The Empty Block: Why Crypto's Bear Market Is Really a Data-Integrity Crisis

Hook

Last week, a risk desk I advise in Copenhagen ran its routine diligence on a mid-cap lending market — the kind of venue that still advertises double-digit stablecoin yields in a market that has no business offering them. The pipeline returned empty. Not "no signal." Not "insufficient confidence." Empty. The oracle feed had gone stale, the TVL dashboard was reading a cached snapshot from eleven days prior, and the governance forum had not logged a post since the token's price fell below its treasury runway. The analyst who owns that pipeline did the correct thing: she refused to fabricate. She wrote one line — input invalid, analysis withheld — and closed the file.

The market, of course, did the opposite. Within the same seventy-two hours, that lending market's token traded 34% higher on a single influencer thread citing "record deposits." The deposits were real. They were also a rehypothecation loop between two wallets controlled by the same deployer. Behind every transaction is a map of human greed, and the map that week was drawn on a void.

That gap — between a pipeline that fails safely and a market that prices the void — is the story of this bear market. It is not a story about price. Price is the symptom. The disease is that an extraordinary share of what passes for market information is, on inspection, empty input wearing a confident face. I have spent thirteen years watching this industry build cathedrals of analysis on foundations of sand, and I have never seen the sand shift as violently as it has over the past four quarters.

Context

Let me be precise about what I mean by a data-integrity crisis, because the phrase sounds like consultant jargon and I do not traffic in it.

Every crypto asset is priced by someone, somewhere, acting on an input. That input is supposed to be information: a price, a flow, a balance, a vote, a peg, a hash. The entire promise of public blockchains — the thing that separates them from the permissioned ledgers that came before — is that these inputs are verifiable. You do not have to trust a custodian's screenshot. You recompute the state from the chain and check it yourself. This is not a marketing claim. It is the actual engineering proposition, and it is the only reason an asset with no cash flow and no legal claim on anything can be worth more than zero.

But here is the uncomfortable arithmetic. The chain verifies state transitions. It does not verify meaning. It can tell you that wallet A sent 500,000 tokens to wallet B and that the balances updated correctly. It cannot tell you whether that transfer was a genuine purchase, a wash trade, a collateral shuffle, a governance attack in progress, or a bridge exploit's first leg. Meaning lives off-chain, in the interpretation layer, and the interpretation layer is where this bear market is being lost.

I watched this distinction get stress-tested in 2022. When TerraUSD broke, the chain worked perfectly. The state transitions were valid. The oracle reported the de-peg accurately. What failed was meaning — the shared assumption that a token's price and its reserve backing were the same thing. The mechanics executed flawlessly on the way to zero. The chain reveals what words hide, but it also hides what words assume.

A bear market does not create the interpretation gap. It exposes it. In a bull market, everyone's input is "up," and the crowd is right often enough that nobody audits the reasoning. In a bear market, the inputs diverge — some protocols bleed, some hold, some are already dead and simply have not been marked yet — and the quality of your interpretation layer becomes the difference between surviving and becoming someone else's exit liquidity. This is why I am writing about data now, in the middle of a drawdown, rather than about price targets. Survival is a data problem before it is a capital problem.

So let me walk the map, section by section, the way I would walk it for a desk that pays me to be right. I am not going to tell you which tokens to buy. I am going to show you where the inputs are empty, where they are poisoned, and where they are simply misread — because in this market, those three failure modes look identical on a chart and have completely different consequences for your capital.

Core

The Oracle Is the Market

Start with the most load-bearing input in all of DeFi: the price feed. Every lending market, every perpetual exchange, every algorithmic stablecoin, every vault strategy reads a price from somewhere and makes an irreversible decision based on it. Liquidations, in particular, are not triggered by the oracle. They are the oracle. There is no committee, no circuit breaker, no human in the loop. The feed prints, the contracts fire, the collateral is seized.

This means that for a large fraction of DeFi, the oracle is not a data source. It is the market. It is the thing that decides who lives and who dies, and it decides in milliseconds, with no appeals process. Yields are not gifts; they are risks wearing suits — and the oracle is the tailor.

I have been auditing oracle dependency since 2020, when I led a backtest of Aave v2 yield strategies and discovered something that still governs how I read every lending market: impermanent loss in volatile pairs erased roughly 40% of headline APY for retail. That finding was not about yield. It was about inputs. Retail was reading the advertised number and the realized number as if they were the same quantity. They were not. The advertised number was a snapshot of a moment; the realized number was an integral over time that included every bad print the oracle delivered.

