Parabolic moves: exhaustion, halts and why both sides get hurt
A parabolic move is easy to see and hard to time. Define it measurably, separate observable exhaustion from inferred exhaustion, and size for the halts that hurt late longs and early shorts alike.
Start in plain English
A parabolic move is trader shorthand for a price run that steepens: each minute climbs faster and covers more range than the last, until the chart curves upward like a parabola. Small-cap gappers produce these more than most stocks. They attract two kinds of trader at once: late buyers who do not want to miss the move, and short sellers who expect it to end.
The core idea of this lesson: parabolic moves end in ways that are hard to time, and the end is often a halt rather than a smooth turn. That makes both sides expensive. A late long can be caught by a reversal and a pause on the way down; an early short can be caught by a halt on the way up and a reopening far above the stop. The useful work is to describe "parabolic" in measurable terms, separate what you can observe from what you infer, and size for the halt row.
This lesson uses one measurable description:
- Rate: price at least 40% above the regular-session open within 20 minutes or less.
- Range expansion: each of the last three 1-minute bars has a high-to-low range at least twice the average range of the ten bars before them.
A stock that meets both is "parabolic" for this lesson's count. The definition is a choice, not a standard.
Two paths from the same start
All prices, sizes and costs in this example are Illustrative. Ostrava Therapeutics is fictional. The plan shown is a short-side fade, the trade most often discussed around parabolic moves; short mechanics, locates and SSR are covered in Shorting small caps. The late-long mirror appears in the cost table.
The setup at 09:44 ET. Opened at $1.60. By 09:43 price had reached $2.46, the 09:43 high, +53.8% from the open in 13 minutes (the 09:43 close of $2.38 was +48.8%). The average 1-minute range from 09:31 to 09:40 was $0.07; the 09:41, 09:42 and 09:43 bars ranged $0.17, $0.19 and $0.17, each more than twice that. The stock met both parts of the definition. The 09:44 bar then closed at $2.25, below the 09:43 low of $2.29: the first close below the prior bar's low, this lesson's fade trigger.
The plan, written before 09:44. Short entry with a marketable limit order (floor $2.22), filled at $2.24 at the bid; locate obtained. Buy stop at $2.49, three cents above the $2.46 high. Target $1.74 (2R). Time stop 10:00. Planned risk $0.25 per share. Size: $50 risk ÷ $0.25 = 200 shares. Costs: $0.005 per share round trip for commissions and fees plus a $0.03 per share locate fee, $0.035 in total (Illustrative), plus any slippage.
Path AExhaustion and reversal
Price slides to $2.10 by 09:47 and keeps falling. At 09:55 it trades $1.73, and the $1.74 buy-to-cover limit fills.
Path BA halt-up continuation
09:45 reverses to $2.41; 09:46 trades $2.47, and trading pauses under LULD before the $2.49 stop can execute. The stock reopens at 09:51 at $2.78. The buy stop triggers on the reopening and fills at $2.80.
Path CSideways
Price chops between $2.18 and $2.36 for 15 minutes. At 10:00 the time stop covers at $2.21.
The review note for all three: "Plan followed; trigger as defined; risk 1R; outcome A, B or C." The decision was the same. Only the draw differed, and one draw cost more than twice the plan. The rest of this lesson covers what the archive shows after large runs, which exhaustion signs are observable, and how halts, SSR and borrow change the cost on each side.
Observable vs. inferred exhaustion
Traders describe the end of a parabolic move with words like "exhaustion", "blow-off" and "climax". Some of the signs behind those words can be measured; the conclusion cannot.
| Observable at the time | Often inferred from it |
|---|---|
| The first 1-minute close below the prior bar's low after a run | "The top is in" |
| A bar with the session's largest volume and a long upper wick | "Buyers are exhausted" |
| Range expansion: bars two or three times their earlier range | "It can't go much further" |
| Price 40% or more above the open, or far above VWAP | "Overextended, due to come back" |
| A LULD pause on the way up | "Pauses mark the top" |
Each observation on the left is consistent with a reversal and with a pause before another leg up. "Overextended" in particular has no fixed threshold: a stock 50% above its open can go on to double. Record the observations; treat the right-hand column as hypotheses.
Odds box
This box selects days by their final high, which is only known after the close. At 09:44 a trader cannot know whether $2.46 is the high or a stop on the way to $3.50. The 81.2% therefore describes how such days ended, not how a short entered at any particular point would have done; a fade entered below a later, higher high counts as a Path B in a trade log and as a "give-back" day in this box. Backtesting intraday ideas covers this kind of look-ahead.
Halts, SSR and borrow: why both sides pay
- LULD pauses. Under the LULD Plan, a stock that moves outside its price band enters a limit state and can pause for at least five minutes, then reopen in an auction. Bands are wider for lower-priced stocks, and for Tier 1 stocks and Tier 2 stocks at or below $3.00 they double between 3:35 and 4:00 p.m. ET (LULD Plan, as of 2026-10-05). Stops do not execute during a pause. The reopening price can land far beyond them in either direction. FennEdge's LULD article and Trading halts cover the rules; Halt resumptions covers the reopening.
