Crypto

Crypto Market Structure: Break of Structure (BOS) and Change of Character (CHoCH)

12 min read · Updated 9 Nov 2026

Before any order block, gap or sweep, Smart Money Concepts ask which way the structure points. A trend is a run of higher highs and higher lows (or lower highs and lower lows), and it changes when price closes through a swing it was not supposed to break. SMC calls such a close a BOS (break of structure) when the trend continues and a CHoCH (change of character) when it may be turning.

The crypto Market Structure page lists every such break on liquid USDT pairs over the last five days. On crypto the first thing it shows is how closely the coins move together: on 30 October 2026 Bitcoin and most large coins broke on the same day. This guide explains the rule, reads that day, follows a CHoCH that failed the next day, and compares the daily trend with the weekly.

Swings, trend and the two kinds of break

Diagram: BTCUSDT daily candles from 10 October to 8 November 2026 with the 79,220.74 swing low of 24 October, the close of 76,918.91 on 30 October that broke it, a bearish BOS on 2.53 times volume, and the order block the break left from 80,397 to 81,823; a side panel lists the breaks of the same day: bearish BOS on BTC, ETH, BNB, LINK, SHIB and LTC, bearish CHoCH on SOL, XRP and INJ
One close under a swing low, and the same day on most of the market.

A swing high is a daily candle whose high is above the highs of the five days before it and not below the five days after; a swing low the same way with lows. It is known only five days later. Then:

  • Bullish break - a daily close above the newest swing high not yet broken.
  • Bearish break - a close below the newest unbroken swing low.
  • BOS - the break goes the same way as the trend: the trend is continuing.
  • CHoCH - the first break against the trend. From that close the page counts the trend as turned.
  • Trend - the direction of the last break.

Three details matter on crypto:

  • A close, not a wick. Crypto prints long wicks, often from liquidations in the futures market. A wick through a swing that closes back inside is a sweep, not a break.
  • The close is 23:59 IST. The daily candle ends at midnight India time, not at midnight UTC, so a break here can differ by a few hours from one on a UTC chart.
  • Each swing breaks once. After a close through it the level is used; the next break needs the next swing.

Each row also carries the order block the break left: between the broken swing and the breaking candle, the opposite-colour candle with the most extreme wick. The chart pop-up shades it.

Real breaks: 30 October to 8 November 2026

How these examples were made: the page's newest day for visitors who are not signed in was Sunday 8 November 2026, which listed no pairs (see "Weekends" below), and past dates need a free sign-in. So the rows here come from running the page's own rules (the same backend functions) on Binance spot candles grouped into IST days, the source of Unfluke's crypto archive. The page's own numbers can differ slightly where its minute archive has a gap.

30 October: one day, most of the market

BTCUSDT closed at 76,918.91, down 4.78%, below the 79,220.74 swing low of 24 October: a bearish BOS, 2.91% beyond the level, on 2.53 times its usual volume, with swings LH · LL (both the highs and the lows already falling). The order block it left was the 28 October candle, 80,397.14 to 81,823.

The same close, the same day:

  • Bearish BOS - ETH (a close of 2,230.34 under 2,282.46, 2.85× volume), BNB (698.65 under 705.86, 5.70×), LINK (4.54×), SHIB (3.11×) and LTC (1.57×).
  • Bearish CHoCH - SOL (94.06 under 98.62, 2.77×), XRP (2.48×) and INJ (down 12.67%). All three still had higher lows; this was their first close down through structure.
  • The exceptions - PEPE was the one pair with a bullish break that week (a BOS the day before), and BCH, SUI and AAVE still had bullish trends and made no break that day.

Two readings follow. First, on a day like this the list is really one event, Bitcoin's, repeated; a pair that did not break that day was holding up better than the market. Second, look at Close vs Level: LTC closed at 55.08 against a level of 55.09, 0.02% beyond it. That is a break by the rule and hardly one by eye.

UNIUSDT, 4 and 5 November: a CHoCH that failed

On 4 November UNI closed at 7.654 (up 5%) above its 7.52 swing high of 29 October: a bullish CHoCH, 1.77% beyond the level, on 1.54 times volume. The swings still read LH · LL, so the turn was one close old. On 5 November UNI spiked to 7.82 and then closed at 6.841: Close vs Level went from +1.77% to −9.04% in a day. Everyone who bought the break was under water by the next close.

BCHUSDT, 7 November: a break through resting liquidity

On 7 November BCH closed at 257.9 below 263.30, a swing low that was also a pair of equal lows: a bearish CHoCH. The row appeared on the page only on 8 November (on 7 November BCH's volume was 0.58 times its average, under the gate), by then 6.72% beyond the level after a close of 245.6. The Equal Highs & Lows guide follows the same level from the other page.

