The most likely outcome
Fund value over time
Value after withdrawals. The bands hold the middle half and the middle 90% of simulated paths. The dashed line is the same Fund held only in sector ETFs, so the gap between the lines is what student selections add or cost.
View as table
Annual withdrawal
View as table
Student holdings and the limit
Return against the S&P 500
Annualized time-weighted return of the Fund minus the S&P 500 Total Return Index, net of all costs. Each bar groups the simulations that landed in that range.
Sector teams
Each team is judged against the sector ETFs its picks replace.
Active return is dollar-weighted across the team's holdings and annualized. The range shows the middle 90% of simulations, with the darker stretch holding the middle half.
The IPS in operation
What the rules did across every simulated month, and where they bind.
One simulated history
How the simulation applies the IPS
Every rule below is enforced in every simulated month.
Base Camp Model IPS 5, 5.3
Each sector budget starts in its sector ETF. An approved buy is funded by selling that ETF, and sale proceeds return to it. Money moves between sectors only at rebalancing.
Sector budgets IPS 5.1
Each budget is the sector's S&P 500 weight times Fund assets. Weights drift with simulated sector returns.
Replacement limit IPS 5.4
20%, 40%, 60%, then 80% of each sector budget, tested at each purchase. A sector pushed over by markets or the withdrawal is not forced to sell, but its new buys wait.
Rebalancing IPS 5.5
Each January and August, by trading sector ETFs only. Student holdings are never touched.
Annual withdrawal IPS 10
4% of the average of the prior twelve month-end values, paid once a year at the chosen rebalancing. Raised from sector ETFs in proportion to budgets, never from student holdings.
Cash IPS 9
Target cash earns the cash yield. Each dividend is reinvested in the ETF of the sector that paid it at month-end, and builds up in cash during breaks. The 5% ceiling is tested every month.
Maintenance Mode IPS 12
No trades in May, June, July or December. Only breach remediation is allowed.
Initial deployment IPS 5.2
All assets, less target cash, go into the eleven sector ETFs in January 2027.
Proposals and voting IPS 4.3, 4.4
Teams submit buy reports in weeks 8 to 13 (March and April, October and November). The five other teams vote on each one: one vote against or none approves it, two send it to the Faculty Advisor, three or more reject it. Each approved buy goes to the team's sector with the most room.
Position review IPS 11.2
A holding 25% or more below cost is re-underwritten at the next trading window: exit (35% of reviews), trim by half (25%) or hold (40%). It is not a stop-loss.
Benchmark IPS 3
The IPS benchmark is the Russell 3000. The simulator compares the Fund with its S&P 500 sector core, which isolates what student picks add, and assumes the core and the Russell 3000 earn the same expected return. Returns are time-weighted and net of all costs.
Prohibited practices IPS 7
No leverage, derivatives, short sales or securities lending exist in the model, so none can occur.
Assumptions and limits
- Market returns come from one common factor with fat tails plus sector-specific moves, calibrated to J.P. Morgan's 2026 long-term assumptions for U.S. large caps: 6.7% compound return and 16.5% volatility, with cash at 3.1% and inflation at 2.4%.
- Each holding earns its sector's return, plus stock-specific noise, plus the skill setting. A skill of 0% means picks neither add nor subtract value before costs, which is what most evidence on active stock selection suggests. That is the most likely case.
- Costs are a 0.09% ETF expense ratio and 10 basis points per stock trade, with a 1.2% dividend yield. Sector weights within each team's pair are illustrative; the pair totals match Table 1 of the IPS.
- Not modeled: new gifts, changes to the withdrawal rule, sector reclassifications, shifting return regimes, and Board decisions beyond the switches shown.
- These are ranges of plausible outcomes, not forecasts. The same seed always reproduces the same results.