Get Your Weekly Digest

Receive the best podcast summaries and insights delivered to your inbox every week.

We respect your privacy. Unsubscribe anytime.

We Study Billionaires - The Investor’s Podcast Network

TIP773: How Systems and Simple Math Shape Better Investing w/ Kyle Grieve

Published November 30, 2025
Visit Podcast Website

About This Episode

Host Kyle shares mental models from systems thinking and mathematics that shape his personal investing approach. He explains concepts like feedback loops, kill criteria, cone of uncertainty, scale, algorithms, critical mass, compounding, power laws, randomness, and regression to the mean, grounding each in concrete investing examples. Throughout, he emphasizes structuring decisions to favor long-term cash flow compounding while surviving volatility and avoiding portfolio blowups.

Topics Covered

Investing Systems thinking Risk and uncertainty Compounding Feedback loops Kill criteria Cone of uncertainty Power laws Randomness Regression to the mean

Disclaimer: We provide independent summaries of podcasts and are not affiliated with or endorsed in any way by any podcast or creator. All podcast names and content are the property of their respective owners. The views and opinions expressed within the podcasts belong solely to the original hosts and guests and do not reflect the views or positions of Summapod.

Quick Takeaways

  • Feedback loops, both balancing and reinforcing, are core to understanding portfolios, businesses, and how compounding works in practice.
  • Kill criteria-predefined "if X by time Y then sell" rules-help close long feedback loops and counteract complacency and emotional decision-making.
  • The cone of uncertainty is a way to align position size with how predictable a company's future appears, concentrating capital in the most certain ideas.
  • Scale creates both economies and diseconomies; as companies grow, new nonlinear problems, costs, and complexity emerge that can weaken or strengthen the business.
  • Algorithms in investing are simple decision recipes that only work if you actually act on their outputs, from rebalancing rules to cash-flow-focused selection.
  • Critical mass and inflection points describe when operating leverage and prior investments start to produce self-sustaining profit and cash flow.
  • Compounding is convex: a few big winners can overwhelm many losers, especially under power-law distributions of returns.
  • Randomness means even great processes can have bad outcomes, so survival-avoiding margin, shorting, over-concentration, and market timing-is essential.
  • Regression to the mean explains why extreme good or bad results tend not to persist and why distinguishing skill from luck requires a long-term view.
  • Layering these mental models together creates a robust framework for long-term, cash-flow-focused investing in an uncertain world.

Podcast Notes

Introduction and episode overview

Asymmetry, power laws, and convex compounding in investing

Over long periods, a handful of stocks drive the vast majority of portfolio returns[0:02]
• Even if half of your investments fail, the big winners can more than make up for the losers due to asymmetric outcomes and power laws
Understanding power laws and convex compounding changes how you think about investing[0:14]
• These mathematical forces operate within real-world systems to create highly skewed return distributions

Episode focus and intended audience

Mental models from systems thinking and mathematics will be explored and tied to investing decisions[0:27]
Topics include feedback loops, kill criteria, cone of uncertainty, scale, algorithms, critical mass, compounding, power laws, randomness, and regression to the mean[0:35]
The episode is aimed at investors who value long-term thinking and want an edge not based on predictions or noise[1:22]

Foundations in systems thinking

Host's introduction to systems and key books

First exposure to systems was through "Thinking in Systems, A Primer" by Donella Meadows[2:10]
• Main takeaway: small changes in one part of a system can cause massive changes in outputs elsewhere, which can be desirable or undesirable
Further learning through "The Great Mental Models, Volume 3: Systems and Mathematics" by Farnam Street[2:39]
• Book compiles numerous mental models from systems and mathematics that influenced his thinking
He also integrated a mental model from Annie Duke's book "Quit" (kill criteria) into this episode[2:53]