In a bear market, this gap widens into a chasm, because the frequency of bad prints rises. Thin order books. Fragmented liquidity across chains. A reference price on one venue that diverges from the reference price on another by more than the liquidation buffer. When the buffer is thinner than the noise, the oracle is not measuring the market. It is manufacturing liquidations and calling them prices.

The tell is always the same. Look for venues where the liquidation volume clusters at round numbers. Look for feeds that update on a heartbeat rather than on a deviation threshold, so that a quiet market produces stale prints that a single trade can shove past the trigger. Look for protocols that cite "decentralized oracles" without ever telling you the quorum, the deviation bound, or the fallback behavior. If the documentation cannot answer "what happens when the feed disagrees with itself," the answer is that your collateral happens to it.

I want to be concrete, because abstraction is how people get hurt. Here is the diligence I run on any lending market before I will let a desk touch it. First, I pull the oracle's update history and compute the realized deviation against a volume-weighted composite of three independent venues. If the protocol's feed is tighter than the composite's natural noise band, it is over-fitted and will mis-liquidate in volatility. Second, I check the liquidation threshold against the 30-day realized volatility of the collateral, not the 90-day. Bear-market vol regimes are shorter and sharper. Third, I model the "cascade window" — the number of blocks between a bad print and the protocol's ability to correct it. If that window is longer than the time it takes a single actor to push the price, the protocol is a target, not a market.

None of this is exotic. It is the minimum. And I would estimate that fewer than one in five mid-cap lending markets in this cycle passes all three. The rest are pricing their depositors' collateral on inputs they cannot defend.

What On-Chain Data Actually Proves

There is a seductive idea that has metastasized through this industry: that on-chain data is true data, and off-chain data is suspect data. This is backwards in a way that costs people money every cycle.

On-chain data proves exactly one thing with certainty: that a specific state transition was included in a specific block and produced a specific new state. It proves inclusion and correctness of execution. It proves nothing about intent and nothing about economic substance. A wallet that receives 10 million dollars in stablecoins and immediately redeposits them into the protocol it borrowed from has produced a fully valid on-chain record of a fully circular transaction. The chain is honest. The meaning is a lie.

I call this the rehypothecation veil, and in this bear market it is everywhere. The mechanism is simple. A deployer controls two wallets. Wallet A deposits collateral and borrows against it. Wallet B receives the borrowed funds as "deposits" and re-lends them back into the same market, inflating TVL. The dashboard reports record deposits. The influencer thread cites record deposits. The token pumps. And none of it — not one transaction — is invalid. The veil is not a bug in the data. It is a feature of reading volume where you should be reading net position.

The defense is a technique I borrowed from traditional prime brokerage and adapted during the 2017 ICO audit that made my early reputation. In that cycle, I cross-referenced the whitepapers of fifteen token sales against global liquidity conditions and found one pre-IPO sale whose market cap exceeded any defensible estimate of real utility by roughly 300%. The number that gave it away was not the market cap itself. It was the concentration of the float — the share of the supply held by wallets that had never interacted with anything except the sale contract. Circulation was an illusion. The token was held, not used.

I run the same test now, updated. For any protocol claiming organic growth, I decompose deposits into three buckets: wallets that have interacted with more than one protocol in the past 90 days (organic), wallets that only ever touch this one protocol (captive), and wallets whose funding traces back to the protocol's own treasury or deployer cluster (circular). In a healthy bear-market protocol, organic dominates and circular is near zero. In the lending market I described in my opening, circular was 61% of deposits. The market priced it as growth. It was incest.

This is the discipline that separates analysis from narration. The chain reveals what words hide — but only if you ask the chain the right question. Volume is a question. Net position is a question. "Who funded this wallet, and what did that wallet do next" is the question. If you are only reading the number the dashboard chose to show you, you are not doing analysis. You are reading a press release with a hash attached.

The Institutional Data Divide

Here is where the bear market gets genuinely interesting, and where most retail readers are looking at the wrong screen entirely.

