Geometric Utility Investing (GUI): My Framework for Conviction-Based Dip Buying

What if investing felt less like gambling and more like a strategic game you actually enjoy playing? Geometric Utility Investing (GUI) is a conviction-based framework that turns market dips into calculated opportunities — by making the rational action and the emotionally rewarding action the same action, at every point in the process. Built on 20+ years of behavioral design research, GUI doesn’t just tell you what to buy. It engineers the decision environment so you never need willpower.

⚡ Speed Run Notes

• GUI optimizes for happiness, not just returns. I engineered this so the rational action and the emotionally rewarding action are always the same action.
• Three core principles: (1) Maximize happiness. (2) Bet on what’s already proven. (3) Design the system so good decisions also feel good.
• Spoon/Plate/Hatchet taxonomy tells me whether a dip is emotional (buyable) or rational (avoid).
• Four Guardians screen every asset before entry: structural moat, secular uptrend, recent resistance line, improving fundamentals.
• The Thirds Strategy sells 1/3 of the current position at each milestone, so I mathematically never fully exit a winning position.
• RPG Party Composition classifies assets by portfolio function (Tank, DPS, Healer, Bruiser, Skirmisher, Bulwark) to survive different crash types.
• Conviction Check system prevents the single fatal flaw: buying all the way down on a broken thesis.
• Conversion Track Record adjusts conviction for heavy reinvestors. A company spending gold on XP looks poor on its character sheet while leveling. The check is whether past level-ups actually landed, and whether this one has a visible progress bar.
• GUI’s two ways to lose: (1) Unrecognized broken conviction — your thesis is wrong and you don’t realize it. (2) Forced liquidity — you need cash during a 2–4 year dip cycle and sell low when you should be buying. Every other risk is manageable.

---

About the Author

Yu-kai Chou is the creator of the Octalysis Framework and one of the earliest pioneers in gamification, having studied the field since 2003. His behavioral design work with LEGO, Microsoft, Porsche, Coca-Cola, Salesforce, and MrBeast has impacted over 1.5 Billion Users.

Yu-kai has shared these insights at Harvard, Stanford, Yale, Tesla, Google, BCG, and IDEO.

His research has been referenced in 3,700+ academic publications including work from Harvard, Stanford, MIT, Forbes, Wall Street Journal, Wired, US Department of Energy, NIST, NSF, NCBI, US Department of Education, and ClinicalTrials.gov. Explore his books here.

Why GUI Exists

Most investing frameworks optimize for returns. GUI optimizes for happiness.

Here’s what I kept seeing: I do the research. I identify the right asset. I buy the dip. Then at 3 AM, the portfolio is down 30%, and I panic-sell at the bottom.

The analysis was perfect. The execution failed because the system wasn’t designed for how I actually feel.

GUI solves this by making the rational action and the emotionally rewarding action the same action, at every point in the process. When markets crash, the system tells me exactly what to do and why that action should feel good. When markets rally, the system takes profit in a way that never leaves me regretting selling too early or too late. The architecture does the work so I never need willpower.

Gambling is optimism about the unknown. GUI is conviction about the demonstrated.

The entire framework is built around what’s already proven: proven resistance lines, proven survival ratchets, proven fundamentals. GUI doesn’t speculate on future highs. It identifies assets where the market has already demonstrated willingness to pay a certain price, and buys them below that price when fundamentals are even stronger.

GUI’s three principles:

1. Maximize happiness and fun. Investing should feel like a strategic game worth playing, not a threat to wellbeing.
2. Bet on what’s already proven, not what is hoped will happen. The opposite of gambling.
3. Design the system so the right decision also maximizes emotional wellbeing. If the system is working, I never need willpower.

This is the same design philosophy behind the Octalysis Framework: instead of telling people “have more willpower” (the equivalent of “have more discipline”), design the system so that the desired behavior is also the path of least emotional resistance.

The Origins: Behavioral Design Meets Investing

I spent two decades studying why people do what they do. The Octalysis Framework was my answer for product design: eight Core Drives that explain all human motivation, from epic meaning to loss avoidance. Companies like Google, LEGO, and Microsoft used it to redesign how billions of people interact with their products.

But I kept noticing the same patterns in my own investing. The emotional traps that make people quit a poorly designed app are the same ones that make investors sell at the bottom: loss avoidance (Core Drive 8) overriding rational analysis, scarcity and impatience (Core Drive 6) driving panic decisions, the absence of empowerment and feedback (Anti Core Drives) leaving investors without a clear next action when things go wrong.

So I asked a different question. What if an investing framework was designed the way a good game is designed? Not “gamified investing” with fake points and leaderboards, but engineered so that every decision point produces both the rational and the emotionally satisfying outcome at once?

GUI is that framework.

I. Long-Term Conviction: “Where Is the World Going?”

Before anything else, before the chart shapes, before the Guardians, before the execution system, I need to answer one question:

“Do I have strong conviction that this asset’s growth is the natural direction of the world?”

Not “is it possible?” or “is it likely?” I need to believe, based on evidence, that the ten-year trend is clear. Conviction here isn’t about price targets — it’s about whether this thing is becoming more dominant in society, not less. Will more people, institutions, and systems depend on it in five to ten years than do today? That’s a fundamentals question, not a chart question.

Ten-year trends are easier to read than one-to-two-year price swings. Interest rates spike and drop. Recessions come and go. But over a decade, the big secular trends tend to hold, because they’re driven by how human behavior is actually changing, not by what the Fed said last Tuesday.

The question isn’t “will this asset’s price go up?” That’s a trading question. GUI is not a trading system. The question is about the underlying reality the asset represents. Is the world moving toward more of this, or less of this?

A note on the word “conviction”: I’m not talking about blind faith. Conviction here means a belief backed by observable evidence, tested against falsifiable criteria (that’s what the Conviction Check in Section V is for). Strong conviction can still be wrong. The framework accounts for that. But without strong conviction as the starting point, nothing else works.

Examples of Strong Conviction

“Ten years from now, more people will buy things online than today, not fewer. As long as Amazon leads e-commerce, Amazon will grow.” The conviction isn’t about Amazon specifically; it’s about online commerce. Amazon benefits because it leads a clear secular trend.

“Ten years from now, more people will drive electric cars than today, not fewer. As long as Tesla leads in EVs, Tesla will grow.” Again, the conviction is about the shift to electric vehicles. Tesla benefits by leading it.

