Sunk Cost: 3 Years of XRP and the Most Expensive Feelings in Crypto

Here's the number that still hurts: $140,000 USD.

Not in a single trade. Not in a crash. Not from a scam. Bleeding out slowly, between November 2020 and late 2023, because I could not bring myself to sell a bag of XRP I'd already decided — on paper, in my notes, to my wife — was a bad hold.

I held. Because selling would have made the loss real.

This is sunk cost fallacy. And it is, honest-to-god, the most expensive emotion I've ever paid for.

I'm writing this because my colleague Sandra is about to make the same mistake with a Peloton position she bought in 2021. I'm also writing it because somebody reading this has an XRP bag (or a Luna remainder, or a BlockFi bankruptcy claim, or a bag of $THETA from the 2021 mania) that they're still holding for reasons their rational brain rejected long ago.

Here's what sunk cost is, how it trapped me for three years, and how to kill it.

---

The Economics 101 Version

Economists teach this in the first week of introductory classes:

The only thing that should influence whether you hold or sell a position today is: what's the expected return from here?

What you paid for the position is sunk. It's gone. It's a historical fact. It cannot be retrieved by any decision you make today. If the asset's forward prospects are bad, you sell. If they're good, you hold. Your entry price is irrelevant to this calculation.

Every economist agrees with this. Every rational-choice theorist agrees with this. Every behavioral-finance textbook cites this as a baseline rationality check.

And almost nobody actually does it.

Because the entry price is not just a number on a spreadsheet. It's the price you felt smart when you paid. It's the price that defines "winning" or "losing" when you sell. Your brain treats the entry price as the reference point for the entire position. Walking away from it "at a loss" is walking away from the identity of "a person who was right."

Sunk cost is not a calculation error. It's an identity defense.

---

My XRP Story (The Short Version)

I got into XRP in late 2017, at the peak of the first crypto cycle. Bought about 28,000 XRP at prices between $0.28 and $2.40. Average cost: roughly $1.08.

The thesis at the time: "Banks will use XRP for cross-border settlement. Ripple has actual enterprise partnerships. SWIFT is a broken legacy system waiting to be disrupted."

That thesis was plausible in 2017. It was questionable by 2019. It was effectively dead by 2020, when:

• Ripple's enterprise pipeline had gone mostly quiet
• Swift had actually improved and wasn't getting disrupted
• Coinbase's CEO publicly stated XRP didn't pass their decentralization criteria
• The SEC filed the securities case, sending the price from $0.55 to $0.17

By early 2021, I had done the research. I wrote in my private journal (still have the file, dated February 8, 2021):

That is a clean sell recommendation, written by my rational self, to my future self. Dated. Filed. Unambiguous.

I did not sell.

For three more years. Until late 2023. Through multiple rallies above $0.40. Through rallies above $0.50 and $0.70 and the brief $0.92 spike in 2023.

Every time the price rallied, my brain would invent a new reason to hold:

• "The SEC case is almost resolved, wait for the news."
• "I've already held this long, what's another few months?"
• "If I sell now I'm confirming I was wrong about it in 2017."
• "The price action suggests accumulation. Maybe I'm missing something."
• "Selling at $0.45 when I paid $1.08 just to lose 58% on purpose? No."

Every one of those was sunk cost fallacy dressed up as analysis. My February 2021 self had already done the analysis. It said sell. My 2022 and 2023 self found reasons to ignore it. Not because the analysis had changed. Because the pain of selling had not.

Then — in 2022 — I made it worse. I FOMO-added $22,000 to the position (documented in the FOMO article). Sunk cost said hold. FOMO said add. The combination was nuclear.

I finally sold the bulk of the position in Q4 2023. Realized loss, including the 2022 adds: approximately $140,000 USD vs cost basis, and approximately $180,000 in opportunity cost vs having held BTC in the same capital over the same period.

The craziest part: selling felt fine. The dread of "making the loss real" was 100x worse than the actual realization. My brain had been torturing me for 3 years to avoid 2 minutes of administrative pain. The pain never actually came. Just the relief of being done with it.