Definition of feedback loops and types

Meadows' definition of a feedback loop[3:16]
• A feedback loop is a closed chain of causal connections from a stock through decisions/rules/physical laws/actions that depend on the stock level, back through a flow to change the stock
Simplified: outputs of a system affect its own behavior[3:36]
• Here "stock" is an abstraction for an amount inside a system, not an equity security
Two types of feedback loops: balancing (stabilizing) and reinforcing[4:36]
• Balancing feedback loops are equilibrating structures that generate stability and resistance to change
• Reinforcing feedback loops are self-enhancing and create exponential growth or collapse

Example of a feedback loop: savings account

Interest-bearing savings account as a simple feedback system[3:49]
• If you leave money in the account, interest accrues and is deposited, increasing the balance
• You can withdraw interest (outflow) to keep the balance constant, or add contributions to grow it faster

Balancing feedback loop example: asset allocation

Host's personal asset allocation as of the discussion[5:28]
• Rough allocation: about 7% crypto, 88% public equities, 5% cash
Inflows and outflows change the stock of each asset class[5:28]
• Inflows: adding cash to brokerage, receiving dividends, price increases in equities/crypto
• Outflows: selling assets, withdrawing cash, or price declines in equities/crypto
Rebalancing as a balancing feedback mechanism[6:04]
• If equities rise from 88% to 90%, he would sell some equities (outflow) to restore target allocation
• In a bad year like 2022, he would buy more equities using cash or by selling some crypto to maintain targets

Reinforcing feedback loops and compounding in portfolios

Reinforcing loops and exponential growth

Reinforcing feedback loops are especially interesting because they involve exponential change[7:00]
Savings account example extended to compounding[7:05]
• If you leave interest in the account (no withdrawals), interest compounds on a growing base
• Regular deposits accelerate the reinforcing effect and growth of the balance

Using reinforcing loops to model portfolio growth and risks

Host's target: roughly 15% annual returns so money doubles ~every 5 years[7:38]
• He notes it does not take many doublings at that rate to reach financial independence, his ultimate goal
Interruptions to compounding as breaks in the reinforcing loop[7:00]
• Examples: selling a long-term compounder too early, withdrawing portfolio cash for spending or emergencies, or buying a house
• He references Charlie Munger's comment that people interrupt compounding in innumerable ways
Reinforcing loops can work against investors in fragile business models[8:47]
• If a business must invest capital every year just to maintain operations, loss of capital access can cause rapid collapse in equity value
He uses feedback loops when forming investment theses and maintaining due diligence[9:17]
• A thesis implies a system with inputs and expected outputs that can be monitored over time

Business example: Sezzle and its key system inputs

He analyzes Sezzle (a buy now, pay later company he does not own) in system terms[9:28]
• Key required inputs for success: increasing gross merchandise volume, growing monthly on-demand subscribers, and maintaining or improving customer credit quality
• If customer credit quality deteriorates significantly, lender relationships and funding could be at substantial risk

Kill criteria as pre-commitment tools in feedback loops

Definition and structure of kill criteria

Kill criteria are a form of pre-commitment contract to make decisions when noise will make them hard in real time[10:26]
Annie Duke's description of kill criteria[10:35]
• Best quitting criteria combine two things: a state (measurable condition/benchmark) and a date (when to evaluate it)
• Typical forms: "If I am in state X at time Y, I have to quit" or "If I haven't achieved X by Y time or resource spent, I should quit"

Role of kill criteria in closing long feedback loops

Kill criteria help close feedback loops that take a long time to resolve, especially in investing[11:20]
• Example: a company making long-term investments (new product rollouts, advertising, R&D, facilities) will show unattractive short-term numbers
• If these investments are likely to generate future earnings, they are good decisions, but results may not be clear for years
Using objective milestones for decisions[12:11]
• Example: after margin compression from investments, set a future target such as margins improving from 5% to 8% within a few years
• If margins are ≥8% by the date, no action; if below 8%, consider full or partial sale

