Get certain. Do your job. Vote. ---Stephen King
Often when I present someone's quote on this blog I don't know the context in which he or she is making it. Gladly, this is not the case with Stephen King's quote: he posted it yesterday on Twitter, with a link to a tweet from USA Today that read, "A new survey found that just 35 percent of Americans ages 18-29 are 'absolutely certain' that they will vote in November." King wasn't very happy about that number, consequently his admonition. I, on the other hand, am astounded in a positive sense with it, as I will explain...
As I showed in my Tuesday article this week, the percentage of registered voters aged 18-29 has been utterly dismal for off-year general elections, i.e. elections in even-numbered years when the presidency wasn't being voted on (like 2018). Their voting turnout for 2002, 2006, 2010, and 2014 was 18.2%, 20.3%, 20.3%, and 16.3%, respectively. So for registered voters in this age range to rise up to 35% for this coming election is certainly encouraging...but that's not all. The USA Today report expressed 35% of Americans 18-29, not just registered Americans in that group! Maybe they misreported this part, but if even if they did, we may be looking at a lot of surprises on election night, November 6th...
Showing posts with label Statistics. Show all posts
Showing posts with label Statistics. Show all posts
Friday, October 12, 2018
Thursday, November 6, 2014
ESPN Analysts Display Lack of Probability Knowledge
Yesterday, on ESPN's Sport Center, they were going on about how badly the Los Angeles Lakers are doing so far this year, going winless in their first five games. One point they brought up was the upcoming schedule for them, with Kobe Bryant and company favored to lose each of their next eight games. There were actual percentage probabilities given for the Lakers to win each game, with the figures hovering around the 30th and 40th percentiles. Then one of the announcers got to the point of all this, which is that they may well lose all eight games and start the season off at 0-13. I beg to differ...
I didn't get all of the percentage predictions of each game (I wonder how they even arrived at them), but I think I'm being pretty conservative at saying they averaged about a 35% winning probability given for each game (it's probably higher). That means that the losing probability for each game is .65: for the Lakers to lose all eight of their next games, you would need to take that probability to the eighth power:
(,65)(.65)(.65)(.65)(.65)(.65)(.65)(.65)=.032 , or 3.2%
In other words, even though the Lakers may be heavy underdogs in each of the next eight games with only a 35% average chance of winning each one, the probability of them losing ALL eight games is only 3.2%!
This is a glaring example of how otherwise very intelligent and educated people can falter the minute they go any distance in interpreting statistical data or anything dealing with probability. Will the Lakers actually lose their next eight games? Maybe, but it's not very likely given the input data...
I didn't get all of the percentage predictions of each game (I wonder how they even arrived at them), but I think I'm being pretty conservative at saying they averaged about a 35% winning probability given for each game (it's probably higher). That means that the losing probability for each game is .65: for the Lakers to lose all eight of their next games, you would need to take that probability to the eighth power:
(,65)(.65)(.65)(.65)(.65)(.65)(.65)(.65)=.032 , or 3.2%
In other words, even though the Lakers may be heavy underdogs in each of the next eight games with only a 35% average chance of winning each one, the probability of them losing ALL eight games is only 3.2%!
This is a glaring example of how otherwise very intelligent and educated people can falter the minute they go any distance in interpreting statistical data or anything dealing with probability. Will the Lakers actually lose their next eight games? Maybe, but it's not very likely given the input data...
Sunday, February 19, 2012
Life Expectancy Figures Skewed by Infant Mortality
On CBS News Sunday Morning today (2/19) they were discussing a former president, William Henry Harrison. It was an interesting history lesson and the announcers seemed to have their facts right. Except for one thing. After mentioning that Harrison was sworn into office at the relatively advanced age of 68, one of the announcers marveled that the life expentancy then was only 39 years; back then it would have been like a 110-year old was becoming president. NO, NO, NO!
The main reason that life expectancy figures were so low back then (and now, in impoverished parts of the world) was because of the enormously high infant mortality rate, which drastically skewed the figures. Being a human being in the first couple of years of life was very, very dangerous, with fatal illnesses a commonplace, tragic feature. Once someone did make it out of early childhood, they stood a good chance of making it into old age eventually (unless they were sucked into a war not of their own making, that is). So being 68 was NOT some extraordinary feat of longevity, as the TV person implied...
The main reason that life expectancy figures were so low back then (and now, in impoverished parts of the world) was because of the enormously high infant mortality rate, which drastically skewed the figures. Being a human being in the first couple of years of life was very, very dangerous, with fatal illnesses a commonplace, tragic feature. Once someone did make it out of early childhood, they stood a good chance of making it into old age eventually (unless they were sucked into a war not of their own making, that is). So being 68 was NOT some extraordinary feat of longevity, as the TV person implied...
Subscribe to:
Posts (Atom)
