Don’t cancel your chess congress - hold it online!

Details of upcoming UK events, please provide working links if possible.
User avatar
Adam Raoof
Posts: 2615
Joined: Sat Oct 04, 2008 4:16 pm
Location: NW4 4UY
Contact:

Re: Don’t cancel your chess congress - hold it online!

Post by Adam Raoof » Mon Jul 06, 2020 11:51 am

You need more information before making an assessment (time control, rating of the player, av rating of the opposition, etc) you can email me if you wish), but you should download the games of all the players and run pgnspy. Then compare this players results to those who finished second or third. If this is a top GM in a standard tournament, it is not suspicious.

This is a sample of a 2300+ recent results over 40 games including games against GMs; on of his opponents accused him of cheating, so I had a look at the games. Compare the results and draw your own conclusions.

My advice is do read the instructions provided with pgnspy. You need to have something to compare the data with, preferably an over the board event played under the same conditions to compare with.

UNDECIDED POSITIONS
Positions: 1426
T1: 379/1009; 37.56% (std error 1.52)
T2: 484/822; 58.88% (std error 1.72)
T3: 521/746; 69.84% (std error 1.68)
=0 CP loss: 730/1426; 51.19% (std error 1.32)
>0 CP loss: 696/1426; 48.81% (std error 1.32)
>10 CP loss: 547/1426; 38.36% (std error 1.29)
>25 CP loss: 394/1426; 27.63% (std error 1.18)
>50 CP loss: 228/1426; 15.99% (std error 0.97)
>100 CP loss: 85/1426; 5.96% (std error 0.63)
>200 CP loss: 34/1426; 2.38% (std error 0.40)
>500 CP loss: 10/1426; 0.70% (std error 0.22)
CP loss mean 29.14, std deviation 83.16

LOSING POSITIONS
Positions: 268
T1: 58/161; 36.02% (std error 3.78)
T2: 76/134; 56.72% (std error 4.28)
T3: 75/117; 64.10% (std error 4.43)
=0 CP loss: 134/268; 50.00% (std error 3.05)
>0 CP loss: 134/268; 50.00% (std error 3.05)
>10 CP loss: 113/268; 42.16% (std error 3.02)
>25 CP loss: 96/268; 35.82% (std error 2.93)
>50 CP loss: 71/268; 26.49% (std error 2.70)
>100 CP loss: 43/268; 16.04% (std error 2.24)
>200 CP loss: 22/268; 8.21% (std error 1.68)
>500 CP loss: 9/268; 3.36% (std error 1.10)
CP loss mean 59.85, std deviation 129.91

Ian Thompson wrote:
Sat Jul 04, 2020 6:30 pm
Adam Raoof wrote:
Sat Jul 04, 2020 2:04 pm
You can try this out for yourself using pgnspy
I have, using pgnspy's default settings for all parameters, and I'm not sure what to make of it. This player scored 8.5/9 in a tournament. Are their results suspicious or not, and why?
Overall results:

pgnspy.png

and broken down by game:
pgnspy2.png
Adam Raoof IA, IO
The Chess Circuit - https://chesscircuit.substack.com/
Tornelo - https://tornelo.com/chess/orgs/chess-england
Simon Williams "The Ginger GM" - https://gingergm.com/ref/106.html
Don’t stop playing chess!

Post Reply