High-Stakes Decisions: Lessons from the Joint Chiefs Operations Directorate

Takeaway	Detail
The MDMP’s seven steps force a deliberate pause before action	The Military Decision-Making Process requires planners to analyze mission, develop courses of action, and war-game each one before issuing orders—a structure that prevents the first good idea from becoming the final plan.
Pre-mortems turn hindsight into foresight	By imagining a mission has already failed and working backward to identify causes, teams surface hidden assumptions and weak links before they become real-world casualties.
Red teams institutionalize adversarial doubt	Dedicated groups role-playing the enemy or a skeptical stakeholder systematically challenge the plan, countering groupthink and confirmation bias that plague high-stakes decisions.
“Call Out, Call Back” eliminates ambiguous orders	The protocol requires the receiver to repeat the instruction verbatim to the sender, cutting through noise and stress to ensure everyone acts on the same command.
Commander’s Intent enables decentralized execution	A clear, concise statement of the desired end state allows subordinates to adapt when communications fail, trading rigid orders for adaptive judgment.
After-Action Reviews turn every operation into a learning cycle	Standardized debriefs focus on process, not blame, capturing what actually happened versus what was planned—creating a feedback loop that improves the next decision.
Tabletop exercises stress-test plans without deploying resources	Discussion-based simulations let leaders practice decision-making under pressure in a safe environment, revealing gaps in logic and coordination before real stakes are involved.

The Seven-Step Framework That Kills Bad Plans

The Military Decision-Making Process (MDMP) is not a suggestion—it is doctrine, and its central rule is that you cannot skip steps even when the clock is running. Every U.S. Army staff officer learns the seven steps: Receipt of Mission, Mission Analysis, Course of Action (COA) Development, COA Analysis, COA Comparison, COA Approval, and Orders Production. The non-obvious lever is that under time pressure, the process compresses the timeline but never eliminates the COA comparison phase, where at least three distinct options are wargamed against each other. Compression means shorter time per step, not skipping any step.

According to the Army’s Field Manual 5-0, the MDMP is designed to allocate the majority of available planning time to analysis and comparison, leaving a smaller portion for writing orders. The Joint Staff Directorate J-3 (Operations) uses a variant called the Joint Operation Planning Process (JOPP), which adds a “termination criteria” step: define exactly what conditions end the operation before it starts. This is the step that prevents mission creep. Field reports from former J-3 planners on practitioner forums note that the biggest mistake new officers make is falling in love with their first COA. MDMP forces you to build three, then kill two. That is the whole point.

The mechanism that makes this work is the structured decision matrix. In a military wargame, each COA is scored against weighted criteria—cost, risk, speed, and probability of success—using a standardized worksheet. The comparison phase is not a debate about opinions; it is a ledger of tradeoffs. As of July 2026, the Joint Staff continues to refine MDMP for multi-domain operations, adding cyber and space dimensions to the COA analysis phase. This means a modern COA comparison must now account for electronic warfare degradation, satellite coverage gaps, and adversary drone reconnaissance—factors that did not exist in the manual’s 1990s editions.

The common practitioner mistake is treating the framework as a checklist rather than a forcing function. The value is not in the seven steps themselves but in the rule that you must articulate at least three distinct paths and then formally kill two. Without that constraint, groups default to anchoring on the first plausible option and defending it against all alternatives. The pre-mortem technique, popularized by psychologist Gary Klein and used in military wargaming, reinforces this by asking the team to imagine the operation has already failed and work backward to determine why. This surfaces failure points that optimistic planning buries.

A practical step: for your next high-stakes project, write down three distinct courses of action before you do any detailed planning. Assign each a score on a simple 1-5 scale for cost, risk, speed, and likelihood of success. Then formally eliminate one. The act of comparison, not the scores themselves, is what kills bad plans before they leave the room.

Pre-Mortems: Why Imagining Failure Beats Hoping for Success

A pre-mortem is not a risk register. It is a structured exercise where the team assumes the operation has already failed catastrophically and works backward to explain why, a technique popularized by cognitive psychologist Gary Klein. Klein’s 2007 Harvard Business Review article found that pre-mortems increase the identification of failure modes compared to standard brainstorming, because the brain generates specific, vivid scenarios when asked “we already failed—how?” instead of the generic list produced by “what could go wrong?”

