The McKinsey Redrock game is a 35-minute data-interpretation task inside the McKinsey Solve assessment. You work through a RedRock Island study, collect useful data, calculate answers, complete a short report, and then solve six quick cases. It is paired with Sea Wolf in the standard 65-minute Solve invitation. As of May 2026, candidates on the longer 85-minute invitation also see Sustainable Future Lab as a third game.
This guide explains how the four phases fit together, what the math looks like, how to think about scoring, and how a full Redrock question feels in practice. The walkthrough follows a Southshore wolves study from the opening text to the final chart and case answer.
Key takeaways
- The Redrock game is a 35-minute data-interpretation task inside the McKinsey Solve assessment. It has four phases (Investigation, Analysis, Report, Cases) grouped into two parts (Study and Cases).
- The Study part is one connected task with one Investigation, one Analysis question set, and one Report. The Cases part is six independent mini-problems.
- The math mainly uses percentages, ratios, weighted averages, and probability. Speed and accuracy matter more than mathematical sophistication.
- McKinsey does not publish a Redrock rubric. For practice, treat every answer field as important, including report blanks, chart choices, chart values, and case answers.
- The most useful practice is a full timed run through all four phases, with isolated math drills as support.
What is the McKinsey Redrock game?
The Redrock game, sometimes called the Redrock Study or RedRock Island, is the data-interpretation part of the McKinsey Solve assessment. McKinsey describes Solve on its official digital assessment page as a game-based exercise used during the screening stage of the application. Redrock asks you to do the work of a junior analyst: read evidence, choose useful data, calculate answers, and turn the results into a short report.
The setting is RedRock Island. The exact species and exhibits change between runs: one candidate might see wolf packs and elk populations, another might see plant biomass, reservoir capacity, or migration patterns. The work stays consistent. You collect data, run percentage and ratio calculations, fill in a written report, choose a chart format, populate the chart, and then face six independent cases under shorter time pressure.
Redrock replaced the older Plant Defense game and is now the standard data-heavy partner to Sea Wolf inside Solve. Compared with classic case interviews, Redrock gives you the structure upfront. Your job is to read carefully, calculate accurately, and stay disciplined under the timer.
Who gets the Redrock game, and when?
Every candidate invited to the McKinsey Solve assessment receives the Redrock game. McKinsey's Solve FAQ updated August 2025 says the length of your Solve assessment is outlined in your invitation email, and each task starts with an untimed tutorial. In current Solve invitations, the total length usually tells you which game mix to expect:
- 65-minute Solve. You receive Redrock and Sea Wolf.
- 85-minute Solve. You receive Redrock, Sea Wolf, and Sustainable Future Lab.
The Redrock timer itself is the same on both versions: 35 minutes for the Redrock game alone. The other games run on their own clocks, so you cannot save time on Redrock and spend it on Sea Wolf. If you are unsure which length you will see, use the invitation email from McKinsey as the source of truth.
If your invitation is for the 65-minute version, this article and our Sea Wolf game guide cover the two games you will actually take. If you have the 85-minute invitation, add the Sustainable Future Lab as the third. For the broader picture of the test, see our McKinsey PSG mini-games overview and our how to prepare for McKinsey Solve guide.
Each game has its own timer. You cannot carry minutes between Redrock, Sea Wolf, or Sustainable Future Lab. Finishing one game with extra time is good, but it will not be added to the next.
How the McKinsey Redrock game is structured
Redrock has four phases that always run in the same order, grouped into two parts in the interface. The timing below is not a rule from McKinsey. It is a practical pacing range, so use it as a starting point and adjust during timed practice.
| Part | Phase | What you do | Time |
|---|---|---|---|
| Study | Investigation | Read the study objective, study information, and exhibits; drag relevant data into the Research Journal | 6–8 min |
| Study | Analysis | Answer 3–4 numeric questions using the in-game calculator | 10–12 min |
| Study | Report | Fill in a short written report, choose the right chart, populate it with values | 6–7 min |
| Cases | Cases | Work through 6 independent mini-cases with their own exhibits | 10–13 min |
The Study part is one connected task. The Cases part is a sprint of six small problems with no shared storyline. The transition between them is one-way: once you confirm the Report, you cannot go back to Investigation or Analysis. Inside Investigation and Analysis, you can move freely until you commit, which is why strong runs usually spend a little extra time gathering and computing before locking the Report.
