"My team is on the floor." -Norman Dale
My wife knows if I’m up late at night, attracted to a new friend online, she can roll over with nothing to worry about because it’s probably someone with a Ph.D. in chemical engineering from Stanford who took a deep dive into data, like Ed Feng. After a surprising turn, he landed in an un-crowded pool of new opportunities in sports analytics and I belly-flopped right into it. He had me at “Data reveals hidden truths.”
Ed wrote a book on solving March Madness brackets. I look at the Stock Market as one giant bracket with teams of tickers. For more than twenty years (after getting promoted from the mail-room) one of the funnest parts of my job turned craft is when I host tournaments of stocks at the end of every single month with fresh data. Rather than predicting what should happen, we untangle what is already happening. Businesses are winning and losing every day. Over time, some of the most consistent operators with innovative playbooks will grow to dominate their competition.
For several big winners, and strict sell disciplines on far more losers - math can do the heavy lifting. My IT guy couldn’t understand why I wanted a photo of where we secure our servers offsite. I just wanted to give credit to my bracket busters.
I owe my entire career to applying only one subjective layer on top of math, the only skillset I was blessed with – curiosity. I am glad I was never smart enough to be convinced of anything. Whatever field math takes me to, from there, I believe whoever turns over the most rocks to study deeply usually wins.
“But Professor, these are the same questions you used last year. Won’t the students already know the answers?” (his secretary)
“It’s all right, you see the questions are the same but the answers are different.” -Albert Einstein
Ed and I had this in common as well. He says he owes his career to being so curious during a research project, that it cost him any future in academia. He is thankful for that lucky mistake. “I’m much happier now,” he told me.
A few notes I made from Ed’s book (thepowerrank.com)
In 2008, he was reading a paper on Google’s PageRank technology. The algorithm that brought order and insight to the complex world of web search. The basic idea was that a website was important if another important website linked to it.
Teams are webpages, and games are links between the teams. In sports, teams earn credibility for beating other good teams.
Then he applied his own research to find margin of victory was more important than records.
He taught his computer how to solve 351 equations simultaneously for 351 variables, one for each team.
His simplest advice is that none of the little upsets matter, but crowds obsess over finding them. You have to pick the tournament winner. Are we talking about teams or stocks? Yes!
Ed makes a clear distinction in his analysis. The best odds are in small pools where you simply need to calculate the best odds throughout. I think this is no different than overwhelmingly the best odds of long-term success for most investors in the Stock Market is indexing, so you know you have all the winners.
Ed explains that his final pick changes if playing in a larger pool. Then, with the same data he will apply one contrarian filter at the end. For example, this year his numbers have Duke’s chances to win at a historically high 33.7% likelihood. But, even higher is the public’s opinion with 40% of brackets entered on ESPN picking Duke.
According to Ed’s numbers, Virginia has a 21.9% chance of winning. A recency-biased fueled public remembers their choke job last year so only 7% of the public brackets picked them. While much less disparity, Gonzaga also has greater odds than consensus believes.
Gonzaga meets Virginia in the Finals in my bracket. This almost certainly will not happen. But if it does, I have a good chance to win far more than I would lose.
There is an unspoken ingredient I call JOMO that is required in any analytics stew. The joy of missing out. If you stick to odds-tilting analysis you must be perfectly satisfied losing. To be clear, I am not normal. Not many people find joy sorting through losses and running back to the nerd cave, bloodied but full of curiosity to see what can be tweaked and improved. Survive and advance.
This month’s stock tournament is honoring a vintage basketball concept I grew up with. It was called a 2-point shot. I think in basketball and stocks, 2's are now underrated. When my wife heard me squeal later that night it was when I found “Non-Transition Effective Field Goal %” rankings and “Points from Assisted Field Goals at the Rim %" rankings. These two tell me who runs the best half-court sets and shares the ball – perhaps the closest clue to finding chemistry. A treasure trove of this data is run by Jeff Haley (hoop-math.com). He knows all about the science of research and development. His day job is at a chemical company, until a university or NBA team hires him away.
In the Stock Market, I have found many of those big winners started as humble 2’s. Not the highest dividend yields, but the fastest growing dividend payments can often be found hanging around the basket of 2% yields. So, I kicked out all the higher yields. I also crossed out companies paying out too much of their earnings, and those not generating a lot of free cash flow to afford large dividend increases.
Teams have to play better offense and defense to win games consistently, and reward stakeholders every year. I have never understood why anybody would want a high growth stock with a ridiculous valuation any more than a cheap stock that is not growing. Why not be patient enough to find both at the same time?
Margins of victory are important. Are the top and bottom lines of growth better than the field, but also accelerating versus its own growth rate? Regardless of how healthy your profit margins are, I want to see if they are still able to expand. Then, is a stock actually beating teams on the court? Recent out-performance of the stock price, relative to its sector, is different clue. I respect fundamental analysts who do not care what technical analysts think, and technical analysts who do not care what anybody thinks. But, I prefer to make stocks compete against each other to find out who can win both.
The final game, after miraculously surviving each one of those rounds, is which remaining team has not yet been rated as a “buy” by the betting public consensus of Wall Street analysts. Superior teams that are also un-loved have my attention.
“It is what you learn after you know it all that counts.” -John Wooden