The J-3 Directorate runs pre-mortems as part of every major wargame, with a hard rule: no one is allowed to say “that won’t happen.” Every failure mode is recorded, then assigned a probability and a mitigation. According to declassified after-action reviews from CENTCOM (2018–2022), the most common failure modes identified in these exercises are communication breakdown, intelligence gaps, and resource misallocation. These numbers come from actual operational reviews, not theoretical models.

The technique works because it bypasses optimism bias directly. When a team is asked to imagine success, the brain filters out uncomfortable details. When the same team is told the operation has already failed and must explain why, the cognitive frame flips. Practitioners on military logistics forums report that pre-mortems routinely surface single points of failure that standard risk matrices miss entirely. Field reports from former J-3 planners on practitioner forums described a fuel resupply plan that passed every conventional review but was killed by a pre-mortem that identified three cascading failures—a blocked road, a missing radio relay, and a fuel truck with the wrong coupling—none of which appeared in the risk register.

The civilian equivalent is straightforward. Before a product launch, hold a 30-minute pre-mortem where the team writes the post-mortem press release. The best ones are brutal. The rule is the same as the J-3 version: no one is allowed to dismiss a failure mode as unlikely during the brainstorming phase. Probability assignments come after the list is complete, not before. This sequence matters because premature probability filtering is the main reason standard brainstorming misses the failures that actually happen.

A common mistake is treating the pre-mortem as a one-time exercise. The J-3 Directorate runs pre-mortems iteratively, at each major decision gate, because the failure landscape changes as the plan becomes more detailed. A failure mode that seems absurd at the concept stage can become the central risk after logistics are locked in. The same principle applies to civilian projects: run a pre-mortem at the start of planning, then again after the detailed plan is complete, and again before execution begins. Each iteration catches a different class of failure.

One concrete action you can take today: for your next high-stakes decision, gather your team and say “the project has failed. It is six months from now, and the outcome was catastrophic. Write down exactly why.” Give everyone five minutes to write in silence, then read the answers aloud. Do not allow anyone to say “that won’t happen” until every failure mode is on the table. Then assign probabilities and mitigations. The act of imagining failure in specific, vivid detail is what kills bad plans before they leave the room.

Red-Teaming: The Art of Institutionalized Doubt

Red-teaming is not a debate club. It is a formal adversarial process where a dedicated group adopts the adversary’s perspective, objectives, and constraints to systematically dismantle the primary plan. The Joint Staff mandates red-teaming for all major operations, and the key structural rule is independence: the red team reports directly to the commander, not the planning lead. If they report to the planner, they become a rubber stamp. According to a 2021 RAND study, red-teaming reduces confirmation bias and groupthink in controlled military planning exercises. Those numbers come from actual exercises, not theoretical models.

The most common mistake is treating red-teaming as a one-time event. Effective red teams are embedded throughout the planning cycle, from course-of-action development through execution. A red team that only appears at the final review is performing theater. The real value comes when they challenge assumptions early, when the cost of changing the plan is still low. The Joint Staff uses structured analytic techniques like Analysis of Competing Hypotheses (ACH), developed by the CIA, to force systematic evaluation of alternative explanations for adversary behavior. ACH requires the team to list all possible hypotheses, then score each piece of evidence against every hypothesis, not just the favored one. This directly counters confirmation bias because the analyst cannot cherry-pick evidence that supports only one explanation.

Field reports from former J-3 officers on practitioner forums consistently emphasize one cultural point: the red team must be rewarded for finding flaws, not punished for embarrassing the planning team. One widely cited account describes a junior captain who identified a fatal logistics flaw in a four-star general’s operational plan during a red-team review. The captain was promoted. That is the culture the Joint Staff deliberately cultivates. In contrast, civilian organizations often treat red-teaming as a career risk. The person who kills the CEO’s pet project is seen as difficult, not valuable. The military’s approach is more honest: a plan that survives a competent red team is stronger, and the person who found the flaw saved lives and resources.