Kill criteria in practice: inflection point bucket and specific stock example

Shorter leash for "inflection point" businesses[12:55]
• He expects about 25% profit growth from such firms and removes them if they cannot keep up
Case: Thermal Energy International[13:10]
• Business provides energy efficiency and emission reduction solutions for the industrial sector
• On September 30, 2024 he wrote criteria for the next year: 37-40 paid development agreements, $35-37M order intake, $22-24M backlog
• Rule: if the business fails to achieve two of these three, then sell
• He judged all three numbers as out of reach, saw insufficient momentum, and sold in February-March 2025, redeploying capital elsewhere

Cone of uncertainty and conviction sizing

Concept of the cone of uncertainty from Nomad letters

Nick Sleep and Qais Zakaria wrote that fewer things will happen than can happen, and investors should exploit that[14:18]
• They focus on company behavior that makes the future more predictable and lowers investment risk
Costco as an example of a narrow cone of uncertainty[14:48]
• Costco's obsession with sharing scale benefits with customers makes its future more predictable and less risky, justifying it as their largest holding
Smaller holdings have wider cones of uncertainty[14:40]
• These may do much better as investments but are less predictable, making position sizes smaller

Applying the cone of uncertainty to portfolio management

Visualization: wider vs narrower cones representing many vs fewer possible futures[15:28]
• A company like Costco can be viewed with a smaller-diameter cone, indicating more certainty about future outcomes
His framework: largest positions should have the narrowest cones of uncertainty[16:02]
• Higher certainty implies better grasp of future cash flows and fewer risks that can derail the thesis
Example from his microcap investing[17:25]
• His biggest winner came from microcaps; initial position was around 1.5% by cost due to a wide cone of uncertainty
• As the business grew and the cone narrowed, he averaged up; even after a 10x, he views its cone as wider than some other holdings like Atopicus or Dino Polska

Scale: how growth changes business dynamics

Definition and nonlinearity of scale

Scale refers to size/magnitude of a system or process and how that size changes behavior, cost, complexity, and dynamics[17:33]
Relationships and costs do not necessarily change proportionally with size[18:03]
• As scale increases, some relationships become nonlinear, which is hard for the human brain to intuitively grasp

Economies and diseconomies of scale

Positive aspect: economies of scale[18:23]
• Example: automation allows higher output without proportional increases in labor costs, improving margins as volume grows
Negative aspect: diseconomies and new problems[19:04]
• Automation may require hiring a full-time engineer, and doubling output may strain shipping partners and logistics

Scale and management challenges

Experience at small scale may not translate to managing at larger scale[18:54]
• Example: a CEO with 30 years on a non-automated factory floor may not be the right person to lead a business through automation-driven scaling
Question of whether a serial acquirer's model scales[23:38]
• He and a mastermind member debated whether a company that bought small niche businesses for a few million could succeed buying $10M companies under the same criteria

Investor expectations vs scaling reality

As production scales, investors may incorrectly assume new capacity flows straight to operating profit[24:44]
• In practice, scaling often necessitates more sales staff, higher SG&A, and sometimes increased R&D to support growth
Tracking R&D and SG&A as percent of revenue[25:24]
• If these ratios shrink as revenue grows, it indicates economies of scale; if they rise, it shows diseconomies of scale

WeWork case as an example of bad scaling incentives

Rapid revenue and headcount growth masked poor unit economics[26:07]
• Revenue grew from $436M to $1.8B in two years but required $1.9B in spending to achieve, a sign of value destruction
Distorted KPIs and community-adjusted EBITDA[26:55]
• WeWork used run-rate revenue by annualizing a single month and invented "community-adjusted EBITDA" that excluded major operating expenses like rent, utilities, and building staff
• Such KPIs incentivized top-line growth "by any means" rather than real profitability

Algorithms as decision recipes in investing and life

Definition and properties of algorithms

An algorithm turns inputs into outputs and forms the working parts of a feedback loop[28:38]
Repeatability: correct inputs into a good algorithm produce the same outputs consistently[28:38]
• Pie-baking analogy: following exact ingredients and instructions yields the same pie; estimating or changing parameters produces a different output

Kill criteria and cone of uncertainty as algorithms

Kill criteria and cone of uncertainty both act as decision algorithms when fed data[29:20]
• They process quantitative and qualitative inputs and suggest actions (e.g., hold, trim, sell, resize positions)

