How good am I? (V2.0) 3862. That is how good I am.

[juventino]

I decided to play some games on same maps and on same settings against opponents who choose to join my games. I made this in order to see how good I was against normal players.

The settings was 2 player games. Italy. Freestyle. Esc, chained.

I played the games as good as I could. With no freestyle crap. No missing turns or play at last second.

I have had played 50 games when I today did my research.


Wins 82 %
41 Games
Avarage gain: 10.15
Avarage opponent: 1159
Total gain: 416


Losses 18 %
9 Games
Avarage loss: 39
Avarage opponent: 1171
Total loss: 351

Gain over 50 games: 65 or 1.3 points per game.

With the same settings and same opponents and winning result:
The score when my avarage win or loss per game is 0.0

My break even score is:

2487



So... That is how good I am with these settings...

[rebelman]

could not fault your logic here but I have two observation to make about your settings freestyle even without the known tricks which you did not use is a game style that suits experienced players much more the same is even more through of escalating as a noob to this site that plays a mix of styles I was trying to get small continent bonuses in esc. games obviously i now know this is not the way to play esc. but it took me some time to realise this i have no doubt several of your opponents made similar mistakes. My final 2 observations are as follows up until recently there was a flaw in the city bonuses on the Italy map, I have no doubt you were aware of this and would have used it to your advantage whereas more inexperienced players may not be as aware. Finally I genuinely believe someone's score although a good indicator does not determine how good they truly are, the reasons for this are many and as they are discussed on numerous threads i will not go into them again here.

[juventino]

absolutly. I agree. That is why i said: This is how good I am with these settings and this map. With a larger sample and with more maps and gamestyles it could be more fair. but this is a 1v1 gametyp and then I think it is more fair then flat rate.

And i couldnt secure that someone used double truns against me. But it is a part of the game so ok if they did. I didnt.

And in a 1v1 I dont think esc is more unfair agaisnt new players then flat rate. the opposit almost

I jsut chose the italian map because it is a fun one.

I chosed to open the games to my self so everyone could join. Would be wrong to chose my opponents beacuse I think I would bo better agaisnt some specieall players then others


(and to your first point.. it is very good to try and take a small bonus on esc... esp in a 1v1.... and soemtimes even in 6ppl (even if it is not the most important thing))

[juventino]

and what is the flaw in city bonus? I dont know about this.

[rebelman]

and what is the flaw in city bonus? I dont know about this.

its fixed now but it was giving a much bigger bonus for holding multiple cities than it should have if yo did not know about it then please ignor that part of my reply.

[juventino]

ah.. nope.. i didt know about that part. I looked at some games and I didnt find i got it. But cool =) it is hard to make an XML flawless perhaps.

I might just add: I dont think 2487 is high or low. I just realized that that is my roof playign thi kind of games. I think that if i chose my own opponents i would perhaps get a bit higher.. I wil perhaps look in to htat next time

[juventino]

With an avarage opponent of 1161 you need a winning percent of 87% to break even at 3000.

and to rbreak even at 3500 you need 90%. So you will need a higher number then that to reach it.

Is it possible? playing fairplay in open and honest games?

[firth4eva]

With an avarage opponent of 1161 you need a winning percent of 87% to break even at 3000.

and to reach 3500 you need 90%.

Is it possible? playing fairplay in open and honest games?

nope. the fear of losing gets to you too much

[The1exile]

You'll probably rake in the feedback if you don't play the same people, though.

[juventino]

nah.. the fear of losing isnt the matter. As long as I am better player then my score i will gain points.

As for me as an example. I have less then my 2487 on italy map. So it would be smart of me to play more. But if i get 2500 i will lose points again. Just beacuse i am not good enough.

[firth4eva]

nah.. the fear of losing isnt the matter. As long as I am better player then my score i will gain points.

As for me as an example. I have less then my 2487 on italy map. So it would be smart of me to play more. But if i get 2500 i will lose points again. Just beacuse i am not good enough.

i know. i was taking the piss of comic boy

[Robinette]

Juv, you sound more like an accountant than a croupier

And I really like what you've done with the numbers here

I'd be really interested to see the #'s with 6p games
Using this kind of calc's, My score right now would be really bad... 23 straight losses in 6p esc.. (ugh) I think need to buy a potato peeler to hang out with the cooks

[sully800]

Nice thread Juve! I like your style, experimentation and analysis of results! It makes a lot of sense and its an interesting concept to test out. If you play enough games in enough different styles you could definitely find your "best" settings and maximize your eventual score. I like that you tested to see how much you could win without any cheap tricks that usually end players at the top of the scoreboard.