A few interface details matter in practice:
- The Research Journal is where you store useful values. You drag values from the study text and exhibits into it during Investigation. The journal blocks duplicates, lets you rename or reorder entries, and stays available during Analysis and Report. When Cases begin, the journal resets.
- The Calculator supports operator precedence and parentheses. Past results can be dragged from history into the journal or directly into answer fields, which keeps your numbers consistent across questions.
- The Report has three steps: a written report with about 8–10 blanks, a chart-type choice (bar, line, or pie), and a chart you fill in with category values.
How the 35 minutes are budgeted
McKinsey gives you one 35-minute Redrock timer, not a timer for each phase. The pacing table above is the target. The notes below explain why those ranges work.
| Phase | Recommended budget | Why |
|---|---|---|
| Investigation | 6–8 min | Read carefully once; drag the obvious useful values, then use Analysis to confirm what else you need. |
| Analysis | 10–12 min | Leave room to calculate, check units, and go back to Investigation for any missing value before you commit. |
| Report | 6–7 min | Most blanks come straight from Analysis answers; the chart step is fast if your numbers are clean. |
| Cases | 10–13 min | Six cases, roughly 2 minutes each. If certainty is not coming quickly, make an educated guess and move to the next one. |
Most timing problems start early. If Investigation or Analysis takes too long, the Cases section becomes a rushed guessing exercise. The fix is to practice full 35-minute runs until you know what a lean Investigation and a 10-minute Cases section feel like. For more pacing examples, see our Redrock time-management article.
What math is on the Redrock game?
Many Redrock answer fields are numeric. The calculations are familiar: percentages, ratios, growth rates, probability, and chart reading, done quickly and cleanly under time pressure.
| Math concept | Example of how it shows up | Common trap |
|---|---|---|
| Percentage of total | Turning category values into chart shares | Forgetting to use the correct denominator (e.g. total land vs. occupied land) |
| Year-over-year growth rate | Comparing changes across several years | Confusing total growth (Y4 vs Y1) with average annual growth (mean of yearly steps) |
| Weighted average | Combining groups with different sizes | Weighting by the wrong column (e.g. averaging by pack count instead of by hunts) |
| Ratio and proportion | Translating phrases like "a quarter of" | Misreading "a quarter of" as 25% of total instead of 25% of a sub-group |
| Probability | Reading frequency tables or case exhibits | Counting cells with NA values incorrectly in frequency tables |
| Reading bar / pie / line charts | Choosing or interpreting a visual summary | Skipping the footnote that defines the unit (e.g. "1 unit = 1 kg") |
If you want a deeper math breakdown with worked solutions, our Redrock math guide goes through each concept with realistic exam-style problems.
How should you think about Redrock scoring?
McKinsey does not publish a Redrock scoring rubric. For preparation, the most useful assumption is simple: every field you submit can matter, and conceptual mistakes usually cost more than small rounding differences.
- Treat every submitted field as important. Analysis inputs, report blanks, chart type, chart values, and case answers all deserve attention.
- Round carefully, but do not over-chase decimals. For example, our simulations allow small numeric tolerance, roughly 2% in most fields, so candidates can work realistically with the in-game calculator.
- Be precise on categories and evidence. Wrong dropdowns, wrong chart types, missing evidence, or extra evidence are concept errors rather than rounding issues.
The practical takeaway is simple: protect the choices that show whether you understood the problem. Chart type, dropdowns, evidence selection, and units deserve as much care as the arithmetic.
