Stats Made Easy

Practical Tools for Effective Experimentation

Tuesday, October 31, 2006

Drinking twice as much reduces heart attack by factor of three?

After suffering a mild heart attack (an oxymoron!) a few years ago, I asked the cardiologist if it's true that a glass of red wine a day keeps the myocardial infarctions away. He said "yes." What about white, I wondered. "That works too," said he. Encouraged by this, I wondered if beer might work too. The answer was affirmative. Next I questioned whether two drinks might be even better. After that got endorsed by the cardiologist, I quit while I was ahead. I've enjoyed one glass of beer or wine, and occasionally a second helping, every day since. Life is good!

Today I was heartened to see in the HeartCenterOnline Newsletter that a study by Beth Israel Deaconess Medical Center in Boston indicates that two drinks daily help men avoid heart attack. At first glance at the following detail I thought I ought to have more than two drinks:
"There were 9 heart attacks in a group of 714 men who drank more than two drinks daily, and 34 in a group of 2,252 who drank less than two a day." Unfortunately, if you do the math and calculate the percents by comparison, this statistic becomes a lot less compelling for those who like their liquor. I am holding the line at one drink every day for sure and two at the most, but only when I want to really live it up.

Thursday, October 26, 2006

Sound from Sand: Rock Music?

At a vastly smaller scale than Stonehenge, stones ground by natural forces to a well-rounded and relatively regular shape can make very distinctive sounds, which you can hear at song of the Atacama Desert dunes in Chile. According to physicist Stephane Doudy of the Centre de la recherche scientifique in Paris, the volume of sand slides can reach a nearly unbearable 110 decibels -- on par with a jet engine. For the complete story of how "self-synchronized avalanches turn piles of sand into musical instruments," check out Dulcet Dunes by Fenella Saunders of American Scientist magazine. What I find fascinating is that it evidently takes a veneer of salt to make sand sing, so if you really want to hear a tune from a dune, head for a desert that spills into the sea.

Thursday, October 19, 2006

Stonhenge blocks demonstrably moved by man -- not magic

My youngest brother, an engineer like me and our father, sent us this link showing how a fellow from Michigan, Wally Wallington, single-handedly lifted a Stonehenge-sized pillar weighing 22,000 lbs. I visited Stonehenge in June and learned that, prior to erecting these really large pillars, earlier builders (2000 BC!) put up 80 bluestones, up to 4 tons apiece, that they mined 240 miles away in Wales. These were thought to have been moved magically by the sorcerer Merlin. More likely these were transported much of the distance by raft and overland on rollers as demonstrated by the Millennium project.
I thought the bluestones were the coolest of all that I saw at Stonehenge, but you must look beyond the larger sandstone pillars to see them and appreciate how much older they are. For more on their history, see the Secrets of the Preseli Bluestones by Dr. Colin R. Shearing.

Any sufficiently advanced technology is indistinguishable from magic.-- Arthur C Clarke

Sunday, October 15, 2006

Lost arts -- slide rules and cursive writing

Now that a whole generation has grown up with personal computers at their disposal, many once-necessary skills have become lost arts. For example, I did most of my college computing with a slide rule, and when I reported to work as summer engineer in 1974, General Mills provided me with a circular one. The University of Minnesota had one Wang calculator that students waited in line to use for doing logarithms to more decimals than possible with a slide rule. General Mills bought one Hewlitt-Packard calculator that did logs and exponential calculations using reverse Polish notation. It was so costly that they bolted it to a table! Nowadays slide rules have become an item for collectors such as fellow U of M alumnus Gary Flom. Aficionados of slide rules formed the Oughtred Society name after William Oughtred, an Anglican minister who invented this calculating device in 1622. My father, an engineer like me, owned a really nice Keuffel and Esser (K&E) slide rule. However, from my quick browsing of the internet, it seems that these go for only about $25 -- far less than what one would have paid originally if adjusted to inflation. As reported in The Death of the Slide Rule by James Redin, the K&E manufactured its last slide rule in 1975 -- the year I achieved my bachelor's degree in chemical engineering.

PS. Another lost art, reported in my Sunday newspaper today by Washington Post writer Margaret Webb Pressler,* is cursive (longhand) writing. She reports that 85 percent of almost 1.5 million students taking their college SAT exams wrote in block letters. Computers have made this style of penmanship obsolete.

*The Handwriting Is on the Wall

Sunday, October 08, 2006

Minnesota Twins beat the statistical odds once again

On May 17 I reported that sabermetrician Clay Davenport computed 200 to 1 against the Minnesota Twins making the Major League Baseball playoffs. Guess what? Not only did they achieve a place in post-season, they won their divisional championship. Granted, it was very unlikely the way the Twins turned their season around, and it was downright surrealistic for them to end up in first after their very last game. As I blogged earlier, statistics be damned by what we now know:
Twins win improbable division title.

One might do well by betting on the Twins when they are down again in future. For example,at the beginning of the 1991 baseball season, odds on 1990’s last-place Twins winning this year’s title were 100-1. They ended up as the World Series champs. Similarly, in 1987 the Twins went all the way in Major League Baseball. Prior to that season I went to Las Vegas for a conference and saw a betting board with odds at 100 to 1 against the Twins winning the championship. Ever since I've second-guessed myself for not betting anything -- even $10 would have netted me $1000! Unfortunately, I am a man of little faith in the face of such overwhelming statistics.

PS. Post season results were not good -- three games and out for the Twins in their playoff series with Oakland. :( This is a triumph for sabermetrics because the Athletics are led by its biggest proponent --Billy Beane.

Sunday, October 01, 2006

Economists shave hairs on whether basketball games are fixed: Any bets on who wins?

In my March 26 blog I reported that 'forensic economist' Justin Wolfers, a professor at the University of Pennsylvania, inferred point-shaving from his statistical analysis of 44,120 NCAA Division I basketball games between 1989 and 2005. This new study by University of Illinois economist Dan Bernhardt disputes Wolfer's contention that statistics indicate point-shaving on college basketball. Perhaps it's only natural that superior teams fall short of expectations on their winning margin. According to Professor Bernhardt "the statistical properties that Wolfers identified in his paper seem to be intrinsic to the game of basketball itself, occurring independently of whether there are incentives to point shave, and are not indicative of an epidemic of gambling-related corruption."
It's good that this new analysis dissipates the cloud of suspicion about point-shaving raised by the first study.