---

The Four Stages of Sunk Cost

Reviewing my own trajectory, every trapped position follows the same four stages. If you've held a loser longer than your thesis would justify, you've been through these:

Stage 1: Thesis Break (acknowledged silently)

You do the analysis. You conclude the trade's thesis is broken. You don't tell anyone. You don't write it down publicly. You don't change your position.

For me, this was February 2021. The analysis was done and clear. The position was untouched.

Stage 2: Rationalization Construction

You spend months inventing narratives for why holding is "actually" the right move. Each narrative is plausible in isolation. Together they're a rotating cover for the underlying sunk cost.

In 2021 I rotated through: "SEC resolution overhang," "XRP is oversold," "crypto winter will rerate everything," "institutional adoption might still happen," "at least partial thesis might hold," "hold for the tax year boundary," "macro is uncertain, not a selling environment."

Every one of those rationalizations had some surface validity. None of them engaged with the actual thesis-breakdown analysis I'd already done.

Stage 3: Defensive Position Management

You stop checking the position. You avoid looking at the P&L. You delete the asset from your main portfolio tracker so it doesn't hurt to see it.

I literally moved my XRP into a hardware wallet with no portfolio tracker integration in early 2022. "For security." That was a lie I told myself. It was so I wouldn't have to see the mark-to-market every day.

Stage 4: Resolution (eventually)

Something forces a decision. A tax event. A liquidity need. A life change. A rally that gives you a psychologically acceptable exit. Or — like me — a conversation with your partner about household finances that finally names the elephant in the room.

My exit trigger was embarrassingly external: in late 2023, my wife asked me to do a formal net-worth review for our family financial plan. I had to put numbers on paper. I had to put XRP on paper. I couldn't hide from it anymore.

I sold that week. Felt fine within 48 hours. Three years of self-torture released in three trading days.

---

Why the Brain Does This

Three reinforcing mechanisms. Each one on its own would be enough. Stacked, they're nearly inescapable.

1. Loss realization is an identity event

You are not "a person who lost money on XRP." You are "a person currently holding XRP." Those feel different. The selling event is the transition. Your brain defends against the transition because it converts "theoretical loser" into "confirmed loser."

Realization makes you face what the market already decided long ago. Loss aversion (the related bias) amplifies the pain of this acknowledgment by 2x.

2. Admission of past error

If you sell a loser, you're saying "I was wrong when I bought this." For many people, this feels worse than the financial loss. The financial loss is external (market did it). The admission of wrongness is internal (you did it).

The defense: if you don't sell, you haven't officially admitted you were wrong. You were merely "still waiting for the thesis to play out." This is a lie, but a psychologically comfortable one.

3. The "compounding commitment" effect

The longer you've held, the more committed you feel, even though the commitment has no forward value. "I've held this for 3 years" feels like a reason to hold another year. It's not — it's a reason to sell faster, because those 3 years of opportunity cost already happened and can't be recovered.

Sunk cost compounds itself. Every year you hold past the thesis break, it gets harder to sell. The position becomes a monument to your own stubbornness. Selling it feels like demolishing part of yourself.

---

The Rules That Would Have Saved Me $140k

If I could send these rules back to 2021-Dominic, they would have saved me the entire $140k plus $180k in opportunity cost. That's $320k of value, from four simple rules.

Rule 1: Thesis review on a calendar, not on price movement

The mistake: I let rallies trigger "maybe it's coming back" analysis, and dips trigger nothing. So I only ever reviewed the thesis when the psychology was biased toward holding.

The fix: quarterly thesis reviews, on a fixed calendar, regardless of price. Write one paragraph: "Is the original investment thesis still valid? Yes/no. Evidence:"

If the answer is "no" for two consecutive quarters, the position sells. Not negotiable. Not dependent on how much it hurts.

Rule 2: The "free cash" test

For every position you're holding that's down more than 20%, ask:

"If I had the current position liquidated to cash, and I could deploy it anywhere — this asset, or anywhere else — where would I put it?"