I built my standing in cross-border payment research on a specific thesis: that the spot Bitcoin ETFs approved in 2024 were not a product. They were a conduit. In my 2024 thesis, I cited roughly 5 billion dollars in initial inflows into the largest of them and argued that the relevant variable was not the inflow number but its correlation with Federal Reserve balance sheet dynamics. I was right about the direction and, more importantly, right about the mechanism: traditional finance had built a pipe, and the pipe had its own flow physics that had nothing to do with crypto-native sentiment.

That pipe is the most important data source in this market, and almost nobody reads it correctly. Retail reads ETF flows as a sentiment gauge — green day, good; red day, bad. This is wrong. ETF flows are a mechanical readout of allocation decisions made by intermediaries whose mandates are set months in advance. A large inflow on a red price day is not a contradiction. It is a rebalancing. A large outflow on a green price day is not capitulation. It is a model-driven trim. The pivot was not a retreat, but a recalibration — and the ETF flow data is where that recalibration becomes visible before it becomes price.

In this bear market, the institutional data and the retail data have decoupled, and the decoupling is the signal. Let me show you what I mean by walking the two layers side by side.

The retail layer is dominated by what I will call liquidity mirages: apparent depth that evaporates the moment it is tested. On-chain DEX liquidity in mid-cap pairs is now routinely 60–70% concentrated in a handful of automated market maker positions that are one volatility spike away from withdrawing. The order book is not deep. The order book is wide and shallow — a thin layer of quotes spread across a price range that no rational market maker would defend. When the mirage evaporates, the price does not fall in an orderly way. It gaps, because there was never anything there to trade against.

The institutional layer reads the opposite. Custody balances at regulated venues have been remarkably stable through this drawdown. Settlement volumes on institutional rails — the cross-border pipes I study professionally — have held. The entities with the longest time horizons are, on net, neither adding aggressively nor fleeing. They are positioning, which is a different verb entirely, and it looks like nothing on a price chart until the positioning completes.

The mistake retail makes is to read the retail layer and assume it describes the market. It does not. It describes the retail subset of the market, which in a bear market is the subset most likely to be wrong, because it is the subset most likely to be forced. Liquidity dries up before the news breaks — and in this cycle, the liquidity that is drying up is retail's, while the institutional pipes quietly widen.

If you want one actionable read from this section, it is this: stop watching the ETF flow number and start watching the ETF flow composition. Are the flows coming from the same handful of authorized participants rotating in and out, or is the participant set widening? Widening participation in a flat price environment is the single most reliable precursor to a regime change I have found in thirteen years. Rotation is noise. Broadening is signal.

Stablecoin Pegs Are Macro Instruments

I developed my crisis framework in May 2022, when TerraUSD broke and the crowd panicked while I sat down and did arithmetic. The arithmetic was simple and, at the time, contrarian: I plotted the stablecoin de-peg against the US Dollar Index and found that the failure mode correlated not with crypto-specific stress but with macro stress — specifically, with high-interest-rate regimes in which unbacked or under-collateralized instruments have no reserve to defend a peg. That observation let me predict the subsequent regulatory crackdown on unbacked assets before it arrived, and it is the framework I still use.

Here is the framework, stripped of its 2022 context and reapplied to today. A stablecoin peg is not a crypto event. It is a foreign exchange event. It is a promise to convert one liability into another at a fixed rate, and like every such promise in history, it is only as strong as the reserve behind it and the interest-rate environment it operates in. When rates are high, the opportunity cost of holding a zero-yield liability is high, and the reserve has to work harder. When the reserve cannot work — because it is algorithmic, or under-collateralized, or locked in an illiquid asset — the peg is not stable. It is latent.

In this bear market, I am watching the peg layer with more attention than I am watching price. Not because I expect a 2022-style cascade — the market has largely purged the naive algorithmic designs — but because the fiat-backed peg layer is now exposed to a risk it was not exposed to before: the risk that its reserves are held in instruments whose marking is itself a data-integrity problem.

Consider the chain of inputs. A fiat-backed stablecoin claims reserves. The reserves are attested — sometimes daily, sometimes monthly, sometimes by an auditor whose methodology is opaque. The attestation is a point-in-time snapshot, not a continuous proof. Between snapshots, the reserve composition can change, and the market has no way to know except by trusting the issuer. This is not a flaw unique to crypto; it is the same opacity that made the 2008 money-market funds fragile. But crypto promised to fix it, and in this specific case, it has not.

