| |
BGonline.org Forums
Giants Voting Info Requests
Posted By: Jake Jacobs In Response To: Giants Voting Info Requests (Bob Koca)
Date: Monday, 15 February 2016, at 4:58 a.m.
Bob,
Carol Joy always run a detailed piece which includes a statement from Yamin, and at least some of the information you ask about. That is the official release, and that is all you or anyone will have to work with. I realize that playing with the data is a lot of fun, and the more you have the better. But the votes are private. Even the limited amount that leaked out was unfair to the voters involved.
Naturally there are various ways of interpreting the data. Rick has posted something below that suggests that the less like his vote, the more "biased" it is.
Not everyone uses his method of choosing giants, nor do we require it.
We have (and won't be sharing, so don't ask) various runs we did. We did not do one based upon either: the error ratings of voters, nor the error ratings of those voted for. For one thing: we don't have the tools. For another,we are not in the "error rating business." What we did do was look at the impact of nationality on voting. No, I won't go into detail on methodology. But it is safe to say a couple of things.
First, if you threw out every ballot except the giants not from Croatia, i.e. 90% of the ballots, which is a drastic effect, you would still find Croatian names. I won't say if one or two dropped, nor will I say if any were or weren't in the top 32, but it is safe to say that the giants may have a higher opinion of Croatians than do some folks who rely solely on error ratings.
Second, we took the percentage of total ballots who "showed respect for Croatians" (more than one country did) and extrapolated from the ballots of various other countries to see what impact those ballots would have if numerous enough to equal the percentage of Balkan ballots, we found that the impact was greater for every country but the US(because the US already voted an even larger percentage). In other words if we got a lot of Japanese ballots there would be a lot more Japanese names. If we got more Danish ballots, the top 32 would be very Danish.
After looking at more than one thousand ballots over the years, and probably fifty thousand individual names, I don't find this surprising. Heck, I have written to that effect.
There are only two things that make the Balkan ballots stand out. One: they don't seem to think that as many of their players are exceptional, as compared with Danes or Japanese or Americans ... (I will stop there before I begin listing more controversial countries and arouse additional anti-national fervor). Which means that where a Japanese voter or Danish voter or American voter(or some other countries' voters)might have as many as fifteen or twenty worthy compatriots on their ballots, the Balkan voters seem to have been impressed by the same four or five, so there was less amelioration. And the Balkan voters did what we asked: they voted.
Otherwise, we looked at the ballots long and hard. There was no evidence of collusion, and no reason to reject any ballots. Unless one thinks: "They voted for a couple of Croatians" is a reason. We don't.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.