top of page
Search
Writer's pictureJason L

When to Stop Listening to the Data and Embrace Guessing

I think the word "guessing" has too negative of a stigma. For the most part, most of our important decisions are guesses -- choosing a career, a life partner, a car, even what flavor of ice cream to buy. No matter how much data you have already consumed and may still have at your disposal, there is eventually that point where you either cannot or should not seek more information before proceeding.

  • A career choice comes with the sense of finality (choosing to work in analytics often involves rigorous training and studying that would likely disqualify you from later applying to work in accounting).

  • A marriage proposal comes with the sense of urgency (you can't realistically ask to go date a few more people just to be sure when the person in front of you is already down on one knee).

  • A consumer choice comes with the sense of practicality and diminishing returns (you certainly could spend all day evaluating chocolate vs. vanilla but will likely cost you a fortune in opportunity cost).

As Warren Buffet once said, "You pay a high price for a cheery consensus."


The 1 Percent Rule


I was formerly a professional poker player in my early 20s. Poker is littered with imperfect information, in that each player only has limited visibility into all of the game's many variables. What makes the difference between elite poker players and novices is the ability to quickly parse the known information, compare to their history of comparable situations they have either played or studied, and derive solutions against the range of potential unknown information (not unlike a Monte Carlo Simulation, similar to my NFL Draft forecast).


What might surprise some is that most elite poker pros do not come to one optimal solution for most poker hands, but in fact they come to what is known in game theory as a mixed strategy. This means they can potentially arrive at the exact same decision point two different times and optimally make competing decisions both times (they may call the bet once and fold the other time). Mixing their strategy often involves some form of randomization to remove bias away from always choosing one decision over the other. But isn't randomization just the consequence of a blind guess? Ah, you are catching on now!


Elite poker players are not mind readers (even if the TV shows might try to suggest it sometimes). The goal of a poker player is to be just slightly better at guessing than their opponents given the same amount of information, and then extrapolate that edge over thousands of smaller decision points. To even just be as little as 1% better at guessing than their opponent can lead to consistent long-term advantages. James Clear dubbed this "The 1 Percent Rule" in his book, Atomic Habits. Clear also suggested that The 1 Percent Rule leads to a winner-take-all effect where rewards for the advantaged tend to be outsized as time goes on.


To demonstrate, consider a basic guessing game where Player 1 is just 1% more informed on guessing than Player 2. Might not sound substantial, but the results for Player 1 after 100,000 decisions on average would look like this:

This run was randomly generated, yet is actually a perfect simulation of the life of a poker player because of the early demonstration of the variance in luck. Player 1 spends about the first 15,000 decisions in the negative despite having the theoretical advantage, yet steadily cleans out Player 2 over the last 85,000. The Law of Large Numbers supports that as long as even the slightest edge in decision-making exists, then the long-term destination of the game was never in doubt even when the short-term results might have suggested otherwise. Also notice the winner-take-all effect that Clear had written about. The units won for Player 1 start out small but eventually ends up winning +1,000 units from Player 2, even with just a tiny 1% advantage.


Suffice to say that we do not create our edge just by blindly guessing. We create it in the preparation before we guess.


Embracing the Unknown


I have learned to embrace the unknown of guessing at major inflection points throughout my own life. For as much as I enjoy the comfort and assurances that can come from structured data, more often than not it's either not available or not enough to provide a path of least resistance. The point where we need to consider guessing usually comes when one of two conditions are true:

  1. There is no more known information.

  2. There is infinitely more known information (or at least seemingly so).

I want to focus more on the latter. When there is no more known information, then there's not much more we can do to better inform our decision. The poker player facing a big bet on the river can only deconstruct the previous actions in the hand and then decide to call or fold. The player does not have the option to seek more information and defer the final decision until later.