“Ten years from now, more institutions and sovereign nations will hold Bitcoin than today, not fewer.” The conviction is about the adoption of digital store-of-value assets. Bitcoin benefits as the established leader.

“Ten years from now, more devices will require advanced semiconductors than today, not fewer.” The conviction is about the expansion of computing. NVIDIA benefits by dominating the critical infrastructure layer.

What Fails the Conviction Test

“Ten years from now, it is unclear if more people will use NFTs or not.” Speculative. No clear evidence the world is moving toward more NFT usage. Without that conviction, buying a dip on an NFT-focused asset is gambling on an uncertain future.

“Ten years from now, it is unclear if more people will use this specific social media platform.” Social platforms can be displaced (Myspace to Facebook to ???). Social media as a trend is here to stay; any specific platform is not.

The Two-Part Structure

This conviction has two components:

1. The trend is clear. More people will do X, not fewer.
2. The asset leads the trend. As long as this specific asset maintains its position, it will capture that trend.

Both must hold. A clear trend with a weak leader means the trend grows but the specific asset doesn’t capture it. A strong leader in an unclear trend means the company is excellent but the tailwind might disappear.

Why this comes first: If the conviction isn’t there, nothing else in the framework matters. The spoon shape doesn’t matter. The Guardians don’t matter. But if the conviction is there, every dip becomes psychologically manageable. The price dropped 30%? If I genuinely believe more people will use this asset in ten years, not fewer, then the drop is a discount on a future I believe in. The conviction does the heavy lifting; the rest of the framework just keeps me disciplined.

II. Pre-Screen: The Upward Facing Spoon

Before running the full Guardian set, GUI visually checks the chart for an “upward facing spoon” pattern.

The shape of a price chart isn’t just a pattern. It’s a compressed narrative of collective human psychology. Each shape tells a different story about whether the previous high was real, and that story determines whether buying the dip is smart or suicidal.

The Spoon (What GUI Looks For)

Imagine a spoon resting on an upward slope: the handle starts low on the left, rises as it goes right, dips into the bowl, and continues rising on the other side. The whole spoon trends upward. That tilt is the secular uptrend. The bowl is a temporary interruption.

The handle (left side, rising): A sustained uptrend over months or years. Each day the price crept higher was a vote. Thousands of independent participants (institutions, retail traders, analysts, algorithms) kept arriving at the same conclusion: “Worth more than yesterday.” A long, smooth handle isn’t noise. It’s consensus building over time through independent analysis.

The bowl (the dip): A sudden, sharp drop. Steep and fast, not a slow grind down. The speed is the signal. Fast moves are almost always emotional, not rational. Rational repricing is slow, because it requires new information to propagate through thousands of independent decision-makers. When everyone sells at the same time, that’s not independent analysis. That’s herd fear.

Why the spoon works: The bowl didn’t erase the reasons the handle existed. If an asset spent two years climbing because institutional adoption was growing, developer ecosystems were expanding, and revenue was improving, a three-day crash because of a Fed announcement didn’t undo any of that. Adoption is still growing. Developers are still building. Revenue is still there. The price just temporarily disconnected from the fundamentals that built the handle.

The Catalyst Test: Ask: “Can I name the specific event that caused the bowl?” If yes (Fed announcement, tariff escalation, contagion selloff, geopolitical shock), that’s a spoon. A nameable catalyst means the drop was event-driven, not a slow realization that the asset is overvalued. If there’s no clear trigger and the decline simply accumulated over weeks without cause, that’s a plate forming.

Disqualifying Shapes

The Plate (The Trap That Looks Like Opportunity)

Imagine a plate sitting on a table. It rises gently on the left side, then slowly curves flat. No sharp bowl. Just a long, gentle decline that eventually levels off.

The most dangerous shape, because it looks almost like a spoon if you squint. People see a previous high on the left, a current low on the right, and think “discount!” The shape tells the opposite story.

A plate means the market didn’t panic. It thought about it. Over weeks and months, independent participants gradually concluded: “Actually, this isn’t worth what we thought.” Analysts lowered price targets. Institutions slowly trimmed positions. Retail interest faded. Each day’s small decline was a rational vote: “Worth a little less than yesterday.”

Why this is dangerous: A slow decline means the market is processing real information, not reacting to emotion. Buying a plate is betting that thousands of participants who spent months independently deciding “this is worth less” are all wrong, and that the less-informed version of the market (at the peak) was right.

How to distinguish plate from spoon: Time and speed. If the decline takes less than a few weeks and drops 30%+ in a clearly identifiable macro-driven event, that’s a spoon bowl. If the decline takes months, grinds slowly with no single clear catalyst, and flattens rather than V-shapes: plate.

Example: RENDER from ATH of $13.53 in March 2024 to $1.30 nearly two years later. No catastrophic event. The market just gradually decided the speculative premium was unjustified. $1.30 IS the informed price. $13.53 was the uninformed one.

The Hatchet (Spike and Fade)

Imagine a hatchet lying on its side: a single sharp spike straight up on the left (the blade), then a long declining handle stretching to the right.

The high was the anomaly. Something caused a one-time, non-repeatable explosion in price: a short squeeze, a viral meme, a speculative mania. The price shot up not because thousands of independent participants gradually concluded “this is worth more” (that’s a spoon handle), but because a self-reinforcing feedback loop briefly took over: price rising made people buy, which made price rise more. Then it stopped, and there was nothing underneath to hold it up.

The “resistance line” at the top of the hatchet spike is meaningless. No one actually valued the asset there through careful analysis. They bought it there because it had just been lower and they assumed it would keep going.

Example: Silver’s spike above $121 in late January 2026, driven by leveraged futures speculation and retail momentum with an estimated paper-to-physical ratio exceeding 300:1 on COMEX. Silver crashed over 31% in a single day (January 30-31). The $121 wasn’t a level the market reached through consensus-building. It was a speculative blow-off. Silver has done this three separate times (1980, 2011, 2026), proving the spikes don’t stick.

The Downward Staircase

A series of lower highs and lower lows. Each recovery attempt fails at a level below the last peak. Structural decline. The market keeps trying to believe, and keeps being disappointed. Do not buy.

III. The Four Guardians (All Must Pass)

Once an asset passes the Long-Term Conviction test and shows a spoon shape, it must pass through four Guardians. Each Guardian protects my happiness (and yours, if you use this) by screening out a specific way the investment could go wrong. If any Guardian rejects the asset, the asset is disqualified regardless of how strong the other signals look.

Guardian 1: Structural Moat

“Could a competitor absorb its use case?”