My read: the peg layer is stable but brittle. The brittleness is not visible in the peg itself, which will hold. It is visible in the attestation cadence — how often the reserves are verified and how narrow the verification window is. A peg is a promise, and a promise is only as good as the frequency of the audit behind it. When I evaluate a stablecoin in this market, I do not look at the peg. I look at the gap between attestations. If the gap is wide enough that a reserve could be impaired and restored between two reports, the peg is theater.

The Yield Void

Nothing in this bear market is more revealing than the persistence of yield where no yield should exist.

A yield is a payment for risk. That is the entire economics of it. In a risk-free-rate environment where the Fed pays a real return on cash, a DeFi stablecoin yield of 12% is not a gift. It is a confession that the protocol has not told you what risk it is running, because if it had told you, the yield would be priced. Yield is just risk in disguise, and in a bear market, the disguise gets thinner every quarter.

I learned this the expensive way in 2020, when my backtest showed that impermanent loss erased 40% of retail's headline APY in volatile pairs. That number was not a market inefficiency. It was the market correctly charging retail for the risk it was taking, and retail not seeing the invoice. The invoice arrives later, in the form of a position that underperforms simply holding the asset — which is the true benchmark everyone ignores.

In this bear market, the yield void has three distinct sources, and they have three distinct risk profiles. Conflating them is the most common error I see from retail readers.

The first source is real but cyclical: lending demand from leveraged longs who are paying to maintain positions. This yield is legitimate, it is priced by a market, and it compresses as leverage unwinds. It is safe in the sense that it is transparent. It is dangerous in the sense that it disappears exactly when you need it, because it is a function of bullish positioning.

The second source is real but hidden-risk: yield paid out of token emissions. The protocol is not generating revenue. It is printing a claim on its own future and calling it a return. This yield is a transfer from future holders to present holders, and in a bear market, the future holders do not arrive. The yield is real until it is not, and the moment it stops being real is the moment the token price collapses, which is also the moment the yield was denominated in.

The third source is not real at all: yield paid out of other depositors' principal, dressed up as a strategy. This is the Ponzi structure, and in a bear market, Ponzis do not slowly deflate. They snap. The snap is always preceded by the same signal: a yield that does not move when the market moves. Real yields breathe. Fake yields are flat, because they are set by the operator, not the market.

I want to make the diagnostic sharp. When I see a stablecoin yield above the risk-free rate plus a reasonable spread, I ask one question: where does the money come from, and can I verify it from the chain? If the answer is a named counterparty paying a named rate for a named reason, I can underwrite it. If the answer is "a delta-neutral strategy" with no chain-level accounting of the hedge, I cannot. And if the answer is a chart showing smooth cumulative returns with no drawdowns, I do not analyze it. I walk away. Smooth returns are the fingerprint of a closed system, and closed systems do not survive contact with a bear market.

The Governance Data Gap

The most under-read dataset in crypto is the governance forum, and in a bear market it becomes the single most informative one.

Here is why. In a bull market, governance is theater. Proposals pass, quorums are met, and nobody reads the text because everyone is making money and nobody wants to rock the boat. In a bear market, governance becomes load-bearing, because the decisions being made are existential: treasury runway, emission schedules, whether to pivot the product, whether to wind down. And in a bear market, the data gap between what governance says and what the chain shows becomes the most reliable leading indicator of which protocols will survive.

The gap has a specific anatomy. A protocol's governance forum will discuss a treasury runway in months. The chain will show the treasury balance in tokens. The two numbers are only equivalent if the token has a stable price, which in a bear market it does not. So a protocol that announces "eighteen months of runway" based on a token price from three months ago is announcing a number that is already false. The true runway is shorter, and the gap between the announced number and the true number is the protocol's denial. The chain reveals what words hide, and in a bear market, the words hide runway.

I run a specific diligence on governance in this market. I take the protocol's stated runway in months, convert it to a token-denominated quantity, and then recompute the runway using the current token price and the actual burn rate from the last 90 days of on-chain spend. The difference between the stated runway and the recomputed runway is a measure of either dishonesty or incompetence, and in this market, it does not matter which. Both kill the protocol.

There is a second governance signal that I find even more telling: voter participation in the tail. In a healthy protocol, participation is broad and shallow — many voters, low concentration. In a dying protocol, participation becomes narrow and deep — a handful of whales voting on everything, because the retail holders have stopped caring. When I see quorum being met by two or three addresses, I do not read it as efficiency. I read it as abandonment. The protocol is being governed by its creditors, and creditors govern toward liquidation, not toward growth.

