After two years...

So know i played two years with the Nicks opmp and Jonny.

With Jonny i played in the last years nearly only championships and rating tournament.

And with opmp only monthly cups.

To see if it is better to play against same rated or mixed players.

Now my rating is with both Nicks like the same.

Next i will play with Jonny long games like 5 and 7 Point and with opmp just 1 point games.

I think that the noise floor is so high that your sample isn't big enough. I think did write a script to play different parametrisations of RoRoRo's engine against itself. I could adapt that so that some participants were never matched with vastly differently rated opponents, and play millions of games, I guess, and see whether there is a difference to be measured or not.

For your next expleriment, my prediction is that there'll be more noise in opmp, and less noise in Jonny.

Game #1124070

This is how close a 7 match game can get

@ phil: Yes i thing that too. But we will see

Within the last year I played nearly against everyone in a lot of 3 point matches and my rating was very low with approx. 1700 (averaged). But my rating is increasing since I'm playing more 5 or even 7 point matches. And I guess that the player strength (rating) is approx. the same. So I see an advantage in 5-7 point matches compared to the 3 point matches.

What is noticable is the winning rate of OPMP (62.0% )compared to Jonny (55.3%). So it looks that Jonny is playing against stronger players and it seems that this results in a higher rating (maybe ~ 1775 averaged about those 2 years compared to ~ 1725 after stabilization) even if more games was lost.

..were ;-)

The realy bad winnig rate of Jonny result from a drop down in the Year 2006. After a discussion about cheating i resign over 50 winnig games.

So the winnig rate had to check since the year 2007 to 2009 from opmp and Jonny. May be a program can do this. then we can say clearly whats going on.

I have a better experiment for you!

One ID plays only normal EWN, the other account plays only backwards capture?

A less luck-dominated game should permit ratings to extend over a wider range, which, as your're a good plaayer, implies that you should end up with a higher rating in the more skillful game.

I think.