The trend I’m betting on must be tied to this specific asset, not just to the sector. If the use case could migrate to a competitor (like social networking migrated from Myspace to Facebook), the asset could decline even while the broader trend grows.

Guardian 2: Secular Uptrend

“Are cycle peaks getting higher over time?”

I check whether the asset’s all-time price history shows higher highs each cycle. If the ATH was 5 years ago and each subsequent cycle peak has been lower, the market is saying it overpriced this thing and keeps correcting further.

Guardian 3: Recent Resistance Line

“Has the market recently shown appetite at this price?”

I don’t use ancient ATHs as the resistance target. I use the high from the past 1-2 years. A recent resistance line reflects current-era market conditions, liquidity, and investor base. An old ATH from a different market regime may represent appetite that no longer exists.

Guardian 4: Improving Fundamentals Since Previous High

“Is the asset stronger now than when it last hit that price?”

The most powerful signal is when current fundamentals are better than they were at the previous high. The “proven appetite” at that price was based on a weaker version of the asset, so returning to that price is actually conservative.

Critical inversion: If fundamentals are worse than at the previous high, the framework’s logic flips. The “proven appetite” may have been overshoot, not floor. Disqualifier. Do not buy.

Important nuance on “proven appetite”: The buyers who pushed an asset to its previous high may be a partially different population than today’s market. Some rotated out, some blew up, some reallocated. The resistance line reflects the appetite of a specific set of participants at a specific moment. Guardian 4 is what bridges this gap: even if the buyer base has rotated, the reasons to buy are stronger now than they were at the previous high. New buyers with better fundamentals should be willing to pay at least what old buyers paid with worse fundamentals.

Guardian 4 Variant: Reinvestment-Phase Assets (Spending Gold on XP)

Guardian 4 has a blind spot worth naming. GAAP accounting systematically understates companies that reinvest heavily in their own future. R&D is expensed the moment the money leaves the building, which crushes current earnings. Capital expenditure is capitalized, which crushes current free cash flow and then leaks into future earnings as depreciation. Both are the same thesis category: growth spend. So a company converting every spare dollar into future capability will look weaker on the income statement than a company coasting on its installed base, even when the reinvestor is the stronger asset.

History is full of examples. Amazon “looked expensive” on P/E for two decades while being one of the great buys of the era, because the earnings were deliberately reinvested away. Meta traded near $88 in November 2022 at a single-digit forward multiple while the market treated Reality Labs spend as pure destruction. Tesla in 2019 production hell traded at a split-adjusted ~$14 while burning cash to build the machine that builds the machine.

Here is the trap: the survivor sample does a lot of work in that list. In real time, the income statement cannot distinguish investment from value destruction. WeWork called it investment. Quibi called it investment. Intel’s foundry buildout and Meta’s metaverse spend both wore the same label. The spending pattern alone predicts nothing.

The discriminator is the Conversion Track Record (Section V-F): demonstrated evidence that this specific company’s past growth spend converted into shipped products, margins, and market position. For a reinvestment-phase asset, the Guardian 4 question “is the asset stronger now than at the previous high?” gets answered with conversion milestones and unit economics instead of the GAAP bottom line. Two hard rules keep this honest:

• Gross margin is the tell. The investment excuse lives below the gross-profit line (R&D, capex, opex). It cannot excuse deterioration in the already-converted, delivered business. If gross margin on what the company sells today is eroding, Guardian 4 fails regardless of how exciting the spend narrative sounds.
• This variant changes what counts as evidence for Guardian 4. Every other gate stays. The asset still needs the spoon, the moat, the secular uptrend, and a price that has stopped believing. A reinvestment story at a premium multiple is the market pre-paying for conversion, and pre-paid conversion offers GUI nothing to buy.

IV. Decision Matrix and Execution

Expected Value Calculation

Expected Value = Upside to Recent Resistance × Conviction of Getting There

When comparing assets:

• Same upside, different conviction → pick higher conviction.
• Higher upside, lower conviction → calculate expected value; higher upside only wins if conviction is above ~55-60%.
• Factor in variance: a 3x at 70% conviction is better risk-adjusted than a 30x at 20% conviction, because the latter is bimodal (either a big win or a big loss).

Weighted Expected Value (WEV)

The simple EV formula above implicitly assumes total loss if the thesis breaks. That’s useful as a conservative baseline, but it penalizes assets with strong downside protection — like Bitcoin, which has institutional floors, or blue-chip equities with dividend support and tangible book value.

Weighted Expected Value = P(up) × Upside Multiple + P(down) × Floor Multiple

Where Floor Multiple = realistic worst-case price ÷ current price. This is the price the asset would settle at if the bull thesis breaks but the asset doesn’t go to zero.

Example: If I’m looking at an asset at $100 with a resistance line at $300 (3x upside), I assign 60% conviction it reaches resistance, and I believe the worst-case floor is $40 (0.4x). Simple EV = 0.60 × 3.0 = 1.80. WEV = 0.60 × 3.0 + 0.40 × 0.4 = 1.96.

The difference matters most when comparing assets with similar upside but different risk profiles. A crypto asset and a dividend-paying stock might have the same simple EV, but the stock’s floor protection gives it a higher WEV — which better reflects the actual risk-adjusted return.

Important: WEV is always higher than simple EV. That’s not a flaw — it’s because simple EV was already penalizing by assuming total loss. WEV just replaces that zero with a realistic floor. The improvement measures how much downside protection the asset actually has. Assets with hard floors (institutional support, tangible assets, dividend streams) get the biggest WEV boost over their simple EV.

What Disqualifies an Asset

• Chart doesn’t form an upward facing spoon (no sustained uptrend before the dip, or the decline is slow/grinding rather than sharp)
• No real fundamentals (no developer ecosystem, no protocol revenue, no utility beyond speculation)
• Declining cycle peaks (each bull market high is lower than the last)
• Inflationary tokenomics without offsetting demand
• Previous high driven by one-time cultural phenomenon rather than sustainable demand
• Competitor has absorbed its use case
• Fundamentals worse than at previous high
• Fundamental downtrend (the water is draining). If the asset’s role in society is shrinking, not growing — losing developers, losing mindshare, losing market position to competitors cycle over cycle — the bowl depth is meaningless. A deep bowl only matters if the water is rising. An asset in fundamental decline has a deep bowl because people are leaving, not because there’s a temporary panic. No amount of upside math overcomes a structural decline. Remove it from the scorecard entirely.