McKinsey reviews Solve results alongside the rest of your application. The Solve FAQ states the result is one input into the screening decision, not a standalone pass mark. A weaker Redrock run is reviewed in the context of the full application.
A walkthrough from a real Redrock simulation
The fastest way to understand the Redrock game is to see one. The example below uses a Southshore wolves study and shows the full workflow from Investigation to Cases.
Investigation: the study you would actually read
The study screen opens with the objective and the study information. The bracketed numbers are values you drag from the body text into your Research Journal.
Researchers want to know how Southshore's land is divided between Territories, a Buffer Zone, and Unclaimed Land. They want to know if the buffer zone area's growth rate was comparable to the 3rd largest packs' territory growth rate over the past 4 years.
Southshore occupies a land area of [2,550] hectares (ha) of Redrock Island and is home to several wolf packs, as well as lone wolves. Note that lone wolves are not associated with a wolf pack.
Over time, researchers have noted that three of the largest wolf packs, Boreal, Sable, and Meteor, have continually expanded their territory and currently occupy about [60]% of Southshore territory. Several smaller wolf packs and lone wolves have also claimed land. Their combined territory is about [a quarter] of the total amount of land area claimed by the three largest packs. The remaining land in Southshore is unoccupied by wolves.
In year 1, researchers designated [189] hectares for the buffer zone. In year 4, the buffer zone was expanded to [206] hectares. Land that is not occupied by wolves or dedicated to the buffer zone is called "unclaimed land."
Then comes Exhibit #1, the only exhibit in this simulation:
Exhibit #1: Pack Territory Size By Year (hectares)
| Wolf Pack | Year 1 | Year 2 | Year 3 | Year 4 |
|---|---|---|---|---|
| Boreal | 458 | 476 | 492 | 515 |
| Sable | 432 | 453 | 465 | 473 |
| Meteor | 475 | 492 | 526 | 542 |
In Investigation, your job is to choose the values you will actually use. For this study, the key inputs are the five bracketed numbers above plus the Year 4 column from the exhibit. Collecting the full exhibit is fine. Dragging every Year 1 value when the questions only ask about growth rates usually slows you down.
Analysis: one full question with the math worked through
The first Analysis question in this simulation asks for the Year 4 area of each land category. The question expects four numeric inputs.
Calculate the area at year 4 for the following:
- Territory, 3 largest packs (ha):
- Territory, small packs and lone wolves (ha):
- Buffer zone (ha):
- Unclaimed land (ha):
Working through it:
- 3 largest packs (Y4): Boreal + Sable + Meteor = 515 + 473 + 542 = 1,530 ha.
- Small packs / lone wolves: the study said their combined territory is a quarter of the three largest packs', so 1,530 ÷ 4 = 382.5 ha.
- Buffer zone (Y4): taken directly from the study text, 206 ha.
- Unclaimed land: total land minus everything else, so 2,550 − 1,530 − 382.5 − 206 = 431.5 ha.
That is one Analysis question. A full simulation has three to four, and most have two parts. The calculations become much easier once the useful values are in the journal. Reading speed and unit consistency create more risk than the formulas themselves.
Carry calculator values forward. The journal lets you drag previous results into a new answer field. Re-typing 1,530 from memory two questions later is how candidates introduce avoidable errors.
Report: choosing the chart and filling it in
After Analysis, the Report phase asks you to summarize the findings, pick a chart type, and populate the chart. In this simulation, the chart-selection prompt is:
Chart the land breakdown in Southshore in year 4.
The four categories are the 3 largest packs, small packs and lone wolves, the buffer zone, and unclaimed land. Together they sum to 2,550 ha, which is the total area. A pie chart fits because the four categories form the whole.
