Friday, June 24, 2011

Posts to Kobe, Tell Me How My Stats Taste may be a little slow this summer (and so is the audience, so it might not matter) but there is a reason behind it. My internship for the summer is with, and I've been busy doing projects with them. I've had a couple blog posts, but the big contribution so far has been working on the new Tennis Rankings and Wimbledon odds. They went up today, and I'm pretty proud. If you're looking for a fix from The Taste, here's where to go:

Sunday, June 19, 2011

Margo Dydek

A few weeks ago on May 27th former WNBA #1 overall draft pick Margo Dydek passed away from a heart attack. It was a tragic death for many reasons, including the fact that she was just 37 years old. Although her name may not be familiar to average sports fans Margo was one of the most unique basketball players of all time. She was picked first in 1998, and despite leading the league in blocks for many years never quite became a star. Although her stats don't set her apart, something else does: Margo was 7'2". 7'2" is extremely tall for a guy (an inch taller than Shaq), but astoundingly tall for women. If Margo were the same height as a male, how tall would she be?

For people of average height, a good rule of thumb is a +5 inch rule; add five inches to a women's height, and that's roughly how tall she'd be as a man. This would put Margo at 7'7", on par with the two tallest NBA players of all time, Manute Bol and Gheorghe Muresan. It turns out however that the standard deviation of female and male heights is not quite the same. According to a study from the early 90's, American men have an average height of 69.3 inches (5'9") and women average 64.1 inches (5'4") (Margo was actually Polish, but we'll ignore that for now). For men the standard deviation was 2.92 inches, but it was just 2.75 for women. This means that when you get to the upper ranges of height (tall basketball players) it becomes a +6 inch rule, making 7 foot tall men roughly equivalent to 6'6" women.

For Margo, the rule is even slightly higher. With a height of 8 sigma greater than the mean (statistically impossible basically) she outpaces even the tallest male NBA players. This chart shows how she compares to varying men.
Margo's equivalent height among men would've been an amazing 7'8.6". This height was likely less a blessing than a curse though, allowing her to play professional basketball but probably contributing to her health problems. RIP Margo Dydek, the tallest basketball player of all time.

Monday, June 13, 2011

Dallas Mavericks: NBA Champions

Congratulations to the Mavericks on their first ever NBA championship. They are one of the few teams that fully embrace the statistical side of basketball, and it paid off for them in the finals. They over came long odds to win it all, but somehow pulled it off with only 5 losses along the way. I'll be wearing the Dirk jersey for about a week straight in celebration. Well done Dallas.


Wednesday, June 8, 2011

2011 NBA Finals Odds-Update!

Through game 4, these are the likelihoods of the remaining outcomes:

Miami in 6: 31.5%
Miami in 7: 34.0%
Dallas in 6: 17.6%
Dallas in 7: 16.9%

It sure seems like the Mavericks have made quite a statement already taking two games from the presumptive favorites. The truth is though, 2-2 was the most likely outcome at this point in the series. Dallas's odds are better now, but not by much: Miami is still expected to win 65.5% of the time. Some significant work has been done, but the big game happens tomorrow as over half the series swings on game 5.


Dallas's comeback win last night brought the series back to 2-2, and is promising to make this an especially memorable finals. People have torn apart Lebron and Jason Terry's play, while praising Dirk and Dwyane Wade. Turns out though, sports reporters have missed the single best predictor in the series: whether or not I wear my old Mavericks jersey. I wore it for games 1 and 3, but not for 2 and 4. Clearly me not wearing the jersey corresponds exactly to Dallas winning. Is there something there? Or is this an example of me having to much time to think of wacky things now that finals are done?

Traditionally, statisticians use 5% as the probability cutoff for correlation. If there's less than a 5% chance something would happen randomly, they assume it's a non-random process. The chances that my fashion choices would correctly predict 4 games in a row is just 1/16, or about 6.3%. That's not statistically significant, but it's close. If my jersey gets one more game right however, the odds move to 3.1%, a level that actually is significant. I won't be wearing the jersey for game 5 (go Mavs!) so a Dallas win will be more than just an important finals game, it'll be validation for all who ever believed that their particular scrap of cloth influences their team's play.

*Okay, clearly my shirt doesn't influence the game. What's wrong with this analysis?
First, I set the null hypothesis halfway through. Flip a coin 5 times, then hypothesize those results happen, and your test will be significant. If I say my jersey is a predictor one way or another, it could predict a Dallas win or loss, so 6 games is the level of significance, not 5.