Selling: The Thirds Strategy

The old rule, “sell everything at the resistance line,” is too extreme. It creates a binary outcome that guarantees regret: if the price keeps going up, you hate yourself for selling everything. If it crashes right after, you hate yourself for not selling sooner. The Thirds Strategy eliminates both regrets.

The Core Rule: Always sell 1/3 of the CURRENT position. Never 1/3 of the original position.

This distinction is critical. Selling 1/3 of current creates an asymptotic curve: you mathematically never reach zero. As long as the conviction holds, there’s always skin in the game.

At the Resistance Line (1-2 year high):
Sell 1/3 of the current position. Now holding 2/3 of original.

• If it drops: Happy. Took profit at the top. Smart move.
• If it keeps rising: Happy. 2/3 is still making money. Still in the game.

At the same upside above the resistance line:
Sell another 1/3 of current (which is 1/3 of the 2/3 still held = 2/9 of original). Now holding 4/9 (~44%) of original.

“Same upside above” means: if resistance was $100 and the buy was at $50 ($50 of upside), the second third sells at $150.

If it keeps running, repeat the pattern:

Each sell is smaller in absolute terms (1/3 of a shrinking base), so progressively less comes off the table as the asset proves it deserves higher prices. The system rewards conviction with continued exposure.

Why “never fully exit” matters: The most painful outcome in investing isn’t a loss. It’s selling a winner and watching it 5x without me. The Thirds Strategy makes that impossible. There is always a position. The position shrinks with de-risking, but it never reaches zero. Over a lifetime of investing, this eliminates the single biggest source of investor regret.

Why this works psychologically: At every decision point, I can construct a narrative where I made the right call. That’s not self-delusion. It’s genuine emotional hedge. The worst outcome in a binary sell-all/hold-all system is catastrophic regret. The worst outcome in the Thirds system is mild suboptimality. Over a lifetime, that emotional stability compounds into better decisions on every subsequent trade.

Case Study: Tesla 2019

In 2019, Tesla stock dropped to around $14.67 per share in today’s split-adjusted terms (~$220 pre-split). The prior resistance line was ~$24 split-adjusted (~$360 pre-split), set when Tesla primarily had the Model S and X. By 2019, Tesla had the Model 3 and was delivering over 50% more vehicles year over year (367,500 vs. ~245,000 in 2018), with Model 3 deliveries alone up over 200%. The fundamentals at $220 (pre-split) were massively better than at $360. The market was offering a better company at a lower price. Textbook GUI spoon: long handle, sharp bowl, all Guardians passing.

Note: Tesla has since undergone a 5-for-1 split (August 2020) and a 3-for-1 split (August 2022), for a cumulative 15:1 adjustment. All figures below are split-adjusted to current share equivalents for clarity.

What actually happened: I entered around ~$14.67/share (split-adjusted). Tesla rose to ~$29 and I sold half. Then at ~$36, COVID struck and I exited the remaining position entirely. Total return: ~2.27x. Excellent. But Tesla then rallied above $167/share (split-adjusted), meaning the full potential was a 10x+ that was missed due to COVID panic.

What the Thirds Strategy would have prescribed:

At the COVID crash moment, I would have already banked $23,100 in realized profit. The remaining 667 shares at $23 = $15,341. Total portfolio value: $38,441 on a $22,000 investment. Up 75% even in the crash. No reason to panic.

The Conviction Decomposition would show:

No sub-beliefs actually broke. They were temporarily stressed but structurally intact. Combined with having already locked in $23,100 of profit, the emotional calculus shifts from “I need to protect my gains” to “my gains are already protected. The 667 remaining shares are my conviction bet.”

Tesla runs to $167+: 667 shares × $167 = $111,389. Plus $23,100 already banked. Total: $134,489 on a $22,000 investment. Over 6x return. Compared to the actual ~2.27x from exiting fully at $36.

My analysis was right in 2019. Every Guardian confirmed the trade. The only thing that went wrong was execution under stress. The Thirds Strategy exists precisely to make execution stress-proof.

One more thing this case study proves, which I only formalized later as the Conversion Track Record (Section V-F). On GAAP numbers alone, 2019 Tesla was borderline unbuyable: losses, cash burn, and a factory sleeping in tents. What made the conviction rational was the conversion history. The Roadster’s money became the Model S. The Model S’s money became the Model 3. Shanghai went from mud to producing cars in about ten months. Every prior round of gold spent on XP had turned into visible levels. The 2019 entry is the archetype of a Conversion Track Record buy: a proven converter, priced as if the converting had stopped.

Buying: The Chunk System

Just as selling is graduated, buying is graduated. Don’t deploy all capital at once. Reserve follow-on ammunition for deeper dips.

On entry (initial spoon identified, framework filters pass): Deploy 1 chunk.

On further dips, every 20% drop from the recent high: Deploy 1 additional chunk at each 20% level.

Reserve 3-4 total chunks as follow-on investments.

Example: If resistance is $100 and the price has dropped to $80 (20% below resistance):

• Chunk 1: Buy at $80 (bowl is forming)
• Chunk 2: Buy at $60 (dip deepens)
• Chunk 3: Buy at $40 (deep conviction territory)

Critical rule: Before deploying each subsequent chunk, re-run the Sunk Cost Firewall test and check Conviction Decomposition status. The chunks are not an obligation. They’re a conditional plan. If conviction is breaking, stop buying.

Why chunks and thirds work together: The chunk system lowers the average entry on the way down, while the thirds strategy locks in profit on the way up. Combined, they create an asymmetric payoff: average cost keeps improving (making the eventual recovery more profitable), and exits are staggered (ensuring profit regardless of where the top actually is).

V. Conviction Check

The Conviction Check is something I run on a schedule and after major events. One simple question: “Do I still believe this outcome is where the world is heading?”

I run it quarterly and after any major event (crash, regulatory change, competitive shift, earnings miss). The question isn’t “will the price recover?” It’s “ten years from now, will more people use this, or fewer?” If the answer is still “more,” hold. If it’s actually changed, reassess.

The problem with conviction is that it feels the same whether it’s justified or whether I’m anchored to a position I’ve already committed to. The more I’ve bought on the way down, the harder it is to honestly answer “do I still believe?” So the checks have to be structural, not emotional.

A. The Survival Ratchet

Core principle: Each crisis an asset survives at a weaker stage permanently raises the conviction floor at its current stronger stage.

