Apr
8
Runs in Sports, From Jim Sogi
April 8, 2008 |
I heard a spot on NPR yesterday about a statistics study of major league baseball that concluded that all the runs and records in major league baseball were consistent with randomness. I find that hard to believe, and that brings up the general feeling that many of statistics' conclusions are hard to believe. But there they are. A random generated time sequence has many apparent trends when looking at a chart. The tests of market statistics seem hard to believe, but are verifiable within its model. This phenomenon highlight the problem of purely subjective trading and the place of feelings in trading. Vic and Laurel's book talks about runs and hot streaks. Many of the conclusions arrived at scientifically are counterintuitive at first. It's like the greatest pain entry point of a trade is often the best entry.
Victor Niederhoffer remarks:
Jim is talking about the randomnes of basketball, and how the runs and streaks are consistent with randomness. They're consistent the same way the market is consistent. If you don't take account of the lead at the time or the changing significance as other markets move and time passes within the day and week, and you're an ignoramous, then yes, things are completely random.
Comments
12 Comments so far
Archives
- May 2013
- April 2013
- March 2013
- February 2013
- January 2013
- December 2012
- November 2012
- October 2012
- September 2012
- August 2012
- July 2012
- June 2012
- May 2012
- April 2012
- March 2012
- February 2012
- January 2012
- December 2011
- November 2011
- October 2011
- September 2011
- August 2011
- July 2011
- June 2011
- May 2011
- April 2011
- March 2011
- February 2011
- January 2011
- December 2010
- November 2010
- October 2010
- September 2010
- August 2010
- July 2010
- June 2010
- May 2010
- April 2010
- March 2010
- February 2010
- January 2010
- December 2009
- November 2009
- October 2009
- September 2009
- August 2009
- July 2009
- June 2009
- May 2009
- April 2009
- March 2009
- February 2009
- January 2009
- December 2008
- November 2008
- October 2008
- September 2008
- August 2008
- July 2008
- June 2008
- May 2008
- April 2008
- March 2008
- February 2008
- January 2008
- December 2007
- November 2007
- October 2007
- September 2007
- August 2007
- July 2007
- June 2007
- May 2007
- April 2007
- March 2007
- February 2007
- January 2007
- December 2006
- November 2006
- October 2006
- September 2006
- August 2006
- Older Archives
Resources & Links
- The Letters Prize
- Pre-2007 Victor Niederhoffer Posts
- Vic’s NYC Junto
- Reading List
- Programming in 60 Seconds
- The Objectivist Center
- Foundation for Economic Education
- Tigerchess
- Dick Sears' G.T. Index
- Pre-2007 Daily Speculations
- Laurel & Vics' Worldly Investor Articles
I don't know a lot about statistics but I know alot about baseball. Of course, streaks occur and they occur all the time. Ask any baseball player. By the way, they are probably the most superstitious of all athletes and they will tell you that when you are on a streak you respect the streak. This is from Bull Durham the movie when Kevin Costner is explaining this to Susan Sarandon. He told her that if you think that wearing womens garter belt or breathing through your eyelids is the reason why you are on a streak then guess what you wear garter belts and try to breathe through your eyelids. A great line in A League of Their Own, Tom Hanks who plays the manager enters the clubhouse and inquires "What is that smell? One of the women replies "That is Alice, she hasn't washed her socks since we started winning." Hanks replies "Alice, your killin' me with those socks, you're G*d D**n killin' me! Due to the nature of the game, baseball lends itself probably to more streaks than any other sport. In all likelihood it is because baseball games are played every day so it is easier to keep momentum or lose momentum that say football. Some of the greatest streaks in baseball Joe Dimaggio 56 game hit streak in 1941. This fascinatingly was also the same year that Ted Williams hit .406. Ty Cobb hit over 300 for 23 straight seasons. The New York baseball Giants won 26 games in a row in 1916. The evil empire aka the New York Yankees won 5 straight World Series from 1949 to 1953. Don Mattingly of the New York Yankees hit a home run in 8 consecutive at-bats in 1987. In the following year Orel Hershiser of the Los Angeles Dodgers, pitched 59 2/3 scoreless innings. This surpassed the previous record which was held by Don Drysdale coincidentially also of the LA Dodgers. Johnny Vandemeer pitched 2 consecutive no hitters in 1938. On the side of futility. The Chicago Cubs last won the World Series when Teddy Roosevelt was President in 1908. The second longest streak was the Boston Red Sox who won the World Series in 1918. This of course was the year that they traded Babe Ruth to the Yankees. Anthony Young of the New York Mets lost 26 straight games in 1992 to 1993. The point is made. However the problem with streaks is that they are impossible to predict when they will start and when they will end. I am sure some gamblers can offer some theories as to good ways to bet streaks however, I am not one of them. As they say in baseball " Some times you win, some times you lose and sometimes you get rained out." P.S. the information obtained above is from a variety of sources. For those stats people out there. sl.
Kelly Slater sure had a good run in surfing. I don’t think there was any randomness about his winnings. I remember watching him surf at Sebastian Inlet when he was just a grom, and being blown away by his raw talent. He’s an athlete on par with Michael Jordan, Joe DiMaggio, Wayne Gretsky, Tiger Woods, or Gayle Sayers. Too bad that surfing doesn’t get any respect….wait….I’m glad surfing doesn’t get any respect.
Jeff
Rather than random, I believe market activity to be chaotic. It seems to swing from stable (trending) to unstable (reversing/sideways) states in an unpredictable fashion, and over many such transitions, would probably seem to be random. I think the key is in trying to understand the feedback mechanisms that hold the stable periods (trends) in place. Not that I’ve found that key.