Once the chart type is set, you populate it. The percentages come straight from the Year 4 numbers above:
| Category | Hectares | % of total |
|---|---|---|
| 3 largest packs | 1,530 | 60.00% |
| Small packs / lone wolves | 382.5 | 15.00% |
| Buffer zone | 206 | 8.08% |
| Unclaimed land | 431.5 | 16.92% |
Chart choice matters here. Choosing a line chart for a parts-of-a-whole breakdown is a concept error, so use the prompt language before entering values. "Breakdown" and "share" suggest a pie chart, "comparison across categories" suggests bars, and "trend over time" suggests a line chart.
Cases: one example of the post-Report sprint
After the Report, six Cases appear. They are independent of the main study. The Research Journal resets for each case, and you cannot revisit the previous one. Here is the first case from the same simulation:
Wolf Population Count by Age and Region (Exhibit #1) shows juvenile, reproductive, and senior wolves across four regions on Redrock Island.
Select the best equation to calculate future reproductive wolf population growth, given that the future reproductive population is the current juvenile plus reproductive count, and the population doubles every 8 years.
The four equation options use either 500 or 355 as the starting count, and either an (n/8)² or a 2^(n/8) growth model. To choose:
- The starting reproductive population is the current juvenile + reproductive total, summed across the four regions: (25 + 20 + 13 + 22) + (85 + 63 + 50 + 77) = 355.
- "Doubles every 8 years" describes exponential growth, which is the
2^(n/8)form.
So the answer is p = 355 × 2^(n/8).
That is one of six cases. The pattern is consistent: a self-contained exhibit, one or two questions, and the same kinds of math (percentages, weighted averages, probability, equation selection). Cases become hard because six small problems demand fresh attention back to back, with no shared scenario to lean on.
How Redrock differs from Sea Wolf and Sustainable Future Lab
If you have the 85-minute Solve, all three games show up on the same sitting. They test different reasoning styles and run on separate clocks, so each one rewards a slightly different kind of practice.
| Game | Timer | Core reasoning style | What practice should target |
|---|---|---|---|
| Redrock Study | 35 min | Quantitative reasoning under structure, plus chart and report communication | Reading exhibits fast, percentage and ratio fluency, calculator habits, chart selection |
| Sea Wolf | 30 min | Constraint-based optimization and decision making | Filtering options, scoring trade-offs, picking sites under limits |
| Sustainable Future Lab | 20 min | Narrative judgment under uncertainty, with team and stakeholder dynamics | Reading scenarios, weighing trade-offs, consistency across decisions |
Each game rewards a different prep habit. Extra time in one game stays there, so plan separate timed reps for each game on your invitation. The all-in-one McKinsey Solve bundle covers the full mix in one place.
Common Redrock mistakes
These mistakes usually hurt more than the arithmetic itself.
| Mistake | Why it happens | Fix |
|---|---|---|
| Over-collecting in Investigation | Candidates drag every value into the journal, "to be safe" | Read the questions first if the interface lets you, and collect only what they need |
| Re-typing values from memory | Calculator history feels slower in the moment | Drag past results from the calculator into the next answer field |
| Picking the wrong chart category | Treating chart selection as a guess | Match the chart to the prompt language: breakdown → pie, comparison → bar, trend → line |
| Burning time on a single Case | Wanting to be sure on every answer | Hard-cap each case at 2 minutes; make an educated guess when needed and move on |
| Ignoring exhibit footnotes | Skipping small text under tables | Read the unit definition before computing (1 unit = 1 kg, currency, percentage vs proportion) |
| Confusing total growth with average annual growth | Reading "growth rate" as the same in both contexts | Use total growth for "Year 1 to Year 4 growth"; use the mean of yearly steps for "average annual growth" |
Switching strategy mid-test. Candidates who struggle in Investigation often try to "make up time" in Cases by guessing aggressively. The Cases section is where careful reading still pays off the most. Recover by pacing instead of accelerating into errors.
