This was a very interesting article.

"A chinese mathematician figured out how to beat anyone at rock-paper-scissors":

"The pattern that Zhijian discovered — winners repeating their strategy and losers moving to the next strategy in the sequence — is called a "conditional response" in game theory. The researchers have theorized that the response may be hard-wired into the brain, a question they intend to investigate with further experiments."

Jordan Low writes: 

If I am reading it correct, the losers are basically choosing to display what won recently. Isn't it like using 3Y returns to pick mutual funds? Or did I get it wrong?

Steve Ellison writes: 

I don't play that game, but a good strategy might be to make selections as randomly as possible, for example by memorizing long sequences of digits of irrational numbers (Arthur Benjamin has a memory aid for how to do so).

Bill Walsh, the coach of the San Francisco 49ers football team in the 1980s, scripted the first 25 plays of each game in advance. This strategy made it harder for opponents to guess what play might be coming next. I have an idea that, if I could ever find enough trading systems with positive expectations, it might be good to randomly pick a sequence of systems to use in advance, in order to keep the crocodiles guessing.



 Back in the day in the pits, if one could predict when the market was going to go from half bid to half sellers with a 90% confidence rate, one had a huge edge. That was not uncommon either. Since all the visual and auditory clues have disappeared, along with open outcry, things have changed because of the screen. But it is still possible to tell when a market is going from half bid to half seller's and the edge remains for the patient. To capture the edge, you need to adapt and think outside the box. It's not as easy as when we traded against order flow, but it's not impossible.



Speaking of trends how does the May-Oct (April 30-Oct 31) period perform as a function of the YTD performance as of April 30?

SP500 (1951-2013) Dec 31-Apr30 returns were sorted:

"-" is down more than 2%. 
"+-" was flattish, return between -2% and +2% (2014*). 
"+" was up more than 2%.

Sorted this way, the returns for the 6 months following YTD-Apr 30 appear to continue the trend:

One-Sample T: May-Oct-, May-Oct+-, May-Oct+

Test of mu = 0 vs not = 0

Variable      N     Mean     StDev   SE Mean       95% CI             T
May-Oct-   17  -0.0408  0.12458 0.0302  (-0.1048, 0.0232)  -1.35
May-Oct+-  11   0.0059  0.0897  0.0270  (-0.0543, 0.0662)   0.22
May-Oct+   35   0.0389  0.0680  0.0115  ( 0.0155, 0.0623)   3.39

* YTD as of 4/30/14 is 1.93%



 The documentary for rent on itunes about Muscle Shoals recording studio is absolutely amazing and a requirement for any music buff. It rivals the 20 feet from Stardom, and the Funk Brother's stories. The movie documents a small recording studio in Alabama, it owner and producer, Rick Hall, and their amazing studio musicians They produced well known recordings such as Wilson Pickett's Mustang Sally, Aretha Franklin's RESPECT, Rolling Stones, Bob Dylan, Paul Simon, Dire Straights, Lynard Sknrd, Steve Winwood, Joe Cocker, and the list goes on and on.

« go back


Resources & Links