Art vs science in using algorithms

Quantitative data is the "science"; interpreting and acting on it is the "art"[29:39]
• Example: making a judgment about whether the cone of uncertainty is widening or narrowing is subjective
• If kill criteria dates arrive and the business appears close to milestones, deciding to extend or enforce criteria is an art decision
Algorithms are useless if you ignore their outputs[30:02]
• If an algorithm signals action but the investor refuses to act, the model has no real-world applicability

Charlie Munger's "secret algorithm" and host's life algorithms

Munger said the secret algorithm to life is doing the right thing, doing more of what is working[30:24]
Host's personal "what's working" list[31:53]
• Spending time with family, especially his son
• Investing in a mix of long-term and short-term opportunities
• Practicing gratitude as often as possible
• Actively working on becoming kinder each day
• Optimizing for happiness alongside doing his job well, and prioritizing physical, emotional, and psychological self-care

Buffett's perspective on a great life and algorithm for success

Buffett wrote that greatness does not come from accumulating money, publicity, or power[32:38]
• He emphasized that helping others in thousands of ways improves the world, and that kindness is costless but priceless
• Buffett praised the golden rule as a guide to behavior, religious or not
Host links Buffett's internal scorecard and golden rule to finding one's own life algorithm[33:33]

Cash-flow-focused investing algorithm

Long-term algorithm: focus on cash flow generation[33:39]
• If a business generates increasing cash flow over time, its share price is likely to perform very well
Preference for operating cash flow and owner's earnings over free cash flow alone[33:39]
• He owns many businesses that reinvest heavily; investing cash outflows depress reported free cash flow
• Using Buffett's owner's earnings helps reveal true cash generation before growth investments
Evidence from the Magnificent Seven[35:28]
• He notes the average 10-year CAGR of cash from operations for the seven large tech leaders is over 27%
• At a 27% CAGR with a fixed multiple, a business can roughly 11x in value over 10 years
Selling long-term cash-flow compounders is a major error[37:22]
• He stresses repeatedly that if a business can compound operating cash flow for many years, selling is a big mistake
• However, along the 10-year path, any single year can look very different due to volatility and randomness

Critical mass and inflection point investing

Definition of critical mass and Tesla example

To fully benefit from a compounder like Tesla over 10 years, investors had to endure multiple 40%+ drawdowns[38:38]
Critical mass defined[39:18]
• It's the point where enough participants or resources converge so a system becomes self-sustaining or gains radically increased momentum

Thresholds and the "magic third" from tipping point thinking

Malcolm Gladwell's idea: changing about a third of something can reach a tipping point[40:00]
• Example: on a nine-person board, shifting or replacing two or more members can flip control and enable change
Host prefers to find companies where the right people are already in place rather than trying to change them as a board member[41:04]

Leverage points and serial acquirers

Leverage points: situations where accumulated cash and high-return deployment opportunities can drive a system toward critical mass[42:09]
• For a serial acquirer, starting with one M&A person and evolving into many decentralized units hunting deals can create a self-sustaining system, like Constellation Software or Bergman & Beving

Time, operating leverage, and disappointment risk

Critical mass timing varies by company[42:06]
• A company might be near critical mass, or it might be far away, which affects investor expectations and satisfaction
Inferring operating leverage too early can lead to disappointment[43:09]
• If investors expect margins to inflect in the next few quarters but the real leverage is years away, the mismatch fuels negative surprises

Reversal of critical mass and product-based examples

Critical mass can be reversed if key inputs decline sufficiently[43:28]
• Example: a hit product without a durable competitive advantage invites competition that can erode its advantage and momentum
Case: Cannabis Capital[44:01]
• He owned a cannabis company whose cigarette-shaped products drove short-lived revenue and profit booms
• Investors, including him, mistakenly extrapolated this advantage, but once the product momentum faded, the business lost critical mass and stalled