Cheers,
George
The NPR segment probably discussed this very interesting op-ed in last week’s NY Times– http://www.nytimes.com/2008/03/30/opinion/30strogatz.html The essay contained simulations on baseball history to determine how unlikely DiMaggio’s 56-game hitting streak was:
“…The sport’s most mythic achievement is Joe DiMaggio’s 56-game hitting streak, a feat that has never come even close to being matched…In a fit of scientific skepticism, we decided to calculate how unlikely Joltin’ Joe’s achievement really was. Using a comprehensive collection of baseball statistics from 1871 to 2005, we simulated the entire history of baseball 10,000 times in a computer…we used baseball statistics to calculate each player’s odds, as determined by his actual batting performance in a given year…More than half the time, or in 5,295 baseball universes, the record for the longest hitting streak exceeded 53 games…In other words, streaks of 56 games or longer are not at all an unusual occurrence…”
I’m somewhat skeptical of the calculations. Removing pre-1900 data and making allowances for the inconsistency of plate appearances could motivate very different results. Also, the results in this op-ed match DiMaggio’s hitting streak, but it’s unlikely they match the magnitude by which DiMaggio’s hitting streak exceeds that of anyone else. As a result, the corollary conclusion would be that everyone else’s hitting streaks are less than they should have been, an inconsistent notion to the thesis of this piece.
Referenced in the NYT op-ed is this essay by Stephen Jay Gould that is my favorite essay on baseball and has a rich discussion of cognitive biases– http://www.nybooks.com/articles/4337
This is not the first time this subject has been covered on this site but I have yet to read an adequate explanation. What does it mean to say all streaks in sports are consistent with randomness? Does it mean relative to all pro basketball players Jordan is within some standard error, some outlier that can be ruled out as not fitting within some predetermined distribution of talent? Tiger? Brady? Great teams like the Habs or Yankees just a some happenstance of players that magically clicked with no intelligence on the part of the management that traded, drafted, signed them? What are we to make of Victor's reliance on his racquet sport past for trading ideas? The whimsy of a man that is logically inconsistent? What are we to make of the mathematical axioms that are so dramatically at odds with common sense? What if someone were to postulate that saying "something is consistent with randomness" translates into plain English as I haven't bothered to consider all the possible variables long enough to create a cogent theory. Could someone give a variant theory of randomness where Jordan or Tiger are actually predictable nay even predetermined outcomes of dedicated effort from an early age? Someone like say Laszlo Polgar who trained all three of his daughters to become master chess players? Ayn Rand's ultimate commandment is a is a, reality is the final arbiter. So perhaps Victor and the rest can disregard Jordan, Woods, (Buffet?) as some fiddler on the roof — a black swan so to speak. But reality does not ignore them, the world hangs on their every move…
Dear Merovingian,
It’s useless, this question you ask. Not that I begrudge your doing so, but you will find soon enough that this clique disdains to respond to the less enlightened.
I thought to make a similar query, but abstained as Nigel seems a bit tightly wound as of late and it would pain me to consider that a realized apoplectic seizure was, even vaguely, due to my curiosity.
As well, you might reconsider using the term “Black Swan” in future comments. This terminology much sets some in the room on edge, so much so as to exhibit a pathological aversion for liquids.
I do hope you’ve your Kevlar at hand. You’ll know why soon enough.
Cheers,
lon (the grinning bear)
In 1996 Brady Anderson of the Baltimore Orioles, hit 50 home runs. Collectively he hit a total of 74 home runs in the four prior seasons. In 15 seasons in the major leagues he hit a total of 210 home runs.
Interesting phenomena about runs and streaks happens in poker regularly. In a Texas Hold-Em game the odds are 1-221 or thereabouts that you will be dealt a pair of aces. If you hold K-K there is about a 2.4 per cent chance that someone else at a full table will hold Aces. In a showdown hand if you hold A-A you are a 4-1 favorite to win the hand against any other pocket pair. Yet there are times when I have seen the same player get blazing hot and catch huge monster hands and become in poker vernacular a card rack. Recent winners of the feature event of the World Series of Poker have all displayed this most notably Chris Moneymaker and Robert Varkonyi both rank amateurs. They both took the title and were not much of a factor in the poker world after that. Jamie Gold 1996 champion won the title by defeating nearly 9000 others largely due to good play and great fortune.
In fact, the last time a true professional won the WSOP main event was in 2000 with Chris Ferguson.
It has been said that to win a tournament is 90% randomness or luck (if that is a better term) and 10% skill. To win over a course of a year is 90% skill and 10% luck.
I am a low single digit golfer and I might be able to beat Tiger Woods in a 9 hole contest if everything fell in place. In a 72 hole event barring some act of God, forget it.
sl.
Many trainers and jockeys at the horse track go through amazing streaks of wins AND losing.
I guarantee you it has virtually ZERO to do with randomness and I suspect the slide rule crowd could probably cook up something to claim that it does.
When a computer simulation can ride or train a horse, or hit one out of the park, I’ll be more impressed.
The birth of the individual capable of greatness is a perfectly random occurance. His, or her, rise to greatness is not.
The former is an act of God; the latter of Man.
We can never know, fully, God; we can only too well know man.
lon
Dear Lon,
Don’t worry about me I’m just an exile - a glitch in the matrix so to speak. I’ve been enjoying your old posts by the way, Gold + the Genome and the John Law situation. If you’re ever in Toronto we should discuss.
All the best,
M
Mervongian,
As I exist the perpetual wanderer, I imagine that soon enough I will find myself in your neck of the woods. Given that eventuality, yes, let’s discuss.
And, by the way, thanks.
lon
Absent any information save the previous shot(s) going in or not what can we conclude? Answer: not much. On one hand I understand the value of such an inquiry (especially to shake the trend following habit) on the other there sure seems to be a lets ignore the forest and find the trees aspect to all of this…