One place I recently found things to be different was in searching for a new rental property. I already own several properties, but I had to approach this one from scratch because of budget limitations taking me out of the area my other properties were in. I needed to find another city that profiled similarly to that area, but still early enough in its expansion cycle to fit the budget I had now. That's a daunting task when your canvas is the entire United States. Luckily, I had plenty of data at my disposal that I looked into between real estate, rental, and economic data. I narrowed down my list to five major cities in the US that would fit all of my initial criteria for a city I might want to invest in (note that I am not advising that any or all of these places would necessarily be a good place for you to invest; only that it made it through my own initial screening):

  1. Tulsa, Oklahoma

  2. Greensboro, North Carolina

  3. Kansas City, Missouri

  4. Indianapolis, Indiana

  5. Mobile, Alabama

All five cities were complete unknowns to me, as I had never been to any of them and really had nothing to associate one from another. Most of the data I had pulled was fairly similar, in that there was no real smoke signal to lead me towards picking one immediately. I was pretty overwhelmed at this point because it didn't seem as if anything was going to stand out from the data I had available. I looked at local investor forums on BiggerPockets to see what others were saying. I connected with a realtor in each market, but they all basically said the same rosy things about how hot their city's market was. I was hoping an answer would eventually present itself to me. Weeks went by without any real movement towards any direction when I finally just had the epiphany:


"I should just shut up and pick one already."


I really felt I was at the point where I was looking for the signal that was not coming or would not be worth the time to search for more data. The truth about the latter point is that even if I used more data to find that signal, then who's to say a different set of data might provide a conflicting signal and circle me back to the start again? It felt like I was already at the point of equilibrium. All five cities had hit my initial criteria. If I trusted in my initial criteria, then how badly could I realistically do if I just picked one at random? It's just money, right?


Improving by Failing


Mistakes are inevitable with imperfect information involved. Poker players frequently get major decisions wrong, as do real estate investors. My first choice -- at random -- was Greensboro. Literally, at random. The final decision went something like, "I guess Greensboro sounds okay?" It was about as thought-out as The Simpsons episode where The Itchy and Scratchy writers had to come up with a name for Poochie The Dog:

I told the realtor I had previously connected with of my intentions. At his behest, I booked a weekend flight to come see the city in person. There were no direct flights, so I had to fly to Charlotte, rented a car, and then drove 100 miles each way to Greensboro. Long story short, the realtor that invited me to come from Florida to look at houses that were 10 minutes from his office with him had then forgotten I was coming. About $500 in write-offs and 12 hours of travel later, I saw the inside of Greensboro's Whole Foods more than I saw the inside of any potential leads.


Despite the obvious annoyance and disappointment, the trip was not a total loss because I did pick up two additional data points that I didn't have before which would aid me in continuing my search. The first data point was the newfound desire for a city with a direct flight. The trip from Florida to North Carolina would be a really easy one had it been direct, but having to either layover or drive some of the way essentially made the trip about as far as Florida to California would've been directly. That isn't something you'd really think about much until you actually do it. I scanned the original list of cities and realized that only one had a direct flight path out of West Palm Beach, which was Indianapolis.


The second data point I picked up were the early warning signs of a bad realtor. In fairness, I shouldn't necessarily refer to the Greensboro realtor as "bad" at his job. He may actually have been highly competent but his customer service was atrocious, and I knew that element was especially critical as an out-of-state investor. However, I probably should have cut bait on him sooner than I did (I kept communication with him for several weeks after the failed trip, mostly chasing my own losses from the weekend I spent there).


Even once I decided to pivot to focusing on Indianapolis, it still took me three attempts to find the right realtor for me. The big adjustment I learned from Greensboro was to have quick hooks on the first two. If they showed warning signs early in our relationship (the time where they theoretically should be working the hardest to earn my business), then I moved on instead of chasing the losses of the time already put in. Thankfully, my third realtor in Indy turned out to be a winner.


What had started months earlier with a blank canvas of the entire US and an endless amount of data to potentially absorb (not to mention a global pandemic in between), concluded with me finally closing on my house in Indianapolis back in September. I will never really know if I would have made a better return if I went with one of the other cities. However, I also did not let the uncertainty paralyze me into never moving forward with any of the options until I was sure because that assurance was likely never coming. It helped in trusting my initial screening because I was not a total novice to the real estate process, but I also gained more confidence in my decision-making from the bad experience in Greensboro than I did from any spreadsheet.


I am not here advocating for flipping coins on significant life or business decisions. I am saying that if you're stuck evaluating a seemingly equal decision, then remember that you might gain more clarity from guessing wrong than you will waiting for the answer to present itself. Just make sure you have put the time in before blindly guessing at it.


0 comments

Comments


bottom of page