Inflection point strategy as quantitative signal of critical mass

He often looks for businesses with two consecutive quarters of positive cash flow or profits[45:06]
• This usually signals that past expenses are now supporting growth, and incremental revenue is flowing through above fixed costs
He manages downside risk by buying such names cheaply and with smaller position sizes than more established compounders[45:23]

Compounding, convexity, and power laws in portfolios

Visible and hidden compounding

Compound interest is the most visible form of compounding[46:52]
• Reinvesting interest leads to earning returns on a growing base, with mind-boggling potential over long periods
Hidden compounding and credit card interest example[48:07]
• Many cardholders misunderstand that credit card interest is compounded daily rather than once a year
• Example: $5,000 balance at 20% APR compounded daily grows to about $6,100 after a year with no payments, not just $6,000
• Minimum payment practices allow issuers to profit from compounding interest on carried balances

Convexity of compounding and Gautam Baid's insight

Compounding is convex to the upside and concave to the downside, leading to positive asymmetry[50:05]
Two-stock thought experiment with +26% and -26% CAGR[50:16]
• Starting with two $100 stocks, after one year the winner is $126 and the loser $74, totaling $200
• After 10 years, the portfolio CAGR is around 18%, even though one stock is almost worthless, because the winner's gains dominate
Gautam Baid's conclusion: he can be wrong about 50% of the time and still earn great returns due to this asymmetry[50:32]

Contribution analysis of real portfolios

Gautam's example: 80% of his returns came from just four of 23 positions[50:34]
Host's year-to-date contribution analysis[53:04]
• Top 4 positions contributed about 53% of his gains; top 2 contributed 42% of portfolio returns across 19 positions held that year
Longer-term contribution since portfolio inception[53:23]
• About 57% of returns came from just five businesses; bottom five positions contributed -17%
• This illustrates convex winners dominating concave losers in overall compounding

Power laws in businesses and diversification

Power law definition[53:55]
• A power law describes a distribution where a small number of outcomes account for a large proportion of results
Power laws at the business level[54:17]
• One product or service can catapult a company into the stratosphere, with upside far larger than the total possible downside (which is capped at 100%)
Imagination is necessary to see optionality and avoid selling potential power-law winners too early[53:46]
Power laws and portfolio rebalancing[54:47]
• Capping position sizes via frequent rebalancing can mitigate risk but may also constitute poor capital allocation by truncating upside of winners
• He believes power-law-following businesses should be among the largest positions, and trimming them reduces future return potential

Shopify case study and avoiding misleading averages

Using industry averages can mislead when one company is an emerging outlier[56:21]
2015 Shopify scenario[55:01]
• If investors assumed Shopify's high growth would regress to the industry average ~20% and sold after a small gain, they missed a power-law outcome
• In reality, Shopify built strong network effects, an ecosystem, and top-decile merchant base, compounded revenue at ~50% CAGR and GMV at ~63% CAGR, becoming category-defining

Randomness, survival, and risk management

Nature of randomness vs compounding's smooth curve myth

Compounding appears as a smooth exponential curve on paper but unfolds in a volatile, uncertain world[57:31]
Randomness defined as lack of predictable pattern in single events[58:04]
• We may understand the distribution of outcomes, but any individual outcome remains uncertain
Short-term performance is dominated by noise[58:06]
• Daily stock prices, quarterly earnings, analyst expectations, and sentiment are mostly noise in the short term
• In the long run, fundamentals drown out randomness rather than the other way around

Randomness and power-law returns

We cannot know in advance which holding will become the power-law winner[59:52]
• We can, however, understand that return distributions are skewed and design portfolios to let outliers matter

Portfolio design: expose to upside randomness, protect downside

Position sizing based on conviction and certainty[1:00:13]
• Narrower distributions of outcomes (higher certainty) justify larger position sizes; adding risk by reducing those positions is, to him, nonsensical
Key rules: don't cap winners, limit downside, and avoid over-diversification that dilutes impact of big winners[1:01:15]

