Your chances of getting your predictions and therefore your decisions right? Still on the topic of prediction and reliability. According to Nate Silver in his book ” The signal and the noise” which I featured in my discussion on ” How confident are you?” the most successful sports better in the world, Harlobos Voulgaris only gets 57% of his bets right. This pretty much sets the upper limit for predictions in a world of uncertainty, doesn’t it?
So think again about your business predictions!
Returning to a previous issue. I blogged some weeks ago on how the media screams at new scientific ” discoveries” published in prestigious journals, but then neglect to report as vigorously when those discoveries fail to live up to their promise (http://blogs.fin24.com/bertieduplessis/2012/09/26/not-so-even-handed/). Nate Silver now reveals an even more serious concern. He refers to a paper by John P Ioannidis in 2005, ” Why most published research findings are false.” He focused on bio-medical research and concluded that most of the reported findings would likely fail, if repeated in the real world. ” Bayer Laboratories recently confirmed Ioanndis’s hypothesis, they could not replicate two-thirds of the positive findings claimed in medical journals when they attempted the experiments themselves..” (Silver, p11).