If an asset couldn’t be killed when it was fragile, it almost certainly can’t be killed now that it’s stronger. Survival compounds. Every existential threat that fails to destroy the asset is evidence that the next threat will also fail, because the asset now has more infrastructure, more users, more institutional support, and more regulatory clarity than it did when it survived the last crisis.

Maintain a Survival Ledger for each asset:

The inverse, Survival Discount: If an asset has never been seriously stress-tested (only existed during a bull market, or too new to have faced a real crisis), conviction should be discounted. Untested survival is not the same as proven survival.

Example: Bitcoin’s Survival Ratchet

Conclusion: BTC survived existential-grade crises when it had almost nothing. At its current stage, with spot ETFs, sovereign interest, institutional custody infrastructure, and regulatory frameworks, it would take something categorically worse than anything it’s faced to kill it. The “goes to zero” scenario has been empirically stress-tested and failed repeatedly, at weaker stages.

B. Conviction Decomposition

Don’t let conviction be a single number. Break it into specific, falsifiable sub-beliefs, each one independently verifiable, each one checked without reference to price. Price is the one thing you can’t use as a signal, because price dropping is the whole reason you’re buying.

For each asset, write down 3-5 sub-beliefs before buying:

Rules:
– Every sub-belief must have a concrete, observable break condition.
– Break conditions must be checkable without looking at price. Price dropping is not a break condition; the reason for the drop might be.
– If one sub-belief breaks, reduce position sizing and re-evaluate.
– If two or more break simultaneously, stop buying entirely.
– Review on a fixed schedule (monthly or quarterly), not reactively after price drops.

Example: Bitcoin Conviction Decomposition

Current conviction: All 5 intact → 82%. If “institutional adoption” breaks (sustained ETF outflows), conviction drops to ~65%. If both “institutional adoption” and “regulatory environment” break simultaneously, conviction drops below the 55% threshold. Stop buying.

C. The Sunk Cost Firewall

The Survival Ratchet and Conviction Decomposition are analytical tools. But analysis alone isn’t enough; there’s also a psychological safeguard needed against the most dangerous bias — what gamification calls the Sunk Cost Prison: maintaining conviction because money is already invested, not because the evidence supports it.

Here’s the test: “If I had zero position right now and saw this asset at this price with these fundamentals, would I buy it today?”

If the honest answer is “no,” conviction is being propped up by sunk cost, not analysis. That’s the signal to stop buying and possibly reduce.

When to run this test:
– Every time you’re about to make a new dip-buy
– After any sub-belief in the Conviction Decomposition changes status
– After any drawdown exceeds 50% from your average entry price

D. The Kill Switch

The Thirds Strategy means I mathematically never fully exit a position. That’s a feature 95% of the time. But there are rare scenarios where full exit is correct.

The Kill Switch fires when a sub-belief is not just weakened but destroyed:

• Fraud or structural dishonesty discovered. (FTX, Enron, Wirecard.) If the fundamentals were fabricated, every Guardian assessment is invalid. Exit completely.
• Regulatory destruction with enforcement. Not “proposed regulation” (that’s a spoon catalyst). Actual, enforced prohibition that eliminates the asset’s use case.
• Thesis-destroying technological obsolescence. A competitor doesn’t just “absorb the use case” (Guardian 1). The entire category becomes irrelevant.
• Conviction Decomposition shows majority of sub-beliefs destroyed (not stressed; destroyed). If 3 of 5 sub-beliefs have their break conditions triggered with no path to recovery, the thesis is dead.

What the Kill Switch is NOT for: Market crashes. Temporary regulatory uncertainty. A competitor gaining ground. Fear or discomfort.

The Kill Switch is the emergency brake. The asymptotic rule assumes conviction, however diminished, is still above zero. When conviction reaches genuine zero, the math changes.

E. Maximum Concentration Rule

Even with perfect conviction, you are still human and could be wrong.

Hard rule: No single asset should exceed 35% of total portfolio via the chunk system, regardless of conviction level. Even at 90% conviction, there is a 10% chance of being wrong. A portfolio that’s 70% in one asset has catastrophic downside if that 10% materializes.

F. The Conversion Track Record (Spending Gold on XP)

Some of the most interesting assets in any cycle are the heavy reinvestors: companies pouring everything into R&D and capacity while their reported earnings look terrible. In RPG terms, they are spending gold on XP. The character sheet looks poor while leveling, and anyone judging by gold count alone will call the character weak at exactly the moment it is growing fastest.

The Conversion Track Record is a conviction modifier for these assets. It asks one question with two halves: did past level-ups land, and does the current one have a visible progress bar?

Did past level-ups land? Look for demonstrated conversion: specific instances where this company’s past growth spend became shipped products, better margins, or durable market position. A long conversion record raises conviction the same way the Survival Ratchet does. A record of spend that quietly evaporated lowers it. And absence of any record means the Survival Discount applies: unproven conversion is worth very little, no matter how confident the narrative sounds.

Does this level-up have a progress bar? This is where the Decomposition rules from Section V-B take over, and they are non-negotiable here. Every “we are investing in the future” claim must decompose into falsifiable milestones: named deliverables, dates, unit economics, all checkable without reference to price. If a growth-spend story cannot produce a progress bar, it has produced an unfalsifiable excuse, and unfalsifiable excuses are precisely what the Conviction Decomposition exists to kill. “Investing in the future” with no break condition is how investors ride a broken thesis to zero while feeling patient.

Applied honestly to the ecosystem everyone argues about (status as of August 2026):

• Tesla, pre-2020: elite. Roadster money became the Model S, Model S money became the Model 3, Shanghai went up in about ten months, and the energy storage business emerged from the same machine. This is the record that justified the 2019 case study above.
• Tesla, since: mixed. The Cybertruck converted into a niche product (20,237 U.S. sales in 2025, down 48% from 2024). The Dojo supercomputer program was wound down in August 2025, then partially revived as Dojo3 in early 2026. Solar mostly never converted. Meanwhile robotaxi shows an actual progress bar: roughly 2.4 million cumulative paid miles by mid-2026, though quarterly growth has flattened. A mixed recent record demands smaller conviction than the pre-2020 record earned.
• SpaceX: cleaner and more recent. Booster reuse converted into launch-cost dominance. Starlink converted from a capex sink into an $11.4 billion revenue engine in fiscal 2025. Starship remains pre-conversion (6 successes in 11 integrated test flights through late 2025), so it earns milestone-tracking, and no credit yet.
• xAI: unproven. A $20 billion Series E in January 2026 at a roughly $230 billion valuation is gold entering the account. Until spend visibly converts into product and revenue milestones, the track record is empty and conviction should treat it as a capital sink with a story attached.

