| |
BGonline.org Forums
Revised Giants List (Long)
Posted By: Jake Jacobs
Date: Sunday, 13 March 2016, at 12:06 p.m.
On The Shoulders Of Giants
Let me begin with an “official” statement by the Giants Committee. Because of the extraordinary circumstances this year – an unusual voting pattern and a flurry of rumors about it – after releasing the list we audited some of the ballots again. We did more due diligence this year than perhaps we have done in the previous eleven voting years. After our audit was complete we rejected eight ballots. All eight were rejected not because of their content, but because close scrutiny led us to conclude that these eight players had failed to meet our requirements for voting.
The impact of removing those eight ballots has affected the bottom half of the list, with some names moving down, and some moving up. Some of the effects might come as a surprise to the backgammon community. We strongly suspect that many will remain unhappy with the final list as there were over fifty ballots they wished to see purged, and a number of names forced off the list. Under our rules we could not do that. We will be reviewing things over the next two years to see what sort of changes might need to be made to avoid similar incidents in the future.
The revised list is now available at http://www.flintbg.com/giants.html. Yamin Yamin’s statement and some statistics will appear in the winter issue of the Flint Area BackgammoNews.
That was the official statement. What follows is my personal statement.
I read something in a recent issue of The Economist. The USA is not the only country with an upcoming election. Iran is holding one, and according to the article in preparation for it the mullahs will be determining the people’s list of choices. That isn’t so different from the US, as its “mullahs” most years make the important decisions, and tell the people they may choose from Column A or Column B after making sure that names from Columns C onward have been purged. And in both countries the voters are for the most part comfortable with the process because they fear, or have been taught to fear, their fellow voters. They would rather a mullah tell them for whom to vote, than leave things up to their fellows.
There is an important difference though. In America we give lip service to the idea that the people can be trusted. Throughout our history the vote has been extended to more and more people: minorities, women, and voters in their late teens. More than lip service, many Americans believe in inclusion and that once you are eligible to vote: it is your right; that your vote should count as heavily as the next man’s; that your vote should be kept private. Those beliefs underlie the Giants List.
Now we find ourselves in crisis, facing the question: will those beliefs save the list, or destroy it?
This year rumors have arisen that Croatian and Bosnian voters voted in large numbers and by doing so affected the results. It is not a violation of their privacy to say that anyone looking at this year’s list can guess that this is true. In mid-January we were well behind the voting totals of previous years, but had already received fifteen or twenty Balkan ballots. That was when rumors of an attempt by “Croatians” to manipulate the vote began swirling. Whether this motivated others to vote is unknown, but by the January 31st deadline we had received more votes than ever before, even without the Balkan voters. The Balkan ballots had themselves continued to arrive, until we had nineteen Croatian ballots and thirty-four Bosnian ones.
We had never faced quite this situation, and wondered what, if anything, we should do about it. We did three things. One would have happened anyway: we checked the eligibility of all new and unfamiliar voters. The second was to take a closer look at the ballots to see if there was an indication that the submissions were traceable to a common source, rather than independent votes. While the feature that first caught our eye remained, that a few Croatians placed very high on all ballots, it was also apparent that each ballot had been filled out by a different individual. Fifty-three different people spent time thinking about and filling out the ballots. And while the same four or five players were common to the ballots, and were in high-ranking slots, the order varied from ballot to ballot, something one can guess from the published lists.
The third thing we did was to compare some other nationalities with the Balkan ballots by taking the ballots we had from them, and extrapolating as though those voters had voted in numbers comparable to the Balkan voters. We found that had any of the other nationalities voted in such numbers the list would take on a noticeable bias such that instead of four Croatians there would be quite a few more of any of the other nationalities. I have heard it argued that this isn’t a sign of bias, because some nationalities “have better players.” While that may be true, and my own biases lead me to believe that Japanese and Danes probably play better than Croatians, that is not part of our official voting rules, nor should it be. What if two married players voted for each other? Could anyone claim that if they were Croatian bias was involved, but if they were Japanese or Danish, there wasn’t?
