A spike is a rapid rise in implied probability; a crash is the mirror move on the downside. Headlines love both because they look like drama on a chart. For you, they are classification problems: did the market receive new information, did thin liquidity exaggerate a small trade, did the crowd overreact, or did a false rumor briefly hijack the tape?
Neither a spike nor a crash is automatically a buy signal. The useful question is what kind of move you are watching, and whether your own forecast should change before you touch the order ticket. The journaling chapter at the end of the prior module exists precisely so you can compare tape timestamps to what you believed before the wire hit.
What you are actually seeing
On a standard binary contract, price approximates the market’s current belief that the event happens. When YES jumps from 34¢ to 52¢ in under a minute, traders are repricing the chance of YES upward—often after news, sometimes without it. The move is a posterior jump, not a stock-style percent return on a fungible equity.
The first minutes after a headline usually show the sharpest repricing. Fast traders and market makers react to wires and official releases; cross-venue arbitrageurs close gaps when the same event trades in more than one place; retail apps and slower accounts extend or fade the move over the next hour. On a central limit order book, aggressive buying walks the ask and prints worse average prices than the mid you saw on television. On automated market maker pools, a modest notional swap can move the curve sharply, then half-revert when the next trade is smaller.
That microstructure detail matters because your fill is the trade, not the mid. Order-book chapters explain walk-the-book mechanics; here the takeaway is timing: the person who lifted the spike paid for urgency.
Four buckets: information, liquidity, overreaction, hoax
Information moves deserve a real update to your probability. Sister contracts and comparable venues should move in a coherent direction; volume often rises above recent baseline; credible sources (official data, court filings, wire services) align with the contract’s resolution rules.
Liquidity moves happen when the book is thin. A few thousand dollars can shift mid several cents without changing the underlying world. Volume may look elevated relative to a dead week but still be tiny in dollar terms, and other venues may not budge.
Overreaction is when the market’s jump exceeds your disciplined Bayesian update. You might believe YES is more likely after news, but not by eighteen cents. Contrarian strategies live here—only when expected value after spread and fees is still positive.
Hoaxes and false rumors often show isolated venue moves, weak volume, quick reversion, and silence from primary sources. Chasing a crash driven by an unverified post is how journals fill with regret.
Separating these buckets uses volume, spread behavior, cross-venue checks, and the resolution text—not vibes. When in doubt, wait one tightening spread print before sizing.
A real spike: indictment headline
Imagine YES on “Candidate X wins the party nomination” trading near 34¢. Your research-based probability is about 38%. At 10:02 ET an Associated Press headline hits; within forty seconds YES lifts toward 52¢.
You check the source tier (wire, not anonymous social), twenty-four-hour volume (several times a normal day), and a second regulated venue (similar move within a couple of minutes). A related contract—general election YES—ticks up modestly in a direction that makes narrative sense. After updating your belief you might land near 46%, still below the 52¢ ask. Spread widens from a tight two cents to six: makers are uncertain. The disciplined play is not a market buy at the top; it is a patient limit bid, a decision to wait, and a journal note with the spike timestamp compared to your pre-news row.
If you are flat, the spike is information you may have missed—not an invitation to pay the ask. If you are already long, the spike may be liquidity to sell into with limits, not proof the event is nearly certain.
A false crash: rumor and reversion
Now YES sits near 71¢ and you hold a modest long from last week. An unverified account claims withdrawal; YES falls to 58¢ in ninety seconds. Volume is only slightly above baseline. On another venue the same contract barely moves. Eight minutes later the campaign denies the story and price drifts back toward 69¢.
The lesson is procedural: thesis-based invalidation should reference official resolution inputs, not every social tick. If you sold into the hoax, tag the journal entry for recency bias review—the market gave you a liquidity event, not a new posterior. Stop-loss chapters distinguish price stops from thesis stops; near resolution, thesis stops usually win.
When the move keeps going versus snaps back
Drift often follows confirmed information: official confirmation, polls moving the same way, sustained volume over days, and arbitrage gaps that stay closed. Momentum-style approaches assume drift when the information layer is thick and cross-venue agreement persists.
Reversion is common when the headline is walked back, sister markets undo the move, volume dies while spread tightens, or the first wave was narrative without capital behind it. Contrarian approaches need a positive edge at the bid you can hit, not annoyance that the crowd panicked.
Information cascades—people trading because others traded—often produce thirty-to-sixty-minute reverts when second-wave readers disagree. The spike is the opening act; your edge, if any, is in the second act.
Spikes and your edge math
A move is tradable only if your updated probability minus executable cost (ask if buying YES, plus fees) is positive. Buying at 61¢ when your posterior is 58% is negative expected value even if you are “right” by election day. Post-news widespread mid is especially misleading.
Compare the market’s delta to your delta. If you would move five points on the evidence and the market moved eighteen, overreaction is plausible. If you would move fifteen and the market moved six, you are behind the news—research before hero trades. Zero personal update with a twelve-cent jump means you missed information; standing aside is valid.
Platform texture
Regulated order-book venues tend to show classic walk-the-book spikes on macro and election days. Global pool-and-book hybrids can gap on-chain before the limit book catches up. Capped retail markets may lag or freeze extremes. Play-money sites can spike on memes with little capital at risk—useful for narrative labelling, weak for bankroll decisions.
Choosing a platform chapter themes apply: news-day analysis on thin venues is often liquidity archaeology, not forecasting science.
What to log and review
When you trade around news, record time of first print, source tier, whether the move was already drifting into the release, cross-venue agreement, your pre- and post-news probability, spread at decision, and whether you chased, faded, or stood aside.
Weekly, list large moves in watched markets, match journal trades within fifteen minutes, tag behavior, and compute profit and loss conditional on tag. One rule tweak per month—such as no market orders in the first five minutes—beats repeating the same chase.
Common misreads
“Big spike equals sure thing” confuses urgency tax with edge. “Always fade” ignores real drift on confirmed news. “Mid was fifty cents” hides that you paid fifty-four. “One venue moved” may be manipulation or local listing risk. “I will journal later” loses the pre-news probability anchor.
Reading the chart as a time series
A spike is one point; the path is the story. Was price drifting into the release—suggesting leaked positioning? Did the move hold the close of the hour? Did sister markets confirm overnight? Charts without volume and venue context are Rorschach tests.
Core concepts to remember
Classify before clicking. Executable price beats mid. Cross-venue and sister contracts separate information from noise. Drift and revert are different regimes. Journal timestamps beat memory.
Overnight gaps and stale books
A spike at the open may be catch-up to overseas news while you slept, not fresh information at your breakfast. Check timestamp of the underlying fact before classifying. A crash on thin Sunday night liquidity may mean nothing Monday when makers return.
Opening this module
Reading the Signals assumes you already trade with journals, limits, and sizing discipline. Here the market talks through price paths, liquidity, clocks, and parallel venues. Spikes are the opening lesson because every other sensor—volume, open interest, spread—makes more sense once you classify violent moves calmly.
What comes next in this module
Reading the Signals continues by asking how much actually traded—volume is loud, but easy to misread when a small market has a big-looking percentage jump.
Practice classifying three historical spikes in markets you follow: label each information, liquidity, overreaction, or hoax, and note what evidence convinced you.
Next: Volume Analysis: What High Volume Really Means