Two guardrails keep this modifier from becoming a loophole:

Guardrail 1: Gross margin is the tell. Growth spend lives below the gross-profit line. Whatever a company claims about the future, the business it already delivers shows up in gross margin today. Deteriorating gross margin in the converted business is a Guardian 4 failure that no investment narrative can excuse.

Guardrail 2: The price must not already credit the conversion. Every great reinvestment-phase buy in the historical record shares one feature: the market had stopped believing. Meta near $88 at a single-digit forward multiple. Amazon after the dot-com crash. Tesla at a split-adjusted ~$14 in 2019. Buying a proven converter at a moment of emotional disbelief is the entire trade. When an asset trades at 250-300x trailing earnings, the market is pre-paying for conversions that have yet to ship, and the asymmetry GUI depends on is gone.

Notice this is the spoon logic restated. The spoon works because emotional loss of faith temporarily disconnects price from intact fundamentals. The Conversion Track Record extends the same logic to reinvestment-phase companies by redefining what “fundamentals” means for them: conversion milestones and unit economics, instead of the GAAP bottom line that accounting rules have already distorted. The bowl you want is the market losing faith in a converter whose progress bars are still filling.

A ledger template for tracking this lives in Section IX.

VI. Party Composition: Asset Role Classification

GUI borrows from RPG party design to classify assets by the function they serve in a portfolio under stress — their actual job during a crash, rather than conventional risk labels or asset class categories. In every RPG, a party of all DPS wipes on the first boss. Portfolio construction follows the same rule: role diversity is what lets you survive different encounter types (market conditions) without panic-selling.

Most retail investors build all-DPS portfolios (high-growth tech, leveraged crypto, meme stocks) because DPS is exciting. Then they get wiped in the first real downturn. I’ve been guilty of this myself. The role system makes the error visible before it becomes expensive.

The Core Roles

🛡️ TANK — Absorbs Damage So the Party Survives

Portfolio function: Holds value (or loses least) during market crashes, giving emotional and financial stability to not panic-sell other positions.

Its job isn’t to make money. It’s to not lose money when everything else is bleeding. It’s the position I look at during a crash that lets me breathe.

GUI characteristics: Highest score on Guardian 1 (structural moat). Lowest correlation to risk-on selloffs. Longest Survival Ratchet history. Modest upside to resistance line, but highest conviction percentage.

Examples: Gold, Treasury bonds, assets with millennia-long or government-backed track records.

⚔️ DPS — Deals Damage, Wins the Fight

Portfolio function: Generates the outsized returns that actually grow wealth.

GUI characteristics: Highest upside to resistance line (1.5x-3x+). Strong but not impervious fundamentals. Higher volatility, larger drawdowns during crashes. Conviction typically 50-70%.

Volatility Tags: Not all DPS are equal. DPS assets carry a volatility tag that adjusts allocation sizing:

• High Volatility: Swings 50%+ in major selloffs, outperforms dramatically in rallies. Always size smaller. (SOL, leveraged sector plays, high-beta altcoins)
• Moderate Volatility: Swings 20-40% in selloffs, solid consistent growth. Standard sizing. (COIN, MU at right entry)
• Low Volatility: Swings 10-25%, steady grower. Can be sized larger. Rare for pure DPS.

Critical sizing rule: Tier measures quality within role. Volatility tag determines allocation ceiling within tier. An S-Tier DPS (High Vol) like SOL gets a smaller allocation than an A-Tier Bruiser like BTC, because the role and volatility profile demand it. Don’t put the glass cannon on the frontline no matter how much damage it deals.

💚 HEALER — Restores the Party’s Resources Over Time

Portfolio function: Generates steady cash flow that replenishes reserves, funding chunk deployments when dip opportunities arise.

In games, you don’t notice the healer until they stop healing. Then everyone dies. Same in a portfolio: yield-generating assets feel boring until a crash creates chunk-buying opportunities and you need cash to deploy.

The DOT Test: Can Your Healer Out-Heal the Damage?

In every raid, bosses apply DOT (damage over time): a constant drain on party HP. In investing, inflation is the DOT. A “healer” yielding 3% while inflation runs at 4% isn’t healing. It’s just reducing the rate at which you’re dying.

GUI evaluates healers on real yield (nominal yield minus inflation), not nominal yield alone.

True Healers (real yield positive): Businesses with pricing power that grow dividends faster than inflation. Staking yields on appreciating assets. Real estate with inflation-linked rent escalation. TIPS / I-Bonds.

Regen Buffs (real yield near zero): Standard dividend stocks during high-inflation periods. Money market funds. Better than holding pure cash, but treading water.

Fake Healers (real yield sharply negative): Token yields funded by inflation/dilution. High-yield bonds from deteriorating companies. DeFi protocols paying 15% APY while the token drops 40%. Looks like healing, actually dealing damage.

Hybrid Classes

The best assets don’t fit neatly into one role.

🗡️ BRUISER (Tank + DPS): Genuine upside AND crash resilience. High upside combined with deep Survival Ratchet history and strong structural moat.

Example: BTC. 1.94x upside (DPS-level), but longest survival ratchet of any crypto asset (tank-quality resilience). Not pure tank (still drops 49% in crashes) and not pure DPS (not as volatile as altcoins), but serves both functions.

🏹 SKIRMISHER (DPS + Healer): High upside that also generates yield while you hold it, funding ongoing chunk deployments.

Example: ETH with staking. 2.29x upside (DPS) plus ~2-4% staking yield (healer), generating income you can redeploy into chunks during further dips.

🛡️💚 BULWARK (Tank + Healer): Maximum defense plus resource generation. The position you never worry about.

Example: Treasury bonds / I-bonds. Government backing (ultimate tank) plus interest payments (healer).

The Unicorn: Tank + DPS + Healer? In games, this is the broken meta pick that eventually gets nerfed. In investing, it would be an asset with crash resilience, high upside, AND yield generation. No clean real-world example exists consistently. If someone claims an asset does all three, they’re probably wrong about at least one.

Elemental Diversity: The Correlation Check

Having three DPS is fine. Having three DPS that all drop together in the same scenario is three copies of the same position with different tickers. Three fire mages all fail against a fire-resistant boss.

Within each role, ensure assets have differentiated risk factors. Ask: “Is there a single event that would hit all DPS positions simultaneously?” If yes, they’re the same element.

Poor diversity: SOL + COIN + ETH (all crypto; a regulatory crackdown wipes all DPS simultaneously).