Pattern recognition traps and macro forecasting

Humans see patterns even where none exist, leading to misplaced confidence in forecasts[1:01:14]
Macro forecasters often appear knowledgeable but are usually wrong or just lucky[1:01:22]
• He notes COVID-19 as an example where early expectations about market impact were repeatedly wrong, including assumptions about crashes and subsequent performance

Process vs outcome and humility

A sound process can yield a bad outcome; a poor process can yield a good outcome[1:02:01]
Thought experiment: bet with 90% chance to double, 10% chance to lose half[1:02:27]
• You should take this bet every time; losing in a particular instance doesn't mean the process was wrong, just that randomness prevailed
Randomness fosters humility and acceptance of being wrong often[1:03:33]
Buffett example: out of 300-400 investing decisions, a dozen drove the majority of Berkshire's success[1:04:33]

Survival as the primary objective

Compounding at modest single-digit rates over decades can build a substantial nest egg[1:04:00]
• Too many investors blow up their portfolios and leave markets, permanently ending their compounding potential
Inversion: avoid behaviors that can destroy portfolios[1:04:54]
• Avoid using margin; a leveraged bet gone wrong can take portfolio value to zero
• Avoid shorting; upside is capped at a double, while potential loss is unlimited
• Avoid extreme over-concentration; he has never put more than 15% of cost basis into one position and now is unlikely to exceed ~10%
• Avoid market timing; selling when markets "seem expensive" and buying back higher is a recipe for failure

Regression to the mean and its investing implications

Definition and origins of regression to the mean

Shane Parrish's description: luck is random, so outlier results with luck components are followed by more moderate ones[58:31]
Francis Galton's height studies[58:55]
• Tall parents tended to have shorter children; short parents tended to have taller children, moving offspring heights toward the population average

Sports example: the hot hand in basketball

1985 NBA study including Amos Tversky examined whether the hot hand is real or luck[59:57]
Findings: hot streaks are largely products of randomness for even highly skilled players[1:00:26]
Steph Curry example[59:41]
• Curry's career ~42% three-point average implies around 4 makes per 10 attempts on average, but he may hit four straight then miss six, giving an illusion of a hot hand

Case study: Scott Barbee and Aegis Fund

Aegis Fund experienced a 72% drawdown from 2007-2009 during the financial crisis[1:01:24]
Despite pressure, Barbee believed his system still worked and viewed the portfolio as de-risked with more upside[1:00:40]
Outcome: fund was up 91% in 2009, landing him on the front page of The Wall Street Journal[1:01:29]
• Host interprets this as Barbee benefiting from positive regression after an extreme negative event

Four key lessons from regression to the mean

1) Extreme outperformers are not reliably repeatable[1:01:41]
• Fund managers who crush the index for a year or two often underperform in others; over time results regress toward their true skill level
2) Misunderstanding regression leads to misdiagnosing skill vs luck[1:02:46]
• Observers in 2008-2009 might have thought Barbee lost his edge, underestimating the role of bad luck and extreme events
3) Remember base rates[1:02:50]
• Ola Svensson's research showed about 80% of drivers believe they are above average, though only ~50% logically can be
• As a stock picker, you implicitly claim to be above average; the only way to verify is long-term tracking against a benchmark (he uses the S&P 500)
4) Systems with high variability show the strongest regression[1:03:32]
• A fund that drops 72% can rebound dramatically when conditions reverse; milder volatility would likely produce milder regression
• He suggests that if your portfolio is down substantially but you are confident in fundamentals, you may be positioned for positive regression if you can ride out volatility

Synthesis of mental models and closing thoughts

Interconnection of systems and math models

He reiterates Charlie Munger's point that layering multiple mental models is more powerful than using any single one[1:04:20]
Recap of systems models covered[1:04:57]
• Feedback loops (reinforcing and balancing), kill criteria, cone of uncertainty, scale, and algorithms
Recap of mathematical models covered[1:04:57]
• Compounding (including convexity), power laws, randomness, and regression to the mean

