GOTW - Whackometer Week 5

Well, this week was surely an eventful one in the United States Chess League's Game of the Week. The results of the voting were razor-thin, and content of the comments were razor-sharp... even Nakamura took a judge to task.

That being said, what does the whackometer say this week?

Let's take a look at the results and their interpretations.

Name  Week-5
Jeff   0.104 
Greg  -0.103 
Arun  -0.090 
JimD  -0.116 
Mike   0.516 

Remember that a number closer to 1.000 means that the judge agreed with the collective wisdom of the other judges, a number closer to -1.000 means that the judge disagreed with the collective wisdom of the other judges, and a number close to 0.000 means that the judge neither agreed nor disagreed with the collective wisdom of the other judges.

Four our of five judges have almost no relationship with the collective wisdom of the other judges. Only Michael ranks highly above them all, with pretty good agreement with the collective wisdom of the other judges. In other words, if you average the ratings of the other four judges, they agree pretty well with Michael's ratings.

Michael was the potential whacko, but not any more. Halway through the season, the average correlation of all the judges are...

Name  Average
Jeff   0.331 
Greg   0.243 
Arun   0.376 
JimD   0.422 
Mike   0.262 

...which suggests that the all are pretty similar. Clearly no whacko now.

As a side note, a comment on the whackometer for last week asked if the correlations presented herein used just the games getting votes, or all the games, and I responded that it was just the games getting votes, but I had pointed out back in week 2, that it would be more proper statistically to include all the games (i.e., games where no one rated it in the top five). So, I suppose the whackometer really measures whether judges agree on how the top games should be ranked.