Good diversity: SOL (crypto) + MU (semiconductors) + growth biotech (healthcare). Different sectors, different risk factors.

Raid Bosses: Thinking About Crashes

Market crashes aren’t things that happen to you. They’re encounters your party was built to handle. Different crashes are different boss types:

• Fed Rate Hike / Hawkish Policy = Physical damage boss. Hits interest-rate sensitive assets hardest.
• Liquidity Crisis / Margin Cascade = AoE (area of effect) damage. Hits everything simultaneously. Tests overall party composition.
• Sector-Specific Crash (AI bubble pop, crypto winter) = Single-target nuke. Tests role diversification across sectors.
• Black Swan (pandemic, war, financial system failure) = Enrage timer. Extended pressure. Healers become critical for sustaining through a long fight.

The question before every “raid” isn’t “will the portfolio go down?” It will. The question is: “Is this party composition built so that no single boss mechanic can wipe the whole group?”

That’s what portfolio diversification across roles actually means. Not “own different tickers” but “make sure you have real tanks, not just five DPS wearing tank cosmetics.”

VII. Current Asset Tier List (February 2026)

Based on GUI analysis conducted during the February 2026 multi-asset selloff:

Tier list reflects a snapshot from February 2026. Prices and tiers change as markets move. I’ll update this periodically.

Party composition check: S-tier and A-tier picks include a Bruiser (BTC), a High-Vol DPS (SOL, sized small), a Skirmisher (ETH), a Tank (Gold), and a Moderate-Vol DPS (COIN). Crash resilience from the Tank and Bruiser. Burst upside from the High-Vol DPS (sized appropriately). Sustained damage plus yield from the Skirmisher. Reliable growth from the Moderate-Vol DPS. No single crash type wipes the whole group.

Note on TSLA: Tesla is a fascinating company that GUI cannot currently classify because it fails the framework’s most important filter. The $490 resistance line was set during a period of declining revenue, declining EPS, and compressing margins: the opposite of Guardian 4’s requirement. The handle from $214 to $490 was built primarily on post-election political sentiment and robotaxi/Optimus narrative, not improving fundamentals, making the shape closer to a hatchet than a spoon. If Tesla’s robotaxi revenue materializes and fundamentals begin improving while the stock trades well below its high, it could re-enter the framework as a DPS or even Bruiser. But buying on narrative at 340-370x trailing earnings while the core business contracts is precisely the kind of trade GUI is designed to reject.

August 2026 Status Check

Six months after the February snapshot, here is the scoreboard (prices as of early August 2026, rounded):

Updated note on TSLA (August 2026): The story has evolved in a way that makes Tesla the live test case for the Conversion Track Record (Section V-F). Q2 2026 revenue hit $28.2 billion, up 26% year over year, so the February critique of declining revenue no longer holds. But operating income fell 57%, adjusted EPS came in at $0.33, and free cash flow went negative under heavy AI and robotics capex. That is textbook gold-on-XP spending. The V-F questions now apply directly: the pre-2020 conversion record is elite, the recent record is mixed (Cybertruck down 48% in 2025, Dojo wound down then revived as Dojo3, robotaxi at ~2.4 million cumulative paid miles with flattening quarterly growth). And Guardrail 2 still bites hardest: at roughly 250-300x trailing earnings, the market is pre-paying for conversions that have yet to ship. A proven-converter thesis needs a disbelief price, and $322 with a ~260x multiple is the opposite of disbelief. Tesla remains outside the framework as a buy, but it is now the single best asset for practicing the V-F analysis in real time.

August 25, 2026 Status Check

Three weeks after the early-August scoreboard, here is the tape. Crypto prints are from CoinGecko on August 25, 2026. U.S. stocks are the August 24 close. Gold and silver are Kitco spot.

Updated note on TSLA (August 25, 2026): Q2 2026 is confirmed in Tesla’s Exhibit 99.1: revenue $28.2 billion, up 26%, operating income down 57%, adjusted EPS $0.33, free cash flow negative $1.1 billion. Yahoo’s trailing P/E at the August 24 close of $348.95 is 332.33. Monday’s drop was a nameable Catalyst Test (China recall of about 2.98 million vehicles, plus Canada auto-tariff tape), not a Kill Switch. Conversion Track Record is unchanged: still no disbelief price. Cybertruck 20,237 U.S. sales in 2025 is a Cox/KBB estimate, not Tesla’s own model breakout. Robotaxi remains a chart at about 2.4-2.5 million cumulative paid miles at Q2, unlabeled in Tesla’s text. Tesla remains outside the framework as a buy.

VIII. Intellectual Lineage: What Exists and What GUI Adds

GUI sits at the intersection of several established investing methodologies. The individual components have parallels in existing frameworks. The integration, and especially the behavioral design layer, is the novel contribution.

Closest Relatives

Kelly Criterion / Fractional Kelly. The Expected Value formula (Upside × Conviction) and the principle that “3x at 70% beats 30x at 20%” is fractional Kelly reasoning: size bets proportional to edge, not excitement.

GUI’s addition: Kelly is applied to individual trades with known probabilities. GUI wraps Kelly-like logic into a macro thesis about secular trends and mean reversion.

Value Averaging (Michael Edleson, 1988). The Chunk system parallels Value Averaging: invest more when the asset underperforms its target path, less when it outperforms.

GUI’s addition: Value Averaging is formula-driven and fundamentals-agnostic. GUI layers in a conviction gate: you only keep chunking if Conviction Decomposition sub-beliefs still hold. The spoon/plate/hatchet filter prevents you from value-averaging into a broken asset.

Mean Reversion Strategies. The core thesis (“if it was worth $X before, and fundamentals are the same or better, it should return to $X”) is a form of mean reversion.

GUI’s addition: The spoon vs. plate distinction is a more sophisticated filter. Rather than asking “is this statistically oversold?”, GUI asks “was this dip emotional (spoon) or rational (plate)?” A qualitative judgment about the type of selling, not just the magnitude.

Stan Weinstein’s Stage Analysis. The “secular uptrend with higher cycle peaks” Guardian aligns with Weinstein’s four stages: Accumulation → Advancing → Distribution → Declining.

GUI’s addition: Weinstein is purely technical (price + volume + moving averages). GUI adds fundamental conviction scoring, Survival Ratchets as historical evidence, and explicit behavioral safeguards.