Unifying theme: building robust decision systems

Goal is to reduce adverse effects of emotion, understand which feedback loops you're in, embrace randomness, and commit to long-term compounding[1:05:41]
Long-term success demands surviving short-term extremes[1:07:20]
• Extreme events often hurt in the short term but become tiny blips over long horizons if you survive them
You don't need to see the future; focus on avoiding mistakes that would prevent you from having one[1:07:44]
While avoiding failure, expose your portfolio to positive surprises from unlikely places[1:07:51]

Closing and contact information

He hopes at least one model helps listeners make clearer decisions or avoid costly errors[1:07:59]
He invites listeners to continue the conversation on Twitter at @IrrationalMRKTS and via LinkedIn by searching his name[1:08:11]

Lessons Learned

Actionable insights and wisdom you can apply to your business, career, and personal life.

1

Define explicit kill criteria (state + date) for your investments so that future decisions are guided by precommitments rather than emotions or narrative drift.

Reflection Questions:

  • • Which current investments could benefit most from having clear, measurable kill criteria defined in advance?
  • • How will you decide what states (KPIs, margins, cash flow levels) and dates are both realistic and meaningful for the businesses you own?
  • • What is one position you could sit down and write a concrete "if X has not happened by Y, then I will sell or reduce" rule for this week?
2

Size positions according to your cone of uncertainty: allocate more capital to businesses with narrower, more predictable futures and less to those with wide, uncertain outcome ranges.

Reflection Questions:

  • • Looking across your portfolio, which holdings genuinely have the narrowest and widest cones of uncertainty based on what you can objectively see today?
  • • How might re-aligning your position sizes with your true level of conviction and predictability change your risk and return profile over the next few years?
  • • What specific criteria (e.g., business model durability, customer loyalty, balance sheet strength) will you use to judge whether a company's cone is narrowing or widening?
3

Respect the convexity of compounding and power laws by letting winners run and avoiding behaviors-like constant rebalancing-that artificially cap your best ideas.

Reflection Questions:

  • • When have you sold or trimmed a big winner too early, only to watch it continue compounding without you?
  • • How could you adjust your rebalancing or profit-taking rules so they still control risk but leave room for power-law outcomes to emerge?
  • • Which one or two current holdings have the potential to be outsized contributors to your long-term returns if you give them enough time and space?
4

Accept randomness and regression to the mean by focusing on the quality of your process, managing downside risk, and prioritizing survival over chasing short-term performance.

Reflection Questions:

  • • Where are you currently judging yourself or others mainly on short-term outcomes instead of on whether the underlying decision process is sound?
  • • How might your strategy change if your primary goal shifted from maximizing returns in the next year to ensuring you are still compounding 20 years from now?
  • • What concrete safeguards (limits on leverage, position size caps, rules against market timing) can you implement to protect yourself from portfolio ruin?
5

Use systems thinking-feedback loops, scale effects, and critical mass-to anticipate how businesses will behave as they grow and to avoid extrapolating early successes blindly into the future.

Reflection Questions:

  • • Which companies you follow are at an inflection point where reinforcing feedback loops or operating leverage may be kicking in-or failing to kick in?
  • • How could you systematically track the key inputs and outputs (e.g., unit economics, customer acquisition, margins) that signal whether a business is moving toward or away from critical mass?
  • • What is one past investment where scale introduced problems you did not foresee, and how would a systems-thinking lens have changed your analysis?
6

Anchor your investment decisions on long-term cash flow generation rather than short-term earnings noise, and avoid interrupting compounding unless there is a clear, process-driven reason to do so.

Reflection Questions:

  • • For each major holding, do you know what its operating cash flow or owner's earnings trajectory looks like over the last several years?
  • • How would your buy, hold, and sell decisions change if you made long-term cash-flow growth your primary filter instead of near-term price movements or headline earnings?
  • • What policies could you adopt (such as separate emergency funds or explicit spending buckets) to reduce the need to liquidate compounding assets for short-term cash needs?

Episode Summary - Notes by Phoenix

TIP773: How Systems and Simple Math Shape Better Investing w/ Kyle Grieve
0:00 0:00