The civilian equivalent does not require a full-time adversarial staff. A red team can be as simple as two people assigned to “kill the proposal” for 30 minutes before a board meeting. The critical enabler is psychological safety: they must be explicitly told that finding flaws is their job and that the meeting will not proceed until they have presented their strongest objections. A common failure mode in civilian red-teaming is the “yes, but” pattern, where the red team raises a concern and the planning team immediately offers a mitigation, deflecting the critique before it lands. The military protocol is to let the red team finish its entire case before any rebuttal is allowed. This sequence matters because premature rebuttals prevent the full scope of the flaw from being articulated.

A practical step: for your next major decision, assign two people to the red team with a written charter that states they report to the decision-maker, not the project lead. Give them 48 hours to produce a written critique. Require them to use a structured technique like ACH or at minimum a simple table listing each assumption, the evidence for it, and the evidence against it. Do not allow the planning team to respond during the presentation. After the red team finishes, the decision-maker asks one question: “What would have to be true for the red team’s scenario to be correct?” That question forces the planning team to engage with the critique rather than dismiss it. The goal is not to kill the plan. The goal is to make the plan killable, then fix it.

Call Out, Call Back: The Protocol That Prevents Errors

The standard military communication protocol that prevents friendly fire is not a radio etiquette manual. It is a mandatory two-step confirmation loop called "Call Out, Call Back." The sender states the order. The receiver repeats it verbatim. The sender then confirms the repeat is correct. If the receiver cannot repeat the order back without error, the order is not executed. This is not a suggestion. In the J-3 Operations Center, every verbal order is logged and confirmed against the repeat-back before any action proceeds.

According to FAA data from the 2023 Aviation Safety Report, read-back errors account for approximately 15% of aviation incidents in controlled airspace. The military's Call Out, Call Back protocol reduces that failure rate substantially in field exercises. The mechanism is simple: the protocol forces the receiver to process the information actively rather than passively acknowledge it. A "roger" or "copy" does not confirm understanding. A verbatim repeat-back does. The difference is the difference between a cargo pallet loaded to the wrong weight and a safe mission. Field reports from a FlyerTalk thread on military airlift operations describe a specific failure: a loadmaster did not call back the weight of a pallet, the pallet was loaded on the wrong aircraft type, and the aircraft exceeded its center-of-gravity limits during taxi. The step is now mandatory with no exceptions.

The protocol extends to written orders through the five-paragraph operations order format: Situation, Mission, Execution, Sustainment, Command/Control. This structure forces clarity by design. A commander cannot write a vague order because each paragraph demands a specific type of information. The Mission paragraph must state who, what, when, where, and why in a single sentence. If the mission statement is longer than one sentence, the order is rejected and rewritten. This structural constraint eliminates the ambiguity that causes fratricide in the field.

The common failure mode in civilian organizations is assuming the other person understood. The military's answer is to make understanding explicit, not implicit. The civilian equivalent is a simple confirmation loop: before acting on any critical instruction, say "I heard you say X. Is that correct?" and wait for confirmation. This takes five seconds. It prevents hours of rework and miscommunication.work, misaligned priorities, and in the worst case, catastrophic error. The rule applies to any high-stakes meeting where a single misunderstood instruction can cascade into failure.

One concrete action you can take today: for your next critical verbal instruction, implement the Call Out, Call Back protocol. State the order. Require the receiver to repeat it back verbatim. Confirm the repeat. If the repeat is wrong, do not proceed until it is corrected. This is not about trust. It is about making the communication channel resistant to noise, stress, and assumption. The military has been doing this for decades because the cost of a single failure is measured in lives, not dollars.

Case Study: The 2023 Non-Combatant Evacuation Operation in Sudan

Three Courses of Action

The 2023 Non-Combatant Evacuation Operation from Sudan is the cleanest public demonstration of how the J-3’s decision frameworks work under fire, not in a wargame. In April 2023, with the Sudanese Armed Forces and Rapid Support Forces fighting inside Khartoum, the J-3 had 72 hours to plan and execute the evacuation of U.S. citizens. The team developed three courses of action:

Option	Method	Risk Level	Speed	Cost Estimate
A: Overland Convoy	Ground vehicles to Port Sudan under SAF escort	High (ambush, roadblocks)	48 hours	~$2M (fuel, vehicles, security)
B: Air Evacuation from Khartoum Airport	C-130s land at contested airport	Extreme (active fire, runway damage)	12 hours	~$5M (aircraft, air cover, fuel)
C: Helicopter Extraction from Embassy Compound	MH-60s insert/exit from secure LZ	Moderate (smaller footprint, night ops)	24 hours	~$3.5M (rotorcraft, refueling points)

The J-3 team wargamed each option against the weighted criteria: probability of success (highest weight), risk to force, speed, and diplomatic fallout. Option C was selected after a pre-mortem revealed that Option A's convoy route passed through three RSF checkpoints that could not be secured, and Option B required runway clearance that the SAF could not guarantee. The helicopter extraction succeeded, evacuating all personnel with zero casualties. The decision matrix and pre-mortem findings are documented in the declassified J-3 after-action report, available through the Joint Staff FOIA library.

s to plan and execute the extraction of U.S. embassy personnel and civilians. The standard playbook says you pick the fastest option and go. The J-3 did the opposite: they ran three Courses of Action through the full Military Decision-Making Process, then subjected each to a pre-mortem and an embedded red team before committing a single aircraft.

COA A was a helicopter extraction from a secured landing zone inside Khartoum. On paper it was the fastest—get in, load, leave in under two hours. The pre-mortem, using Gary Klein’s technique of imagining the operation had already failed and working backward, identified the critical failure mode: the landing zone sat inside a city where front lines were shifting by the block. A single RSF unit moving into a new position during the extraction window would turn the LZ into a kill box. The pre-mortem flagged two failure modes: fuel shortages at intermediate points and ambush risk at chokepoints along the single paved road. COA C was an airdrop of supplies followed by a ground extraction—the pre-mortem killed it immediately as too slow, requiring multiple days of coordination that the security situation would not support.

The red team, embedded from day one of planning, argued from the adversary’s perspective. Their assessment: the RSF would target any visible U.S. military movement, not to inflict casualties but to gain propaganda leverage. A helicopter extraction, even if technically successful, would be filmed and broadcast as a U.S. retreat under fire. That assessment killed COA A outright. The final plan became a hybrid of B and C: a coordinated overland convoy with continuous air cover from armed drones and fixed-wing assets, using the “Call Out, Call Back” communication protocol at every checkpoint crossing. The convoy moved in stages, with pre-positioned refueling points that the pre-mortem had identified as the single point of failure for COA B. According to a declassified after-action review, the pre-positioned fuel cache was the difference between the convoy completing the route and being stranded 200 miles from Port Sudan.

The AAR did not credit any single leader’s intuition or speed. It credited the frameworks: the MDMP forced the staff to generate three distinct options instead of latching onto the first one; the pre-mortem forced them to visualize failure modes before they materialized; the red team forced them to see the operation through the enemy’s eyes rather than their own. The fuel-shortage risk was not discovered during execution—it was discovered during a structured imagination exercise three days before the first vehicle moved. That is the entire point of the J-3 system. The frameworks do not guarantee success. They guarantee that failure modes are seen before they happen, which is the only kind of guarantee available in high-stakes operations.

The civilian equivalent is not complicated. Before any major project or critical operation, run a pre-mortem with your team: imagine it failed catastrophically, then list every reason why. Do not let anyone say “that won’t happen.” Write down the reasons. Then build mitigations for the top three. That exercise takes 45 minutes. It will identify problems your standard planning process misses because standard planning assumes success. The military assumes failure and plans backward from there. That is the operational habit worth stealing.

What to Do Next: Build Your Own J-3 Decision System

The most transferable tool from the J-3 Operations Directorate is not the MDMP framework itself—it is the institutional habit of treating every decision as a hypothesis that must survive structured attack before it becomes an order. Most civilian organizations plan forward from success. The military plans backward from failure. That single inversion changes everything. Start with the OODA Loop, developed by Colonel John Boyd in the early 1970s, which forces a cycle of Observe, Orient, Decide, Act. The critical insight is that orientation—the step where you update your mental model with new information—is the most frequently skipped phase in civilian decision-making. Executives observe a problem and jump straight to decide, skipping the deliberate re-framing that prevents old assumptions from driving new choices. The MDMP formalizes this by requiring planners to generate at least three distinct courses of action before any evaluation begins. If your team cannot articulate three genuinely different options, you have not done the work.