The Complexity Tax: Reading Uniswap V4 Correctly

I want to talk about Uniswap V4, because it is the clearest example in this cycle of a protocol whose data has gotten ahead of its meaning, and because the bear market is the exact environment in which that gap does its damage.

V4 introduced hooks: programmable extensions that let developers attach custom logic to pools — dynamic fees, limit orders, custom oracles, whatever the developer can imagine. In principle, this turns the DEX into programmable Lego. In practice, it turns every pool into a bespoke financial instrument with its own risk surface, and it moves the complexity from the protocol layer to the interpretation layer.

Here is the data-integrity consequence, and it is severe. In V3, a pool was a pool. You could compare liquidity across pools because the pools were the same object with different parameters. In V4, two pools with identical token pairs and identical fee tiers can have completely different behavior, because their hooks differ. The data — the pool address, the TVL, the volume — looks comparable. The meaning does not. A hook that implements a dynamic fee can make a pool's quoted yield look attractive in a snapshot and punitive over time. A hook that implements a custom oracle can make a pool's price feed diverge from the market in ways that only reveal themselves during a liquidation cascade.

My technical position, stated plainly because the bear market rewards plainness: V4's hooks turn the DEX into programmable Lego, but the complexity spike will scare off 90% of developers — and the 10% who remain will build instruments whose risk is invisible to anyone reading the dashboard.

This is not a criticism of the design. It is a criticism of the data layer that has grown up around it. Dashboards aggregate V4 pools by token pair and fee tier, which is exactly the wrong aggregation, because the hook is the risk. The right aggregation is by hook behavior, and nobody has built that yet, because hook behavior is not a field in the pool contract. It is an emergent property that has to be reverse-engineered. Until the interpretation layer catches up to the protocol layer, V4 liquidity data is a mirage: it tells you how much is there without telling you what it will do to you.

For a bear-market reader, the practical implication is this. If you are providing liquidity to a V4 pool, the number that matters is not the advertised APR. It is the worst-case behavior of the hook under the volatility regime you are actually in. If you cannot articulate that behavior in a sentence, you do not understand the position, and in a bear market, positions that are not understood are positions that get liquidated.

The Distribution War: Reading the L2 Landscape

I hold a specific and, I think, under-appreciated view about the Layer 2 landscape, and the bear market has made it undeniable: the real difference between the OP Stack and the ZK Stack is not technical — it is who can convince more projects to deploy chains first.

This matters for data integrity because the L2 landscape has fragmented the market's information into a dozen incompatible ledgers, and the fragmentation is now a diligence problem, not just an engineering one. When liquidity and activity are spread across a rollup that uses optimistic proofs and another that uses validity proofs, and a third that is a fork of the first with different governance, the aggregate on-chain data is no longer a single truth. It is a collection of partial truths whose reconciliation requires assumptions the analyst has to make and defend.

Consider what happens when you try to compute a protocol's "total value locked" across the L2 ecosystem. You have to decide which chains count, how to price assets that only exist on one chain, and whether bridged assets should be double-counted. Each of those decisions is an interpretation, and each interpretation changes the number. A protocol can look like it has 2 billion dollars in TVL under one set of assumptions and 800 million under another, and both numbers can be technically correct. This is not a data problem that better tooling solves. It is a data problem that only better interpretation solves, and interpretation is a human function.

The distribution war compounds this. The stack that wins will not be the one with the better cryptography. It will be the one with the better network effect — more projects, more liquidity, more tooling, more data. And the winner-take-most dynamic means that the losing stack's data will become orphan data: real, on-chain, and increasingly irrelevant, because nobody is building on top of it. We do not predict the wave; we engineer the vessel — but a vessel on a route nobody sails is just a very expensive object.

For the bear-market reader, the implication is a filter. When you evaluate a protocol deployed on multiple L2s, ask which chain carries the real activity and which carries the echo. The echo is often the louder one, because it is the one being subsidized. The real activity is where the users are, and the users are where the other users are. Distribution, not cryptography, decides which ledger's data will still matter in two years.