We were not surprised by our findings because we have looked at more than a thousand ballots over the past twenty-two years. We have seen instances where we strongly suspected that two players agreed to name each other number one. We have seen ballots where the first sixteen names were all from the same country, or, hardly better, the last sixteen names were. We have seen lists that were alphabetical, and lists that we didn’t understand at all. All of it in keeping with one of our most basic principles: We don’t tell people how to vote. And despite those odd looking ballots the lists were the lists every year.
We also looked at the effect of counting the votes from only the Giants who were not Croatian, a drastic change since ninety percent of the ballots weren’t counted. Even then one Croatian just missed making the top 32 and the others were all among distinguished company. Clearly the current Giants have some respect for the Croatians. (Some of this information we shared early on, but were handicapped by our respect for voter privacy.)
In the end we could find no violation of our rules that would allow us to throw out the Balkan ballots.
After the final tally was in, we could see that four Croatians of the five who were favored had made the final thirty-two, one placing very high on the list. We looked over the voters once more, doubled checked the credentials of one, and then finding no cause to reject votes, published the list.
The expected outcry came, and with it came calls to purge Balkan ballots, and for us to act as mullahs, or to appoint some if we wouldn’t, to tell the people how to vote.
Privately we heard from elite players who claimed to have a “smoking gun,” a well-placed insider who could confirm a Balkan plot. This was followed by an “incendiary post” (I am quoting) also alluding to foul play. We spent quite a bit of time conferring with well-placed insiders, and here is what we learned.
It seems the Croatians had a team that won the European Team Championships. That is almost the only fact that emerged. There were five players on the team, the five that turned up on Balkan ballots. There are rumors that one member of the team may have lobbied Balkan players to vote for the champions. Or maybe he lobbied his wife and son. Or maybe he lobbied his wife. If he did lobby the Balkan voters, he must not have urged them to put him atop the list, because the person named as having lobbied was among the bottom half of the five on the list.
So we asked ourselves once again: Was this information sufficient to reject the ballots, and the answer was that it was not. If our thinking isn’t clear to you, I will lay it out. We have no rule saying that a ballot may not show bias. How could we? How would we distinguish between a Japanese ballot naming a Japanese player Number One, an American ballot naming an American, a Danish ballot naming a Dane, and a Greek ballot naming a Greek? “Oh, these are all fine, but not yours: you are Croatian!” We have no rule saying that if two Americans, two Danes, etc. name Americans, Danes, etc. we accept them, but reject the others “because they are Croatian.” And we have no rule saying that a player cannot lobby, asking voters to vote for him, or to vote for someone else he names. We tried to find a rule other than “because you are Croatian.” We did consider “because there are too many of you.” But how many is “too many,” and is that the rule for everyone else, or only for “Croatians?”
There were also rumors as to the qualifications of the Balkan voters, rumors that they do not play with the doubling cube, etc. It was not easy to track down and verify precisely what experience the players have, but as noted above, after extraordinary effort, and possibly bending our own requirements a bit, we found eight ballots to reject. This means that we accepted forty-five ballots, even though we anticipate further outcry. We chose principle over expedience.
Let me go further. There are limits to what we can do. We cannot establish with certainty how well any voter plays, nor can we establish whether the voter is an ethical person. But the Balkan voters have now been through more rigorous vetting than any voters in our history. Anyone who says the Balkan players are not qualified to vote: is wrong. Those saying it are at best ignorant. And anyone who says Balkan players are unethical is spreading a scurrilous rumor that no one has substantiated. We don’t know the character of the Balkan players, but we can form an opinion of the characters of those who spread such rumors.
Having defended the right of the Balkan players to vote, and to vote as they choose, let me now make something very clear, to the Balkan voters and to any others who may have done what the Balkan voters are accused of doing. You may vote as you see fit, but we don’t have to like it. The purpose of the list is to determine who the backgammon community believes are the world’s best players. It isn’t a popularity contest or a competition between nationalities. Whenever you vote for someone other than the one you truly believe belongs in that slot you are shaming yourself. It doesn’t matter if you did it because of national pride, or just plain laziness: you did something wrong. If you have sixteen of your countrymen in the top sixteen slots, do you really believe, even if they are all terrific players, that those are the best sixteen in the world? I doubt it. And if you don’t believe it: why are you better than the Balkan voters?
