The Forecast

See what your students get, and why it lands.

No mockups. Every page below is real, pulled straight from the files you download. Tap any one to see it full size.

Your students are not solving for x. They are making a call the city is counting on.

This is a real daily release from the case. Your students work as a data analytics firm hired by Ridgemont, a town of 60,000 people. Each day a new dataset lands on their desk: population, tax revenue, school enrollment, housing starts, crime rate, water usage. They do not get a worksheet with a known answer at the back. They get raw numbers that curve, plateau, spike, and lie. From that evidence they choose a function family, fit a model, and forecast forward. The page asks for a defended decision, not a calculation. Pick the wrong model and the forecast they hand the city council falls apart in public. That pressure is the point.

Tap to see a case day

They are graded on how they reasoned, not on whether the forecast was lucky.

Here are the Dual Rubrics your students are measured against. One rubric scores the math: did they fit a sound function, check residuals, report a range instead of false precision. The other scores the reasoning traced in their daily journal: did they name the bias pulling at them, did they update when the evidence shifted, did they choose a domain-aware model over a flashy statistic. A team that picks a simple, honest model with a stated range of uncertainty can outscore a team that chased a perfect R-squared and got the number right by accident. A sound process behind a careful answer beats a lucky guess. The rubric makes that grading philosophy explicit, so the boring choice wins on paper too.

Tap to read the grading rubric

Let the workbook hold the busywork so the energy goes to judgment.

This is the student Workbook, a spreadsheet that does the arithmetic so your students can think. Each scenario arrives pre-loaded, so nobody spends the period typing in ten years of numbers. Students choose a function family, run the regression, and watch the R-squared and residuals update live. The tool surfaces the statistics, then steps back. It will not tell a team that a cubic fit with R-squared of 1.000 is a trap, and it will not warn them that their model predicts 930 million gallons of water for a town of 60,000. That call belongs to the student. By taking the computation off their plate, the workbook frees the whole session for the harder work: deciding which model the city should actually trust.

Tap to see the tracker

Every day is already scripted. You bring the facilitation, the simulation brings the rest.

This is the Teacher Guide, 13 playbooks that run minute by minute. Every session follows the same three-block rhythm: a 15-minute mini-lesson on the day's function family, 15 minutes of modeling and prediction in the workbook, then 15 minutes of journaling and pair share. The guide hands you the prompts, the discussion moves, and the reveal keys, including the Year 11 actuals you release on Day 9 to stress test every model in the room. You do not prepare a single dataset or invent a single prompt. Prep is light because the scripting is heavy. Your job is to facilitate the debate when a team's favorite model starts to crack, and the guide tells you exactly when that moment is coming.

Tap to read the teacher guide

The lesson plan is already written.

Every simulation comes with a fully editable, admin-ready lesson plan. Standards alignment, daily pacing, learning objectives, differentiation, and an assessment plan are already done, so you can hand it to an administrator or adapt it to your district template in minutes.

Tap to preview the lesson plan

The day a perfect model predicts an impossible number.

On Day 6 the water usage data lands, and it is seductive. The numbers climb for years, peak, then ease into a steady decline. Maya's team tries a cubic polynomial and the result is electric: R-squared of 1.000, a curve that threads every single point. They have never seen a fit this clean. They lock it in, sure they have beaten the case. On Day 9 the Year 11 actuals are revealed and they feel vindicated, because their model predicted 112.3 million gallons, almost exactly the real figure. The flashy choice looks like genius. Then the teacher asks one quiet question: what does your model say for Year 20.

Maya runs the forecast and the screen returns 930 million gallons, for a town of 60,000 people. The number is absurd. Her team faces the pivot every analyst eventually meets: defend the model that fit the past perfectly, or trust what the evidence is now screaming. They could rationalize the outlier away. Instead they pull the cubic, choose a simpler curve that respects how a real town actually uses water, and report a range instead of a single confident point. Once they let the perfect fit go, they see it plainly. The model was memorizing history, not predicting the future.

A data analytics firm hired by the city of Ridgemont.

Your students are not students for these two weeks. They are consultants at a data analytics firm, hired by the city of Ridgemont, population 60,000, to forecast its future. Infrastructure, housing, and public safety planning all depend on the models they build. Each day the city releases a new dataset, and the firm must turn raw numbers into a defended forecast. They work in teams, debate the evidence, and ultimately stand before a city council to justify the calls they made. The math is real. The pressure is the role.

Engineering better thinkers.

In a world run on data, the dangerous skill is not running a regression, it is knowing which model to trust. The Forecast pairs each day's math with the bias most likely to corrupt it, then names the capacity that defeats that bias under real pressure.

13 days of evidence, models, and one hard reveal.

Three pre-simulation days build the toolkit: spotting signal versus noise, the function families and the R-squared warning, and a bias experiment on anchoring. The ten case days then march through Ridgemont's data, each introducing a new function type and a new trap, building toward the Day 9 reveal and the Day 10 defense.

Standards alignment.

The Forecast is built on the Algebra II modeling strand of the Common Core math standards. Students distinguish linear, exponential, logarithmic, and polynomial growth and choose among them, which is CCSS.MATH.CONTENT.HSF.LE.A.1. They fit functions to data and use those functions to solve problems, which is CCSS.MATH.CONTENT.HSS.ID.B.6, elevated here because students must judge which features of a dataset matter, such as excluding the Day 5 crime outlier to find the honest trend. The work also draws on interpreting regression and residuals under HSS.ID.B.6.B and reasoning about rates of change in HSF.IF.B.6. These standards are not merely met. They are exercised as professional judgment.

Turnkey, classroom-ready.