My only problem with this is that its a relatively small sample, and therefore might not be accurate. Of course that means your break even score could be lower or higher than reported, so I'm not trying to say that you aren't that good. It would be interesting to see if these results continued to hold fairly constant for another 50 or 100 games.

[juventino]

Juv, you sound more like an accountant than a croupier

I actually studied for ½ year with the accountants...


I have done some more research about this. Yet again only 1v1 games since they are often the quickest game type.

I have come to the conclussion that this would be a fun feature to the scoring system. Then you can see who is the best 2ppl, 3ppl, 4ppl, 5ppl, 6ppl, trips or doubles player...

You can have a limit of games. Min 100 perhaps. And you can have "latest 100 games" to see who has a good form for the moment. (100 games can be what ever number you like)

[Coleman]

You look outrageously familiar.

I've been to Sweden by the way, this is not outside the realm of possibility.

[juventino]

ever been to a casino in Sweden?. I guess you have seen me then....

and.. we have a waying in sweden. "everybody knows the monkey but the monkey knows nobody"

[Aerial Attack]

I guess the real question is how many people only play 1 specific set of options on a specific map or two.

I'll post my game counts at every option/map later.

EDIT: Also, there is a potential flaw with your mathematics.

Unless you immediately clicked on their profile as soon as your game ended - you have an incorrect score for your opponent. If you just ran 50 games and calculated the avg score after everything was complete - then based on your win % your opps avg score was actually higher in your wins and lower in your losses.

An easy solution to this would be the following change to the Game Log (archiving of Ranks).

Instead of

"Aerial Attack loses 13 points
Juvento gains 13 points"

It should be:

"Aerial Attack loses 13 points from 1940
Juvento gains 13 points added to 2474"

[juventino]

I know that is a strange way to count. But i just wanted a fresh start. I have done this with 50 games where I join some games and thefore chosed my own opponenst. And I think I will get a better score then. But funny to see I chosed to play with lower ranked players then the players who joined my games.

But it is not my biggest point. I just think this is a fun way to count points. To see how really good I am at game type have I done it at many types of games now (with old games). 3 ppl, 4 ppl, 5 ppl and 6 ppl.

And... I must just say... I am goooood

of course just a joke... but i have always wanted to type with smal letters like this...

[juventino]

Instead of

"Aerial Attack loses 13 points
Juvento gains 13 points"

It should be:

"Aerial Attack loses 13 points from 1940
Juvento gains 13 points added to 2474"

But that is what is done. It is calculted from the start points. So after evry game I put it in to a form. That is why it will not be entirely correct to look at old games. I realized this half way into my experiment. It was actually a better idea to put them into the form directly after the game. Much easier

[Aerial Attack]

*lol*

Juv,

The part of the previous post that you quoted wasn't meant for you at all. It was actually meant to show lack a method of "archiving" player ranks at time of point distribution. Because if this was implemented, then all someone has to do is to look at each completed game once and gather all relevant information.

No, my quibble was that if you had waited until the 50th game was complete to then go and check everybody's scores, the first few players could have vastly different scores [currently] than when they completed the game against you.

Since you put all the scores in a spreadsheet at the start of each match, your margin for error decreased dramatically.

[juventino]

I have now done the second part. 50 1v1 games against opponents I chosed (i joined second). It looks like I prefer a bit lower ranked if you lcompare to those who joined my games.

Wins 94 %
47 Games
Avarage gain: 8.57
Avarage opponent: 995 (with low 625 and high 1799)
Total gain: 403


Losses 6 %
3 Games
Avarage loss: 53.67
Avarage opponent: 957 (with low 690 and high 1280)
Total loss: 161

Gain over 50 games: 242 or 4.84 points per game.


With the same settings and same opponents and winning result:
The score when my avarage win or loss per game is 0.0

My break even score is:

3862

Notes:
Number of ? played: 6
Number of times I said "gg, you win" and still won: 3


I admit I perhaps had a very good 50 games to pick from. That is why my score is so high. But I will continue to add games and I think it will be lower score when I will check it later. But this only proves it can be done to rach high scores with 1v1 and fair play and not (well. not so much) picking on new players

[chessplaya]

hmmm great job.....

u think u can get us a cure for cancer

hhehehehe just teasing...

but srsly good job with the numbers!

[juventino]

You can prove everything with numbers.... and I am the first to admit that they numbers are perhaps not fair. It will be fun to see the stats after 200 games or so.