In defending the Balkan voters and keeping their votes, we are hoping common sense will prevail. Despite comments online to the effect that it is easy to stuff our ballot boxes, it isn’t. There were fifty-three Balkan voters. I challenge our voters to recruit five next time. Not for personal aggrandizement, but as an exercise: please go out in 2017 and find five qualified players who are not already planning to vote, and persuade them to study the candidates, rank thirty-two players, and fill out a ballot. Carol Joy Cole and I have been on the front lines for the past twenty-two years, and if that Croatian player did as rumored my hat is off to him as a motivator. And if he did as rumored, what was the result? Did those fifty-three voters result in the top five Giants being Croatian? Not even close. With one exception who would have been only a few slots down from where he ended up without a single Balkan ballot, only three of the remaining four team members made the top thirty-two, and all three were near the bottom. After the eight ballots were discarded two of those dropped out of the top thirty-two. The results wouldn’t make ballot box stuffing acceptable, but it makes the danger that “next time the top sixteen will all be Croatian” rather silly. More to the point, while they made the list, they have been shamed, derided as barely competent, and otherwise treated as pariahs. As have the voters who placed them there. Are they really motivated to repeat the process with interest in 2017? Are other nations inspired by this example to copy them?
I would like to defend our rules one more time. "To be eligible to vote, you must be an Open/Championship-level player or a tournament director/organizer at the regional, national or international level." We require that voters vote for themselves as number two, and name at least sixteen players. And this year we have added the unwritten rule that we will reject duplicate ballots should we ever receive them.
Matt Cohn-Geier has done some terrific commentary here about the wisdom of crowds: https://www.youtube.com/watch?v=ybc6U0iHSxU&feature=youtu.be
Of course if the Balkan voters did stuff the ballot box it would interfere with the wisdom of the masses, though as we have seen, the effect isn’t as large as people fear. But there are other aspects to this. Besides curtailing the eligibility of voters, there have been calls for the mullahs to issue a list, eliminating all the choices some find unacceptable. Not that there is consensus: some think only PR rates should matter, others that tournament results are important. Some think only players who have played in tournaments in the past two years should be counted. Perhaps some would like to limit it to rated matches within the past two years.
When the list started some Giants were very active but didn’t play in tournaments. When I first met Falafel I thought of him as Abe Mosseri’s sidekick. Abe was already getting votes before he had played in his first tournament. When he did play in his first tournament he made post-Crawford cube errors in the finals of the Jackpot (which he won) because he had never been at post-Crawford. He won around a million dollars in his first few years of play; how’d your favorite Giant do? Abe might be winning hundreds of thousands of dollars a year currently. I don’t know how he is doing, but is a set of rules that says: “You can’t vote for him, even if he is the best and winningest player in backgammon” really superior?
This might be more than a hypothetical. You all like rumors, so here’s one for you. There is a certain player who was certainly a Giant when he played, top ten and some thought number one. It has been well over two years since he last played in a tournament. He has no current PR rating. A reliable source tells me he won over fifteen million dollars last year. I don’t expect him to show up on the list, but the mullahs would prefer that even if you knew whom I was talking about, you be prevented from voting for him.
There is one change I do think is a good idea. We found it very hard to evaluate the tournaments players listed as having qualified them. Having a committee tell us whether a tournament was an Open tournament would be helpful. The committee would get a list of events (but not who may have played) and decide which meet their standards. I’d recommend internationalists like Carter Mattig, John O’Hagan, or Bob Wachtel from the US. In Europe Tobias Hellwag, Raj Jansari, Chris Ternel, Maya Peycheva, Leonardo Jerkovic, and Lars Trabolt are all highly respected. For the Middle East, and everywhere else there is Chiva Tafazzoli. East Asia has Robin Swaffield in Hong Kong, Mochy and Kenji in Japan. Those are good choices, but there are many others.
As for other changes …? It really is your list. Perhaps in six months or a year suggestions for change will be solicited, and you may be asked to state the answers to questions such as: How active is active? Is online play an acceptable qualifier? Etc.
Okay, I have had my say.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.