So here's the think about that: The 8% came true. It's why some people bet on the green 00 on Roulette. Small likelihood, big payoff. It doesn't mean the HRC percentage was, per se, wrong.
-
-
Replying to @Snowman55403 @lindsaycrouse
In which case all of these projections are pointless or misleading. Pick one.
24 replies 1 retweet 61 likes -
Replying to @zeynep @lindsaycrouse
It means to me that probabilities are complicated, and not a view of the future, and news outlets should use much more caution in flogging them.
2 replies 0 retweets 26 likes -
It is surreal to me that people cannot grasp probabilities, one of the simplest of statistical concepts.
2 replies 0 retweets 1 like -
Replying to @dontknownowtf @Snowman55403 and
Nah, probability is conceptually very hard. The development of probability took hundreds of years of philosophical debate (still ongoing).
1 reply 0 retweets 2 likes -
Replying to @richardtomsett @Snowman55403 and
I know, but what could be easier than understanding that x has a 91% chance of occurring?
2 replies 0 retweets 0 likes -
Replying to @dontknownowtf @Snowman55403 and
When x is a single event that will never be repeated under the same conditions again, how should we interpret what a 91% chance of x occurring means? How would we distinguish that prediction from one where we said x had a 92% chance of occurring (or even a 19% chance)?
1 reply 0 retweets 0 likes -
Replying to @richardtomsett @Snowman55403 and
If pollsters had a long history of being wrong, I would agree that such numbers are meaningless. This is Twitter, and I certainly wasn’t going to overdo this.
1 reply 0 retweets 1 like -
Replying to @dontknownowtf @Snowman55403 and
it’s difficult with election predictions because of relevant changes in the context of each one and how the methodologies change. The 538 prediction is also slightly different in meaning from The Economist prediction for example-for 538, x is the expected outcome were the ...2 replies 0 retweets 1 like -
Replying to @richardtomsett @Snowman55403 and
Bitecofer claims traditional methodology is wrong due to a rise in partisanship. I was making a small point. Btw, Silver said Trump’s chances were 1 in 4 late in 2016.
1 reply 0 retweets 1 like
Yeah 538 coupled polling errors, better than most. But my point is broader.
-
-
Replying to @zeynep @richardtomsett and
I know. I wasn’t making a big point. I was only asking: what does it mean to say that the odds of something occurring is x? People do not get that, let alone the methodology underlying the prediction
0 replies 0 retweets 2 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.