The Component Comparison

The White Space: Designing for Psychology, Not Against It

Every framework listed above treats human emotion as the enemy to be overcome. Kelly says “bet this fraction” and assumes you will. Value Averaging says “deploy this amount” and assumes you will. Stage Analysis says “sell in Stage 3” and assumes you will.

Brilliant on paper. They all fail at the same point: 3 AM, staring at a 40% loss, brain screaming to do the wrong thing.

GUI is built on a different premise: What if the execution system was designed so that at 3 AM, the rational action and the action that maximizes emotional wellbeing are the same action?

That’s what the Thirds Strategy does. That’s what the Conviction Decomposition does (it gives the panicking brain specific, observable checkpoints to evaluate instead of spiraling into fear). That’s what the Sunk Cost Firewall does (it breaks the emotional chain between past losses and future decisions).

These aren’t add-ons to a trading strategy. They are the core innovation. Guardians and math are necessary but not sufficient. What makes the system actually executable by real humans under real stress is the behavioral architecture.

The same design philosophy drives the Octalysis Framework for product design: instead of telling users “be more engaged” or telling investors “be more disciplined,” design the system so that the desired behavior is also the path of least emotional resistance.

Academic Foundations

GUI’s behavioral architecture draws on established research in behavioral economics and psychology:

• Kahneman & Tversky’s Prospect Theory (1979): Loss aversion, the finding that losses feel roughly twice as painful as equivalent gains feel pleasurable, explains why investors panic-sell in drawdowns and why the Thirds Strategy’s guaranteed partial profit-taking is psychologically powerful. After that first sell, the investor is always playing with “house money.”
• Shefrin & Statman’s Disposition Effect (Shefrin & Statman, 1985): The documented tendency for investors to sell winners too early and hold losers too long. GUI’s Thirds Strategy directly counteracts this: selling is graduated (preventing premature full exit from winners), and the Conviction Decomposition forces evidence-based re-evaluation of losers rather than hope-based holding.
• Edleson’s Value Averaging (1988): The mechanical foundation for GUI’s Chunk System. Edleson demonstrated that investing more when an asset underperforms its target path generates superior risk-adjusted returns compared to dollar-cost averaging, but left the critical failure mode (value-averaging into a structurally broken asset) unaddressed.
• Self-Determination Theory (Deci & Ryan, 2000): The framework underlying the Octalysis Framework’s Core Drives. GUI applies SDT’s autonomy, competence, and relatedness needs to investing: the Chunk/Thirds system provides autonomy (clear rules, investor in control), the Conviction Decomposition provides competence (specific checkpoints proving the investor’s analysis is sound), and the Party Composition system provides relatedness (each asset has a role and purpose in the portfolio).

Why “Geometric Utility”

The name encodes the framework precisely:

Geometric: Reading the literal geometry of chart shapes (spoon, plate, hatchet) as psychological narratives about market behavior. Not pattern-matching for entry signals, but diagnostic reading of whether the previous high was rationally established or emotionally inflated.

Utility: In economics, utility goes beyond usefulness. It is subjective satisfaction. A utility function captures not just outcomes but how someone feels about outcomes. Risk-averse utility functions explain why rational agents prefer a certain $80 over a 50/50 shot at $200, even though the expected value of the gamble is higher. The Happiness-Optimized Execution System is a utility-maximizing strategy: it sacrifices some expected monetary value to maximize expected emotional satisfaction across all possible outcomes.

Investing: Anchors the framework in long-term asset allocation, not short-term trading.

The full description: Reading the geometry of markets to maximize the utility (subjective satisfaction) of investing.

IX. Assessment Templates

Asset Assessment Table

Survival Ledger Template

Conviction Decomposition Template

Conversion Track Record Ledger Template

A Few Things Worth Saying

Don’t invest money you can’t afford to lose. I mean this seriously: if losing everything you put in would change how you live, you’re putting in too much. GUI is supposed to make investing feel good. That stops working the moment the stakes become existential. Size your positions so that a total loss would be annoying, not devastating. Then play the game.

There’s a practical reason beyond psychology: GUI is a patient strategy. Depending on where you enter the cycle, your capital could be underwater for 2–4 years before the thesis plays out. During that window, you need to be buying, not selling. If a medical emergency, job loss, or any life event forces you to liquidate positions in the middle of a drawdown, you’re doing the exact opposite of what the framework prescribes — selling low when you should be buying low. That forced sale isn’t just bad timing. It permanently breaks the asymmetric payoff the chunk and thirds systems are designed to create. This is why the “money you can afford to lose” rule isn’t a disclaimer — it’s a structural requirement of the framework.

The two ways this framework fails:

1. Broken conviction you don’t recognize. My thesis is wrong and I don’t realize it. I keep buying the dip because the conviction feels intact, but the world has actually moved on. That’s why the Conviction Check exists. It’s the most important part of this entire framework. I built the Sunk Cost Firewall and Conviction Decomposition specifically because I know how easy it is to fool myself.

2. Forced liquidity during a drawdown. GUI requires patience — potentially 2–4 years of holding through a cycle before the thesis plays out. If life forces you to sell during that window (because you invested money you actually needed), you crystallize the loss at exactly the wrong moment. The Conviction Check can’t help you if the problem isn’t conviction — it’s cash flow. This is the second fatal flaw, and the only defense is position sizing: never invest more than you can afford to have locked up for years.

On taxes: The Thirds Strategy creates taxable events at each sell. I don’t try to optimize for taxes inside the framework; that’s a separate problem for a CPA. Mixing tax optimization into conviction decisions adds complexity and distorts the system.

On whether this is “financial advice”: This is my personal framework. It’s how I think about my own money. I’m sharing it because I think the ideas are useful, especially the behavioral design angle. Your situation is different from mine. Use what’s helpful, ignore what isn’t, and talk to a financial advisor if you want personalized guidance. I’m a behavioral designer, not a financial advisor.

Related Reading

• The Octalysis Framework — The behavioral design framework that inspired GUI’s approach to investor psychology.
• Top 10 Finance Gamification Examples — How the finance industry uses behavioral design to change money habits.
• Self-Determination Theory: The Complete Guide — The academic foundation behind GUI’s autonomy, competence, and relatedness design.
• Anti Core Drives — The hidden forces that kill engagement — and investor discipline.
• Economy Design Framework — Tokenomics and economic system design through a behavioral lens.
• Game Technique #50: Sunk Cost Prison — The psychological trap GUI’s Sunk Cost Firewall is designed to defeat.

WOULD YOU LIKE YU-KAI CHOU TO WORK WITH YOUR ORGANIZATION?