How to prepare for the Redrock game
Most candidates start preparing two to four weeks before the test. The plan below assumes that window. If you have less time, keep Phase 1 short and spend most of your time on full timed practice.
Phase 1: Learn the format
Before any timed practice, get the structure into long-term memory:
- Four phases: Investigation, Analysis, Report, Cases.
- Two parts: Study (the first three phases) and Cases (six independent problems).
- One 35-minute timer covering the whole game.
- Calculator with history, Research Journal that resets between cases, one-way Report transition.
Run one untimed simulation to see the interface, so layout questions do not drain attention when the timer is on.
Phase 2: Practice with realistic simulations
This is where most of the improvement happens. Real Redrock-style runs train the actual skill mix: reading exhibits fast, computing percentages cleanly, picking the right chart, and pacing the Cases section.
Start with the free Redrock simulation practice for an easy first rep. For deeper practice, the PSG Cracked Redrock simulation pack has 30 full simulations across different RedRock Island themes (wolves, plants, water capacity, migration and more) so you can build pattern recognition across exhibits.
Per session:
- Run the full four-phase flow under the 35-minute clock.
- Note the two phases where you lost the most time.
- Replay only those phases with the same simulation before moving to the next one.
Phase 3: Rehearse to time
In the last few days, switch from learning to consolidation:
- Do one full timed run every day or two.
- Keep solution review short; the final stretch is for timing and consistency.
- If your invitation includes Sea Wolf and SFL, give each its own timed rep on the same day to simulate the real session.
For more on the time pressure specifically, see our Redrock time-management article.
Frequently asked questions
What is the McKinsey Redrock game? A 35-minute data-interpretation task inside the McKinsey Solve assessment. The phases are Investigation, Analysis, Report, and Cases. The setting is RedRock Island, with species and exhibits that vary between simulations.
How long is the Redrock game? 35 minutes for Redrock alone. The full Solve assessment runs 65 or 85 minutes depending on whether the Sustainable Future Lab is included.
How is the Redrock game scored? McKinsey does not publish a scoring rubric. For practice, treat every submitted field as important: numeric inputs, report blanks, chart choices, chart values, and case answers. This is preparation guidance rather than a leaked formula.
Are there leaked Redrock answers? No. Each simulation has different exhibits and different numbers, so a memorized answer for one run does not transfer. Practice runs train the reasoning pattern, which is why simulator volume helps and answer keys do not.
What math do I need for Redrock? Percentages, ratios, growth rates, weighted averages, probability, and reading common chart types. The calculations are familiar, but Redrock tests how quickly and accurately you can apply them under time pressure.
Can I use a real calculator or Excel during the game? You should use the in-game calculator. Bringing calculations into a separate calculator or spreadsheet is a bad tradeoff: it slows you down and pulls you out of the workflow the assessment is designed around. The built-in calculator supports parentheses and result history, which is enough for the Redrock math if you practice with it.
Is Redrock the same as Red Rock or RedRock Island? Yes. The game is sometimes written as Redrock, Red Rock, or RedRock Island depending on the source. McKinsey's assessment platform uses RedRock Island as the in-game scenario name.
Do I need a business background to do well on Redrock? No. The scenario gives you the context you need. Non-business candidates can prepare effectively by practicing fast reading, clean arithmetic, and structured reasoning under time pressure.
Before you practice
- Know the phase order before you start timed runs: Investigation, Analysis, Report, then Cases.
- Keep your calculations clean enough to support the Analysis answers, report blanks, and chart values.
- Practice Cases under a hard time cap, because six short problems can drain the clock quickly.
What to do next
Start with the free Redrock simulation practice if you want one low-pressure practice run. Move to the PSG Cracked Redrock simulation pack when you need full 35-minute reps with feedback.
If your invitation is for the 85-minute Solve, use the all-in-one McKinsey Solve bundle so Redrock, Sea Wolf, and Sustainable Future Lab practice sit in one place. The full set of options is in the McKinsey Solve collection.
