Wharton Investment Competition
I co-captained DMV’s Finest in the Wharton Global High School Investment Competition. Each team of high school students gets $100,000 in simulated money to invest in Wharton’s trading simulator, and a client to build a strategy for. In 2023, 3,200 teams from 53 countries entered, and on April 22 we won the global final in Philadelphia.
Start with the client
That year’s client, Peter Hjemdahl, wanted to give a $5,000 to $10,000 gift to a startup every year and open a yoga studio in Miami within 15 years. We modeled both first. The model said 8% a year would cover the gifts and the studio, so we aimed for 20% to keep the gifts going after the studio was bought. We ran it as a Monte Carlo simulation, so we could see the range of outcomes, not just one line.
511 stocks to 22
Every team got the same approved list of 511 stocks across 11 sectors. We first narrowed it to the seven sectors that fit our client: health care, consumer staples and utilities to hold up in a downturn, and communications, technology, financials and energy to grow. Then we screened on macro factors, keeping companies in countries we graded well, and kept stocks whose alpha beat their industry’s average while their beta stayed below it. Research on the 79 that were left picked the final 22.
Grade the countries
For the macro screen, we graded each of the 22 countries our stocks operated in, from A to F. We weighed political stability, sovereign risk and government policy alongside inflation, inequality and credit ratings. At the semifinals, a judge disagreed with our low grade for China, and we answered in writing.
Score every company
Each shortlisted stock got a DMV Dozen score from 0 to 100, built from twelve numbers in its financial statements and ranked against its industry. We also measured the tone of its recent headlines and graded its management and impact by hand. At the finals, Sprouts Farmers Market, the stock we walked the judges through, scored 75.2, and its headlines were more than twice as positive as the consumer-staples average. Looking back, the score alone barely predicted what came next. The stocks we picked with it and our research did: a median of +109% from the report to October 2026, against +62% for the scored stocks we passed on.
Size the positions
We sized each position with a mean-variance optimizer, the core idea of modern portfolio theory. Five years of daily prices gave each stock an expected return and a covariance with every other stock, and the optimizer found the mix with the least risk for each level of return, within limits on each sector. We ran it for the lowest risk and for the best return per unit of risk, and we recommended the lower-risk mix.
What happened next
If our client had bought the portfolio in our report on December 13, 2022, and held it, $100,000 would have grown to about $227,600 by October 22, 2026: 23.8% a year against our 20% target, and 127.6% in all against 100.2% for the S&P 500. It swung a little more than the market, but it earned more for each unit of risk: a Sharpe ratio of 1.10 against 0.99, and about 4% a year more than its market exposure explains. The riskier mix we passed on, with about 10% of its stocks in NVIDIA, would have done even better: about $290,900, with a Sharpe ratio of 1.22.
3,200 teams, 10 finalists
Of the 3,200 teams, 1,400 turned in a final report, and ten made the final in Philadelphia, where each team pitched its strategy to the judges. We took first place.
