scores will balance it out in the long run.

I think it's a good idea, and should be given serious consideration unless you can produce figures that show otherwise.

[Artimis]

I think this thread is most appropriate for where this discussion is heading, in particular, this post by Cicero: stop inflation

[jpcloet]

It is possible, just unlikely.