By Ray Wallin
How many ping pong balls would it take to fill a 747? How long would it take to count to one million? How many hours will you sleep in a lifetime? How many piano tuners are there in Chicago?
Right now, you think that I am either insane, off my meds, or both. Yet companies like Microsoft, Google, and Goldman Sachs use questions like these with potential employees in their interview processes. These types of questions show the applicant’s flexibility to think on their feet and apply math skills to real-world problems.
These problems are also called Fermi Problems after the great Italian physicist Enrico Fermi, who created the world’s first nuclear reactor and was one of the few physicists to excel in both theoretical and experimental physics. Fermi often challenged the students in his classes with questions such as these by using estimation, common sense, and reasoning to work out problems that were almost impossible to measure. The point wasn’t the answer the students would come up with, but rather the process they used to get to the answer.
By now you are wondering how trying to figure out the number of piano tuners in Chicago can help you make your living playing the races or at least help you win a few bucks at the track. Don’t be skeptical, I once wrote a college essay where I not-so-successfully compared Virginia Woolf to a fish.
This is a word we jokingly use in my line of work, estimating budgets on large commercial construction projects. Often, I am presented with some high-level information – the use of a building, the overall square footage of the building, and what the client or designer wants it to look like. Without any drawings or design documents my colleagues and I have to come up with a narrow range of potential costs.
The information I am given is not perfect, but neither is the information we use to handicap horse races. We can see how the horse did last race, who the jockey was, what the conditions of his previous races were, and a short blurb about how he ran in that race. Beyond that you are left to guesstimate his current form, how he is feeling today, if he has been eating well, and what else may be going on behind the scenes that will help to improve or hinder his performance today.
When I select pace lines for comparisons between horses, I look for the most applicable running line. It may be the last race, two back, or nine races back. I am effectively guessing at how the horse will run today through by reasoning through previous running lines and selecting the most likely.
Some of the criteria I use to select the most likely running line:
Was this past race well above or below today’s class? In either case the running line is more than likely a toss. A horse that takes a field of $5,000 claimers wire to wire and then runs in an allowance race today likely doesn’t stand a chance of having that sort of dominance. Likewise, a $10,000 claimer that failed miserably last out against an allowance field should most likely get a pass for that dismal effort since he was completely outclassed.
When looking down a career sprinter’s running lines I look for what are the most comfortable fractions for him. If you have a horse that consistently runs well (or well enough) in races where the half mile call is in 46 3/5 seconds and you see a race where the half mile went in 44 1/5 seconds, you can toss that race. Likewise, he didn’t fare too well in that pace-less race that hit the half mile in a whopping 48 3/5 seconds.
That early speed horse that couldn’t make the lead in a race where the class was close enough and the fractions where in his wheelhouse often raises a red flag. Before you sound the alarm or use this race for comparison, you need to look at the comments section to see if he was bumped out of the gate, checked in the turn, or had some other sort of trouble. If he does have a legitimate excuse, don’t include this race in your analysis.
This is the most basic rule I use when selecting a running line. It today is a sprint on the dirt, you shouldn’t use a route or a turf race for comparison. While there are par time adjustments available, they are different types of races. My lone exception to this is when you have a one-turn mile race, I treat this as a sprint.
Like most weekend and evening horseplayers, I don’t have a lot of time to commit to handicapping the entire race day card at any track. One strategy of solving Fermi problems is to use effective guesstimation before going into a deeper dive. This is part of the triaging I do to a race card where I look for my best opportunities and attack those races first.
I also like to take a high-level look at the race and scan through the Quirin running types and scan down the fractions that each horse seems comfortable with. I am looking to find if there is a favorite that is mismatched or if there is a horse that has an apparent pace or class advantage given my quick back-of-the-envelope review of the race.
When I find either a race that presents a good opportunity or a horse that appears to have an advantage against the rest of the field, I will take the time to dig deeper into the race. So before you put too much effort into analyzing a race it is good practice to make a rough estimate of how you think the race will run and if you think you may have an edge before committing the time to analyze it further.
By coupling guesstimation, quick assessments, and sound reasoning you can think like Enrico Fermi when handicapping the races. You may not be a rocket scientist or know how many piano tuners there are in Chicago, but you will find the best races to handicap and churn out a nice profit.