This infuriatingly undefined question is one of the reasons I decided I don't have time to struggle with this game. I like having a reputation as a superforecaster from GJP4, but this game seems designed to teach us humility by forcing us in cases like this to forecast the organizers of this game instead of the apparent question. Instead, now I'm concentrating on the upcoming Hybrid Forecasting Competition. I hope to help design one of the competing versions of this game, but if I turn out to not be on a winning proposal team, then I'll jump at the chance to forecast on some other contractor's version of the game. http://www.iarpa.gov/index.php/research-programs/hfc
@000: That is a "loaded" image. @ravel has never claimed to have nibbled through the structure of the system, only to have explored it and reported upon what he found. He's a real life pioneer who is valuable as a navigator for us all. In terms of that image, I find that the apparent nibbling requires an explanation.
@ravel I once went through the first 60 forecasters by fiddling the URL, but I never thought to see if I could see their rankings. Cmeinel had mentioned that many Superforecasters from past seasons were in the first 200 registered, so I was looking for them. I found that most of those who were early to register are inactive.
Odd that they disabled your hack to see the rankings. I also found it odd that GJO put our Accuracy/Relative scores on our profiles, then took them off, as if they're trying to hide people's rankings. Rankings should be public, but we can only see the top 20 in each challenge plus the page we are on. I can understand if they don't want to spend the resources making full ranking tables, but why disable the hack?
@000 People who have just a few big scores will fall down the rankings as more questions are resolved, though, as accuracy scores are cumulative. But avoiding questions that are too risky and staying close to the median until you are sure seem to be a big part of doing well in the challenges.
I should say that @mparrault is an exception. He answers all questions and does very well, at least in Geopolitical.
@praedico, they do explicitly say somewhere that picking your questions is part of the art of being "super". However it has nothing to do with what I assume to be the reality of a working intelligence analyst which (under my assumption) is to chew on whatever is put on your plate. Maybe they are saying that that reality should be changed, analysts should pick and choose their questions at will. This also assumes that there is no politics whatsoever about who can work on what question in "the office". The latter I'm really not so sure about. My touchstone on all of this is Bridget Nolan's thesis (also a U Penn product, like GJOpen).
@000 Intelligence analysts specialize in a certain area, though. But, yeah, they would have to make predictions on whatever is asked within those parameters. I'm not sure that every analyst has to express an opinion on every question. I don't know how it works. Do they work in teams? If so, some team members will have more to say than others on any given question. In that sense, you could say that the analysts are picking and choosing. After all, they don't want to look bad by expressing strong opinions on subjects that are not in their purview.
This infuriatingly undefined question is one of the reasons I decided I don't have time to struggle with this game. I like having a reputation as a superforecaster from GJP4, but this game seems designed to teach us humility by forcing us in cases like this to forecast the organizers of this game instead of the apparent question. Instead, now I'm concentrating on the upcoming Hybrid Forecasting Competition. I hope to help design one of the competing versions of this game, but if I turn out to not be on a winning proposal team, then I'll jump at the chance to forecast on some other contractor's version of the game. http://www.iarpa.gov/index.php/research-programs/hfc
Anyone who wants to follow me offline as I seek the Holy Grail of a more powerful aid to us forecasters, you can reach me at carolyn.meinel@cmeinel.com. Meanwhile I'll do my best to get OFF this game instead of being listed as inactive. So hasta vista to my followers who aren't already interacting with me outside this game: @Xsess @Dwight-Smith @thmnewman @Gil-Edgar @tmahoney @lsgold7 @Counterintelligence @Enthu @Hochstetler @Delorean @g @howard @facetious @The_Gnome @BG1 @DrStrangelove @Flyn1200 @PianoPicasso @AndersAsa @writeitdown @HW15 @tkimble @Doudtful @firemansghost @mparrault @lindsey @Dima-K @crntaylor @azivkovic @Mos @MattWard @AlexisTocqueville @luckyomari @Ermonic @Pstauble @Edwinian @Clairvoyance @ACurmudgeon @ @ts2m @M3T1tus @dada @GeneH @S1 @ConnorM @peterhansen90 @Bklyn_j @dniewood @deggen @Etsudo @praedico @Spyglass @Manfred @VoxVox @FuturoMAGE @subject1138 @MAA414 @richtyge @gkamstra @RCScheffers @Random @Spandrelbarca @Rene @gstaneff @fifty-sixty @spotter @dbealick @davidk @terobrandstaka @RobK @walt @Rectitude @malcmur @dominich @Konrad @balbec @Ioana @Raisinville @rjfmgy @Rote @TopQuark @ESR @Xu @sharms10k @Paul15 @NickLutz @Ritam @seveDB @Jean-Pierre @Bill @Agent0090 @RolandKofler @Aches @JoeG @madre @Paul_Theron @September @DariusX @AgentCarter @JeanP @tumbleweed @Reynard @SmallTown-Gal
Comment deleted on Apr 16, 2016 08:17PM UTC
@ravel, like this then: https://mobsterratsquad.files.wordpress.com/2014/06/img_3750.jpg
@Resonancia, you are not dating yourself. http://www.redicecreations.com/ul_img/20737bighorn-medicine-wheel.jpg
@Resonancia Nice to know I am not the only one who totally failed the Economist Challenge while doing reasonably well in the others.
@000: That is a "loaded" image. @ravel has never claimed to have nibbled through the structure of the system, only to have explored it and reported upon what he found. He's a real life pioneer who is valuable as a navigator for us all. In terms of that image, I find that the apparent nibbling requires an explanation.
@ravel I once went through the first 60 forecasters by fiddling the URL, but I never thought to see if I could see their rankings. Cmeinel had mentioned that many Superforecasters from past seasons were in the first 200 registered, so I was looking for them. I found that most of those who were early to register are inactive.
Odd that they disabled your hack to see the rankings. I also found it odd that GJO put our Accuracy/Relative scores on our profiles, then took them off, as if they're trying to hide people's rankings. Rankings should be public, but we can only see the top 20 in each challenge plus the page we are on. I can understand if they don't want to spend the resources making full ranking tables, but why disable the hack?
@000 People who have just a few big scores will fall down the rankings as more questions are resolved, though, as accuracy scores are cumulative. But avoiding questions that are too risky and staying close to the median until you are sure seem to be a big part of doing well in the challenges.
I should say that @mparrault is an exception. He answers all questions and does very well, at least in Geopolitical.
@praedico, they do explicitly say somewhere that picking your questions is part of the art of being "super". However it has nothing to do with what I assume to be the reality of a working intelligence analyst which (under my assumption) is to chew on whatever is put on your plate. Maybe they are saying that that reality should be changed, analysts should pick and choose their questions at will. This also assumes that there is no politics whatsoever about who can work on what question in "the office". The latter I'm really not so sure about. My touchstone on all of this is Bridget Nolan's thesis (also a U Penn product, like GJOpen).
@000 Intelligence analysts specialize in a certain area, though. But, yeah, they would have to make predictions on whatever is asked within those parameters. I'm not sure that every analyst has to express an opinion on every question. I don't know how it works. Do they work in teams? If so, some team members will have more to say than others on any given question. In that sense, you could say that the analysts are picking and choosing. After all, they don't want to look bad by expressing strong opinions on subjects that are not in their purview.
@praedico, behold and luxuriate in the details of the only open source datapoint on the matter of what intelligence analysts do at the office:
https://cryptome.org/2013/09/nolan-nctc.pdf
although my favorite fictional version goes like this:
http://rubicon.wikia.com/wiki/American_Policy_Institute