Second, correlation does not imply causation, so it's possible some non-random third party is influencing both. Maybe I put the jersey on only when Dallas is losing (subconsciously). Or maybe Dirk watches what clothes I wear and then plays accordingly just to mess with my head.
Finally, even if something happens only 3.1% of the time from random chance, a random distribution will still give that answer 3.1% of the time. If you have 32 octopi picking 5 games, one of them is probably going to pick all 5 right. Then you can name him Paul and get the whole world to watch him.
Nonetheless, even though it's not true, I still kind of believe in this stuff. The truth is, I won't be wearing the Nowitzki jersey for game 5 because I want Dallas to win. But I won't be placing bets on it.

Thursday, June 2, 2011

Ranking Shaq with the Centers

Only Shaquille O'Neal could steal attention from an exciting NBA finals as an over the hill 39 year old. His impact has been humongous not only on the court but off it too. Just look at the name at the top of the page (Kobe, Tell Me How My ? Tastes!). Watching him over the last 5 years it's easy to think of him just as an outsized personality, but he was maybe the most dominant basketball player of all-time. In the all time pinnacle of centers, where does the Big Aristotle rank? I created a cumulative index to try to find out. And then I changed things a little bit (I'll explain for each change).
My index was a combination of my work on the Top 5 each year, career win shares, MVP share, Basketball-reference's Hall of Fame odds (to include championships), and BBR's Elo player ratings. I altered things a little bit to help some defensive studs out, but pretty much stuck to the results.
There were two big names who maybe should be on the list. Arvydas Sabonis never got a chance to show the world what he could do in his prime, but some suspect he could've been the best center ever. Also, I left Tim Duncan out, even though he's really more a center (I'd probably put him right behind Shaq).

10. Dwight Howard (5 points in the index)
He proved himself the second best player in the NBA, and with it moved himself into the top 10, knocking Bill Walton (or Willis Reed) out. He and Ewing were the only centers under consideration who haven't won a championship, and everybody else besides Malone has won at least two. Howard's entering his prime, and his time will come.
9. George Mikan (7 points)
Mikan is without a doubt the hardest player to rank on this list. He played in a completely different era, and probably would play something like Aaron Gray if he were in the league now. At the same time, he was the NBA's first superstar, and the center piece of the leagues first dynasty.
8. Patrick Ewing (9 points)
He will always be known for failing to break through (or for his Snickers commercials) but he took the Knicks to the finals twice, and led a renaissance of New York basketball. It's just tough to go up against MJ in your prime.

7. Moses Malone (25 points)
Moses Malone is often a forgotten name among centers, quickly replaced by Olajuwon in Houston and never living on Dr. J's level in Philly. He peaked in a way that few players ever have. According to my top 5 calculations, Moses was the best player in the NBA in 1978-79, 81-82, and 82-83.
6. David Robinson (28 points)
Robinson was a physical specimen, and is being appreciated even more with advanced statistics. He consistently shows up at the top of Career PER or win shares lists, and can be seen by some methods as one of the best players of all time. He needed Duncan for playoff success though, and in his prime got it taken to him by the next man on this list.
5. Hakeem Olajuwon (24 points)
While his regular season stats aren't quite on par with Robinson or even Malone, taking both of the titles during MJ's break earned him a top 5 spot among greatest centers ever. He beat both Patrick Ewing and Shaq in those finals, and the way he dominated NBA MVP David Robinson forced me to give him a bump up this list.

4. Shaquille O'Neal (38 points)
Shaq was the best player in the NBA in the immediate post Jordan era, ruling over the league in 1999-00 in a way that few ever have. He could rule games physically, but also had great touch around the hoop. He never quite got all he could out of his basketball potential, but winning 4 rings and being an absolute force for 3 different teams still put him 4th on the list.
3. Bill Russell (37 points)
Russell helped integrate the sport, anchored the greatest dynasty in the history of athletics, and my be the best defensive basketball player ever. A player with 11 championships will never be seen again in any high level team sport.
2. Wilt Chamberlain (44 points)
He occasionally cared more about women or statistics than about winning, but he also once averaged 50 points and 27 rebounds per game, over an entire season. Nobody has had a game like that since him, and that what he used to average! Wilt the Stilt also managed to squeeze in a couple chips around Russell's hogging.
1. Kareem Abdul-Jabbar (46 points)
Kareem was the best player in the NBA for 7 total years, and was the best player in the league for 4 straight seasons from 1970-74, a streak matched only by Michael Jordan and now Lebron James (didn't see that one coming huh?). The NBA's all-time points leader is also the greatest center of all time. For now.