The Machine-to-Machine Data Problem

My current research — the thing I am actually paid to think about in Copenhagen — is the convergence of AI agents and blockchain for micropayments. I am modeling the economic viability of autonomous agents that use zero-knowledge proofs to execute transactions without human intervention, and I have identified a market that could reach roughly two trillion dollars in machine-to-machine commerce if the latency and cost barriers are removed. I am spending this bear market on this problem because I believe the next cycle's data-integrity crisis will be born here, and I would rather understand it before it is priced.

Here is the problem, stated as cleanly as I can. An AI agent executing a payment has to decide whether to pay based on inputs it receives — a price, a service description, a proof of delivery. Those inputs are, in the current architecture, unverified claims from other agents or services. An agent cannot audit the meaning of a counterparty's claim the way a human analyst can. It can only verify inclusion: that a claim was signed by a key, that a proof verifies, that a state transition occurred. The moment machine-to-machine commerce scales, the interpretation layer — the layer where meaning lives — has to be encoded, because there is no human to interpret.

This is the frontier, and it is why I think the current bear market's data-integrity problems are a rehearsal rather than a one-off. Every issue I have walked through — stale oracles, circular deposits, misread flows, hidden yields, opaque governance — is a problem of meaning that humans currently resolve by judgment. In a machine-to-machine economy, that judgment has to become a protocol. And the protocols that solve it will be worth more than the assets they settle, because behind every transaction is a map of human greed — and the first system that can price greed without a human reading the map will eat the entire settlement layer.

I am not going to pretend this is solved. It is not. What I will say is that the researchers and protocols working on verifiable meaning — proofs of delivery, attested inputs, machine-readable counterparty reputation — are building the thing this bear market has proven we need. Everything else is a dashboard.

Contrarian

The consensus view in this bear market is that crypto is failing because the market is failing — that the drawdown is a verdict on the asset class, and the correct response is to wait for sentiment to turn. I think this is precisely backwards, and I want to be explicit about why, because the contrarian read here is the difference between catching the next cycle and being its liquidity.

The consensus treats price as the primary signal and data as secondary. In this cycle, the relationship has inverted. Price is now the lagging indicator, and data integrity is the leading one. The protocols that will survive this drawdown are not the ones with the best price action. They are the ones whose inputs are verifiable and whose meaning survives audit. The ones that will die are not the ones with the worst price action. They are the ones whose growth was circular, whose yields were flat, and whose governance was captured by creditors. You can see all three of those things in the data months before you see them in the price, and the crowd, watching price, will only notice at the snap.

The deeper contrarian point is about what a bear market is. The industry narrates drawdowns as punishment — a cleansing of excess, a return to fundamentals. This framing is comforting and useless. A bear market is not a moral event. It is a liquidity event, and its function is to force the market to re-price inputs it had been taking on faith. Every input that was assumed rather than verified gets tested. Oracles get stressed. Pegs get probed. Governance gets contested. Yields get questioned. The bear market is not destroying the data layer. It is auditing it, at scale, in public, and the protocols that pass are being handed the next cycle on a plate.

The blind spot in the consensus is the assumption that safety and leverage are opposites. They are not. What you think is safety is often leverage — a stablecoin yield that is really a token emission, an "institutional" flow that is really a single participant's rotation, a "diversified" treasury that is really one illiquid asset marked generously. The bear market's job is to convert those hidden leverages into visible losses, and it is doing that job with a thoroughness that the crowd will not appreciate until it is over. The contrarian move is not to hide. It is to run the audit first — to read the inputs the market is ignoring, and to position in the protocols that pass before the crowd realizes that passing was the whole test.

Takeaway

The next twelve months will not be decided by a Fed pivot, a regulatory ruling, or an ETF approval. Those are headlines, and headlines are lagging indicators wearing the costume of catalysts. The next twelve months will be decided by which protocols can survive an audit of their inputs — and most will not, not because they are fraudulent, but because they never asked what their own numbers meant.

So here is the question I would leave you with, and it is the one I ask every desk that retains me: if you stripped the price chart away and looked only at the data — the flows, the feeds, the governance, the yields — would you still be able to tell the survivors from the dead? If the answer is no, then you are not analyzing the market. You are watching it, and in a bear market, watching is the same as waiting to be someone else's exit. We do not predict the wave; we engineer the vessel. Build the vessel. Read the inputs. The wave will do what waves do, and it will not care whether you saw it coming.