Ratings Updated and top player list for kurnik

[Eric Strange]

We have updated ratings to May 14th 2010 and will be caught up to current before you know it. http://cross-board.com/tourneys/?page=2& and http://www.cross-board.com/ratings

My coder through this little gem together in a day and I think it is pretty cool. http://www.cross-board.com/kurnikratings/

I can add anyone's kurnik ID to this list and it updates their rating in real time based on the information in the ID's profile. After 14 days of inactivity it will automatically remove the ID from the list until they play again.
So this is an ACTIVE top player ratings list. We use karate belts for rating levels and even have an olympic winners podium for the top 3. It is a little cheesy but I think it is pretty cool!!! Currently Alex Moiseyev is #1 on the list but he will need to play a rated game soon to stay there!!!\

[Eric Strange]

he is 16th now.

dallas went from 1300's and has been studying like crazy and looking up his losses. Doing much better than he was a month ago. generally he is a low 1400 in kurnik

[waynegober]

Post pulled...

[Eric Strange]

well in showing the tournament it says that Justin gained over 200 points in that tournament. He was in database with his OLD calculated tournaments you told me about when he was younger.

I did find a problem though. When you told me he came in 6th I started sniffing around and found that the cross table Richard gave me had Justin at 15 points and the Cross table on nccheckers.org shows Justin with 17 points. I need to know which it is because in round 5 is where the difference is. NCcheckers shows Justin got 2 wins vs Dr. Robert Shuffett and the cross table Richard gave me shows 2 draws. If in fact Justin won both games in round 5 I can just setup a 1 round tournament and play those 2 against each other to fix it.


My system is not buggy though. I understand you are trying to get Justin to start playing again and his rating is an important factor. If he keeps playing the way he did he will be 2200+ in no time. I welcome you to calculate his rating at 1523 with the elo system and a K factor of 24 and see if you get very different. I would appreciate if you did not throw stones at my system because you don't like the outcome. everyones ratings are drastically changing because of this new system but it has been FULLY tested and is 100% good or I would not have even started calculating. This is the same system that online gaming sites use. So think about that.... if u play 16 games your not going to go from 1500 to 2000 or whatever you expected. The fact that he gained OVER 200 points was pretty impressive.

If the ACF exec. agree I can start Justin as a provisional 1600 for that tournament which can give him a K factor of 32 and not effect his opponents ratings. This would be sure to give him 300 points or so which would make him 1900. But I only calculate the ratings I am given and have no authority to adjust people just because I know they are better or worse. If so I would also like to adjust Dustin Belcher who was 1300's last I checked also because of tournaments he played in very early in his career. What can I say... you pay for your past.

[Chexhero]

Eric, Albert still has it I think lol. He bested me 2-1-1 last time we played. He is a very strong book player. I would be in the top 15 if it wasnt for his thrashing on me. Anyway, I like the top online ratings!

[liam stephens]

Well, you can all dismiss my claim (as seems to be the case) that calculating the ratings for 3 Move Tournaments on the basis of "Games points" instead of "Rounds points" makes a huge difference. The contention being that it has little or no effect. But, lets have a reality check.
The latest rating list above shows:
A Moiseyey 9th
S Mc Cosker 36th

Well, that ain't reality, it's a dream world.

Those two stick out like the proverbial "sore thumb" and there are many others that are similar.

As Alex might say - "The proof of the pudding is in the eating".

Incidentally on this new system, Alex will probably never have a chance of reaching Tinsley's 2800+ rating.

The COMPRESSION SYNDROME rules - OK.

[Eric Strange]

Yes rating are adjusting and GOING to be different than with the old system. I know people do not like change. But as soon as ratings are updated I have also incorporated a stricter rule on how often you must play in tournaments to stay on the active ratings list. So these players that their old ratings are not adjusting along with very active Shane Mccosker and Alex Moiseyev will be dropped from main list until they play another tournament. That looks like it will remove 5 of those players over Alex and many more over Shane.

Ratings will be more compressed as you say Liam. But in all honesty I am not seeing anything wrong with that? In prebious tournaments you have 1800-2650 all playing in masters????????????????????????????????????

now you will have 2000-2300 all playing in masters. The way this system works is after many tournaments are played it will CLEARLY weed out the different divisions. Just go look at kurnik when it has 50 players in it.

12 1000-1200, 10 1200-1300, 10 1300-1400, 8 1400-1500, 5 1500-1600, 3 1600-1700, usually 1 or 2 id's higher than 1700

if we gathered all of them up for a tournament there would be a very clear way to split them into groups. With checkers being so draw heavy you REALLLY don't want your opponent to be 400+ points under you anyways. I think ACF needed "compression syndrome" badly.

Not too mention.... every single pairings system has its flaws. Glicko still has same problem and more so what exactly are you suggesting by saying we are ignoring you?

[JR Smith]

by Eric Strange on Tue Jun 28, 2011 11:43 pm

well in showing the tournament it says that Justin gained over 200 points in that tournament. He was in database with his OLD calculated tournaments you told me about when he was younger.

I did find a problem though. When you told me he came in 6th I started sniffing around and found that the cross table Richard gave me had Justin at 15 points and the Cross table on nccheckers.org shows Justin with 17 points. I need to know which it is because in round 5 is where the difference is. NCcheckers shows Justin got 2 wins vs Dr. Robert Shuffett and the cross table Richard gave me shows 2 draws. If in fact Justin won both games in round 5 I can just setup a 1 round tournament and play those 2 against each other to fix it.

My stuff is correct... look again at cross-tables, finish positions, and source document score-sheet. All of the them reflect the same thing, Justin had 17 game points. I've been emailing you about Justin and several other who are concerned about their ratings. I know it takes time and I personally think you are doing a tremendous job!

http://www.nccheckers.org/NCCA/2010%20T ... esults.htm

[jeff webster]

I agree with JR, this is a huge task you are tackling and you are doing a fine job. Your efforts are appreciated !

[Eric Strange]

Thank you for your appreciation for what me and my coder have been doing.

Now to business, since I do believe that Justin in fact did get the 17 points and the cross table I had was wrong I ran a single round tournament vs Shuffett and Gober and gave Justin his 4 points and his rating has now been adjusted. He jumped to 1791 so he is almost 200 points over where people start at in total he gained just under 300 points from that one tournament which was 7 or 8 rounds I can't remember. If that isn't a huge change in rating I do not know what is. The information I am given is not always perfect or accurate so if anyone does see something that may be wrong I can explain to you why it is not and/or adjust anything that may be wrong. But I assure ALL of you that my system is not flawed or "buggy". It is just not the same system your all used to.

Thank all of you

-Eric Strange