News and Announcements
Posted Jul 4, 2007 11:14am by Lyle.
We have finally received the go-ahead from the executive committee to launch the site. The database is online, our first featured predictions have been posted, and the Google Ads will be up within a few days. I'm sure there will still be some minor glitches to work out, but we felt it was time to release it beyond our group of beta testers and see what happens.
We had some unexpected delays due to staffing problems in the legal department and the research department, but those issues have been addressed through decisive action by the executive committee.
I thought this might be a good time to talk a little bit about the history of the Punditude concept. Early in 2007, there was a lot of back-and-forth in the media, and especially the blogosphere, about predictions that had been made about the war in Iraq, and who had been consistently right, and of course who had been consistently wrong. Our assumption was that there had to be, somewhere on the web, some person or organization who was actively cataloging and analyzing predictions made by pundits, in order to provide some systematic and objective analysis of these charges of prediction inaccuracy. To our surprise, this "prediction repository" did not seem to exist.
I've been watching The McLaughlin Group more or less regularly for more than 20 years. It's a round-table discussion show which almost always concludes with each panelist giving a brief prediction, nearly always related to politics or current events. Again, we assumed that this would provide a perfect opportunity for someone, somewhere, to collect and analyze these predictions. With one exception (a mostly complete collection of predictions, but with a mostly incomplete collection of outcomes, on some guy's blog), we were again amazed to find no such collection.
And so the Punditude concept was born. We initially visualized a large, searchable database of pundit predictions that would be useful to the public primarily as a research tool. This concept was gradually expanded to include the PunditScore idea, and then to embrace user interactivity through participation in the prediction process by agreeing or disagreeing with pundit predictions. Finally, with the addition of the Featured Predictions, allowing us to expand beyond just the predictions being made by professional pundits, all the pieces were in place.
Along the way, we had the usual technical difficulties to deal with. More importantly, however, were the unexpected conceptual difficulties that we encountered. Unfortunately the process of prediction analysis was much, much more difficult than we had imagined. It quickly became apparent that the vast majority of predictions coming out of the mouths of pundits are completely worthless, at least from the perspective of objective analysis. Either the prediction is so lacking in specificity as to end up being unfalsifiable, or it's so peppered with qualifiers and conditionals as to be completely meaningless. See FAQ #7 for more on these difficulties.
Furthermore, there is the matter of interpretation. The analyst must make a determination on what the pundit really meant, whether he meant it as a joke, and whether he would agree with our interpretation and resolution. While it is our intention to remain absolutely non-partisan, objective, and independent, we have no doubt that we will be subject to harsh criticism from both the right and the left regarding our interpretations, prediction inclusions and omissions, and of course prediction resolutions. Obviously, all we can do is our best, and all decisions will be open for discussion via the comments section of individual predictions or blog posts.
Finally, just the process of researching individual predictions was extremely time-consuming. We had hoped to have at least 1000 predictions when we launched the site; we are going live today with somewhere slightly less than 300, including all the McLaughlin Group predictions that meet our standards going back to 9/11/2001.
So feel free to explore the site, set up an account, and start taking positions and building up your PunditScore!
[In case it's not obvious, Punditude Media has neither a "legal department", "research department", nor an "executive committee". What we do have is a couple of guys with jobs who are doing this in their spare time. Enjoy!]
We had some unexpected delays due to staffing problems in the legal department and the research department, but those issues have been addressed through decisive action by the executive committee.
I thought this might be a good time to talk a little bit about the history of the Punditude concept. Early in 2007, there was a lot of back-and-forth in the media, and especially the blogosphere, about predictions that had been made about the war in Iraq, and who had been consistently right, and of course who had been consistently wrong. Our assumption was that there had to be, somewhere on the web, some person or organization who was actively cataloging and analyzing predictions made by pundits, in order to provide some systematic and objective analysis of these charges of prediction inaccuracy. To our surprise, this "prediction repository" did not seem to exist.
I've been watching The McLaughlin Group more or less regularly for more than 20 years. It's a round-table discussion show which almost always concludes with each panelist giving a brief prediction, nearly always related to politics or current events. Again, we assumed that this would provide a perfect opportunity for someone, somewhere, to collect and analyze these predictions. With one exception (a mostly complete collection of predictions, but with a mostly incomplete collection of outcomes, on some guy's blog), we were again amazed to find no such collection.
And so the Punditude concept was born. We initially visualized a large, searchable database of pundit predictions that would be useful to the public primarily as a research tool. This concept was gradually expanded to include the PunditScore idea, and then to embrace user interactivity through participation in the prediction process by agreeing or disagreeing with pundit predictions. Finally, with the addition of the Featured Predictions, allowing us to expand beyond just the predictions being made by professional pundits, all the pieces were in place.
Along the way, we had the usual technical difficulties to deal with. More importantly, however, were the unexpected conceptual difficulties that we encountered. Unfortunately the process of prediction analysis was much, much more difficult than we had imagined. It quickly became apparent that the vast majority of predictions coming out of the mouths of pundits are completely worthless, at least from the perspective of objective analysis. Either the prediction is so lacking in specificity as to end up being unfalsifiable, or it's so peppered with qualifiers and conditionals as to be completely meaningless. See FAQ #7 for more on these difficulties.
Furthermore, there is the matter of interpretation. The analyst must make a determination on what the pundit really meant, whether he meant it as a joke, and whether he would agree with our interpretation and resolution. While it is our intention to remain absolutely non-partisan, objective, and independent, we have no doubt that we will be subject to harsh criticism from both the right and the left regarding our interpretations, prediction inclusions and omissions, and of course prediction resolutions. Obviously, all we can do is our best, and all decisions will be open for discussion via the comments section of individual predictions or blog posts.
Finally, just the process of researching individual predictions was extremely time-consuming. We had hoped to have at least 1000 predictions when we launched the site; we are going live today with somewhere slightly less than 300, including all the McLaughlin Group predictions that meet our standards going back to 9/11/2001.
So feel free to explore the site, set up an account, and start taking positions and building up your PunditScore!
[In case it's not obvious, Punditude Media has neither a "legal department", "research department", nor an "executive committee". What we do have is a couple of guys with jobs who are doing this in their spare time. Enjoy!]
Results: 1
1 
1 