If the answer is "anywhere else," the position is a hold purely on sunk-cost. Sell it.

I asked myself this question in 2023. The answer was "I'd put this into BTC instead, there's no comparison." I was a year late to my own rule. The question is obvious; I just didn't ask it until forced to.

Rule 3: Hard time stops

No thesis is immortal. If you bought an asset with a 12-month expected thesis, and 24 months have passed without the thesis playing out, the thesis is wrong. Your "patient capital" framing is covering for sunk cost.

For me: XRP thesis was "bank adoption, 18-24 months." By month 30 it was clearly not playing out. Hard time stop: month 36. I would have sold mid-2020 with a much smaller loss. Instead I ran the clock to 72 months and bled.

Rule 4: Tell someone your decision in advance

Write your sell decision down, dated. Tell a partner, a friend, an accountability buddy. Make it costly to reverse without a new analysis.

I had told my journal. I had not told my wife. She didn't know the XRP position existed as a separate problem until the 2023 net-worth review. If I had told her in 2021 — "this is my XRP thesis, I'm exiting if it doesn't resolve by Q2 2022" — she would have asked me about it. That social accountability would have forced the decision.

---

The Questions That Trigger Real Reviews

If you suspect you have a sunk-cost position, answer these honestly:

1. "What's my thesis, in one sentence?" If you can't answer in one sentence, you don't have a thesis. You have a bag.

2. "When did I last update this thesis with new evidence?" If the answer is "when I bought," your thesis is stale.

3. "If I sold this today at market, what would I regret — losing money, or admitting I was wrong?" If the answer is the second one, it's sunk cost running the show.

4. "Would I buy this at current price if I had fresh cash?" This is the most important question — rational hold and rational buy should produce the same answer. If they don't, something emotional is in the way.

5. "What's the cost if I'm wrong AND I hold?" Most people ask "what if I sell and it rallies?" Nobody asks "what if I hold and it keeps bleeding?" Both need to be weighted.

---

The Sell Day I Wish I'd Had in 2021

If Dominic-2021 had followed his own analysis and sold that February after the thesis-break note: realized loss approximately $34,000. Painful but survivable.

Instead, Dominic-2023 sold at realized loss of ~$140,000, plus $180,000 in opportunity cost. That's 10x worse.

The difference was 30 months of sunk-cost-driven holding, during which:

• I didn't sleep well about it
• I FOMO-added because "I'm already in, may as well go bigger"
• I avoided honest portfolio conversations with my wife
• I told myself I was "patient" when I was just stuck

All of it avoidable. None of it avoided.

---

If You Recognize Yourself

You probably have a sunk-cost position. Most crypto and retail stock investors do. It might be an XRP-style stuck bag. It might be a DeFi protocol that pivoted away from its original value prop. It might be a stock whose business model clearly broke two years ago.

Here's what I'd do, in this order:

1. This week: Write down the current thesis. In one sentence. If you can't, the thesis is gone.
2. This week: Run the "would I buy today" test. Honestly.
3. Within 30 days: If thesis is gone and you wouldn't buy, sell. Don't wait for a "good exit." There is no good exit. The good exit was two years ago. The next best exit is today.
4. Next quarter: Redeploy the capital into something with an actual forward thesis — even if that's just BTC-DCA or an index fund. ANYTHING with forward alpha beats dead-thesis capital.

You will feel better within 72 hours of selling. I promise. The dread of making the loss real is always — always — worse than actually making it real.

I know. I paid $140,000 to learn that.

---

Related reading:

• Behavioral Biases Pillar — the other 11 self-sabotage patterns
• FOMO: The $140k Tax — how I got INTO the XRP adds in the first place
• Loss Aversion — why selling feels worse than it is
• Beat HODL or Don't Bother — the benchmark that exposes dead-thesis capital

---

Not financial advice. Share your own sunk-cost story or questions in the newsletter thread — every trapped bag starts looking smaller once someone names it out loud.