- SSR. Under SEC Rule 201 (as of 2026-10-05), once a stock falls 10% or more below the prior close, the short-sale price test applies for the rest of that day and the next: short sales generally cannot execute at or below the national best bid. A gapper triggers it only after falling 10% below the prior close, which means giving back the whole gap first; in the Gapper HOD sample, 62 of 671 gappers (9.2%; 2026-07-01 to 2026-10-02; FennEdge archive; as of 2026-10-02) even closed below the prior close. If it does trigger, day two starts under SSR. See FennEdge's SSR article.
- Borrow and locates. Hot small caps are often hard to borrow; a locate can cost several cents per share or be unavailable. That cost is paid whatever the outcome, as Path C shows. FennEdge's Stock locate article and the HTB vs. ETB lesson in US Day Trading Fundamentals cover it.
- Late longs. A long entered after the third expanding bar has its stop below a bar that may be 15 to 20 cents tall. The same pause risk applies on the way down.
What shifts the odds
There is no FennEdge sample yet linking conditions to parabolic-fade outcomes. These are hypotheses to record, not findings:
- How far above the open at the trigger, and how fast it got there.
- How many expanding bars before the trigger, and the volume of the largest one.
- Pauses already taken on the way up, and how close price sits to the current upper band.
- Locate cost and availability, and whether SSR is in effect.
- Time of day. In the Gapper HOD study, 70.2% of final highs printed before 10:00 (n = 671; 2026-07-01 to 2026-10-02; FennEdge archive; as of 2026-10-02); the median final high among the 69 observations in the subset above came at 10:54. Timing is context, not a trigger.
- News and filings during the run (an offering announced into strength is a different start).
Cost of being wrong
Using the Ostrava Therapeutics bars (Illustrative). Short plan: 200 shares, $0.25 risk. Late-long mirror: long filled at $2.39 at 09:44:00 after the 09:43 bar, stop $2.27, $0.12 risk, 416 shares, $0.005 fees.
| Outcome | Exit | R after costs | Dollars after costs |
|---|---|---|---|
| Short, Path A: target | $1.74 | +1.86R | +$93.00 |
| Short, Path B: pause up, reopen fill | $2.80 | −2.38R | −$119.00 |
| Short, Path C: time stop | $2.21 | −0.02R | −$1.00 |
| Late long: ordinary stop fill | $2.24 | −1.29R | −$64.48 |
| Late long: pause down, reopen fill | $2.02 | −3.13R | −$156.00 |
Both pause rows are larger than the plan by a multiple. That is the sense in which "both sides get hurt": the side that is wrong at the moment of a pause pays the gap. Size for those rows; Position sizing shows how.
How many trades before this means anything
The archive subset holds only 69 observations in three months, and its interval spans almost 20 points. A personal log grows more slowly: if a trader sees one or two qualifying parabolic moves a week, 20 observations take months, and with 6 Path A results out of 20 (Illustrative counts) the 95% Wilson interval still runs from about 15% to 52%. With rare, fat-tailed setups, one pause row can outweigh many small wins, so look at the worst outcome as well as the average. How many trades before your numbers mean anything has the method.
What this lesson does not claim
- It does not claim parabolic moves reverse or continue, or that fading or chasing them is profitable or unprofitable.
- The measured give-back is conditioned on the final high and uses hindsight; it is not a trade statistic.
- The definition of "parabolic" here is one measurable choice; LULD band details are summarized, and the Plan is the authority.
- Ostrava Therapeutics, its bars, quote, locate and costs are fictional and Illustrative.
Go deeper
- FennEdge first.
- The Scanner (paid: Plus, Pro, Max; tiers as of 2026-10-05) flags rapid rises and drops, quiet surges (a sharp, high-volume rise after quiet trading), regular-session highs, and Nasdaq-listed LULD halts and resumes. It has no parabolic-specific trigger and no Level 2 depth.
- In-Play Days shows 1-minute charts of past active days, where you can measure rate and range expansion on past runs. Without Max, including as a guest, you can open each ticker's most recent active day in full; earlier days and the two-year history need Max.
- Gapper HOD Timeline shows when each gapper's high printed (guests, Free and Plus: latest completed session; earlier sessions need Pro or Max).
- The Reg SHO threshold list is public; threshold status is context for borrow, not a trade input.
- SSR on day two. A parabolic day that reverses below 90% of the prior close carries SSR into the next session (Rule 201, as of 2026-10-05).
- Borrow on day two. Locate availability and cost can change overnight.
- Related lessons: High-of-day breaks, The first pullback, Halt resumptions.
Exercises
Six exercises follow this lesson: three concept checks, one calculation (a short stopped through a pause), and two scenario questions (reading the measured give-back box, and which factors change the short plan's expectancy before the first trade). Exercises, answers and progress are available after signing in.
Primary sources
- Limit Up-Limit Down Plan. LULD Plan website.
- Nasdaq Trader. Trading halt codes.
- SEC. Key Points About Regulation SHO.
- eCFR. 17 CFR 242.200 (Regulation SHO definitions) and 17 CFR 242.201 (Rule 201, short sale price test).
- LULD Plan text: Plan to Address Extraordinary Market Volatility, Amendment 20 (band percentages and the 3:35–4:00 p.m. doubling).
Sources checked 2026-10-05. Ostrava Therapeutics and all prices, volumes, locate and fee costs, and the trade outcome split are fictional and Illustrative. Measured figures come from the FennEdge archive as stated. This lesson is education, not a recommendation to buy or sell any security. Past patterns do not indicate future results.