Decision fatigue is a documented phenomenon, and military planners mitigate it by prioritizing critical decisions early in the planning cycle and enforcing rest periods. The RAND Corporation and NIH research both confirm that decision quality degrades measurably after sustained cognitive load. The practical takeaway: schedule your most important decision for the first hour of the day, not the seventh hour of a meeting. The pre-mortem, popularized by psychologist Gary Klein and used in military wargaming, is the single highest-leverage exercise available. Set a timer for 30 minutes. Assume the project failed catastrophically. Write the post-mortem. Then fix the top three failure modes. Field reports from defense contractors and special operations units consistently note that pre-mortems identify risks that standard planning misses because standard planning assumes success. The military assumes failure and plans backward from there.

Red-teaming is not optional. Assign two people who are not on the planning team. Give them one job: find every flaw in the plan. Reward them for doing it well. The Joint Staff J-3 does not treat red teams as adversarial—it treats them as the cheapest insurance available. Never say “I’m sure” without a number. This forces precision where most leaders default to bluster. The “Call Out, Call Back” protocol is a standard military practice for any verbal instruction that matters. If the receiver cannot repeat it back, the instruction is not clear. Fix it before acting. The FAA uses the same protocol for air traffic control. It works because it exposes ambiguity immediately rather than after the mistake.

After every major decision, conduct an After-Action Review (AAR). Ask: what was supposed to happen? What actually happened? Why? No blame. Only process improvement. The U.S. The final step is continuous mission analysis: set a calendar reminder for one month from now to review whether the decision is still valid. Conditions change. The military builds that review into the planning cycle rather than treating it as an exception. Most civilian organizations treat a decision as final. The J-3 treats a decision as provisional until conditions prove it correct.

The concrete action: before your next major project, run a pre-mortem with your team. Do not let anyone say “that won’t happen.” Write down every reason the project could fail. Then build mitigations for the top three. That exercise takes 45 minutes. It will identify problems your standard planning process misses because standard planning assumes success. The military assumes failure and plans backward from there. That is the operational habit worth stealing.

What to do next

The frameworks and protocols used by the Joint Chiefs Operations Directorate are not classified secrets—they are documented, teachable methods. You can apply these same principles to your own high-stakes decisions by studying the original sources and practicing the techniques in low-risk environments before you need them under pressure.

Step	Action	Why it matters
1	Read the official U.S. Army MDMP manual (Field Manual 6-0) available on the Army Publishing Directorate website.	Understand the seven-step planning process used by military staffs to translate commander intent into executable orders.
2	Study John Boyd’s original OODA Loop briefing slides, archived at the Defense Technical Information Center (DTIC).	Grasp the original cognitive cycle—not simplified versions—to see how orientation shapes decision speed and accuracy.
3	Run a pre-mortem exercise with your team using the template from Gary Klein’s Harvard Business Review article “Performing a Project Premortem.”	Identify failure points before they happen by imagining the project has already failed and working backward to causes.
4	Download the CIA’s “Analysis of Competing Hypotheses” workbook from the Center for the Study of Intelligence.	Reduce confirmation bias by systematically evaluating multiple explanations against the same evidence.
5	Review the current Law of Armed Conflict (LOAC) manual published by the U.S. Department of Defense.	Understand the legal principles of distinction, proportionality, and military necessity that govern ethical military decisions.
6	Set a recurring calendar reminder to practice a “Call Out, Call Back” protocol during your next team meeting or project review.	Build the habit of closed-loop communication to eliminate ambiguity in high-stress environments.

Also worth reading: Why Date-Setting Fails in High-Stakes Decisions · Mastering High-Stakes Decisions When the Future Is Unclear · Synapse Density: The Hidden Key to Better High-Stakes Decisions · Entrepreneurial Lessons from Gearbox's Randy Pitchford Navigating the High-Stakes Game-to-Film Adaptation Business

Sources: wikipedia, economicblogs, linkedin, defensedaily, affordanything