How to use a break in a trade

Diagram: three panels. UNIUSDT's bullish CHoCH above 7.52 on 4 November 2026 that closed at 6.841 the next day; a table of daily against weekly trend on 6 November with BTC, ETH, SOL and LINK bearish on daily and bullish on weekly; a bar chart of how many of 30 pairs passed the liquidity gate each day from 25 October to 8 November, with weekend days far lower
Wait for confirmation, check the weekly trend, and remember the weekend gate.
  1. Find the break. Today for fresh breaks, CHoCH for possible turns, BOS for continuation.
  2. Start with BTC. Check Bitcoin's own row first. An alt's bullish CHoCH on a day BTC makes a bearish BOS is swimming against the market.
  3. Check the weekly trend. Switch Candles to Weekly. On 6 November BTC, ETH, SOL and LINK were bearish on daily candles and still bullish on weekly ones; daily bearish breaks there are pullbacks until the weekly trend turns too.
  4. Do not chase the break candle. Wait for a pullback to a defined area: the order block of the break, a fair value gap the move left, or the 62–79% band on the OTE page.
  5. Set the invalidation before the entry. For a bullish CHoCH, a close below the swing low the move started from. UNI's failure shows why it must be decided in advance.
  6. Size from that distance, and mind leverage. The share of capital one idea may lose, divided by the distance to the invalidation, gives the size. Keep the liquidation price well beyond it.
  7. Read on weekdays. The liquidity gate can hide a pair's row at the weekend; see the data section below.

Strategy ideas (ideas, not promises)

These are ideas to study, paper-trade and backtest, not recommendations or signals, and not a promise of any result. Every condition below can fail. Crypto trades around the clock and can move sharply at any hour; leverage and options can lose more than you expect, and past behaviour says nothing certain about the next session. Size, entries, exits and risk are your own decisions.

1. Breaks with Bitcoin

  • Setup: a BOS on an alt in the same direction as BTC's last break, with a volume ratio above 1.5.
  • Entry idea: a retest of the broken level from the other side.
  • Invalidation: a close back through the level.

2. The pair that did not break

  • Setup: a market-wide break day like 30 October; list the liquid pairs that are not on the page.
  • Idea: those held their structure while the market broke; study them as relative-strength candidates once BTC stabilises.

3. CHoCH plus a higher low

  • Setup: a bullish CHoCH whose Swings column already shows HL.
  • Reading: the lows turned before the highs broke: one more piece of evidence than a CHoCH with LL still showing, like UNI's.

4. The failed break

  • Setup: a break of the last few days whose Close vs Level has turned the other way, as UNI's did on 5 November.
  • Reading: trapped buyers. For longs, the break is suspect until price closes back beyond the level; for short-side research, it is a possible trap to study.

5. Daily breaks in the direction of the weekly trend

  • Setup: a daily BOS whose direction matches the weekly trend.
  • Reading: two timeframes agreeing; a daily break against the weekly trend gets a smaller size or a pass.

How to test these

Step back through past weekdays with the session date, note the rows that met an idea, and record what price did over the next ten days: did it pull back to the order block, or hit the invalidation first?

Risk per trade - risk a small, fixed share of capital on any one idea. Crypto never closes, so a stop is the only exit while you sleep; with leverage, keep the liquidation price well beyond the stop. When selling options, prefer defined-risk structures (spreads, iron condors) over naked positions.

Test before you trade - write the rules down exactly, test them on past data (see how to backtest an options strategy and the Option Simulator, which supports crypto), then paper-trade at least 20 occurrences before committing real money. Keep a journal of every trade, including the ones you skipped.

Daily or weekly candles

Every Smart Money page has a Candles setting: Daily (the default) or Weekly. Weekly runs the same rules on weekly candles built from the same daily bars.

  • A crypto week is Monday to Sunday in IST: the open of Monday 00:00, the week's high and low, the close of Sunday 23:59, with volume added up. Every date in a weekly row is that week's Sunday.
  • The current week is the week so far. On a Monday its candle is one day old, so a weekly row seen mid-week can change, or drop off, by Sunday's close.
  • Counts become weeks. A swing needs five weeks on each side, ATR 14 is fourteen weeks, and every look-back counts weeks.
  • What stays daily. The liquidity gate (the day's turnover and volume), the last price and change.

A weekly idea for structure: a weekly CHoCH needs a weekly close through a swing that took five weeks on each side to confirm, so it is rare and marks a much bigger turn than a daily one. Applying the page's rules to Binance weekly candles, BTCUSDT's last weekly break was a bullish CHoCH in the week to 27 September 2026, a weekly close above 82,850. Through 6 November that weekly trend was still up while the daily trend was down. Treat the weekly trend as the side you prefer and daily breaks in that direction as the timing.

Every tab, filter and column

  • Tabs - All (every break of the last five days), Today, CHoCH Bullish, CHoCH Bearish, BOS Bullish, BOS Bearish.
  • Order - newest first; on the same day a CHoCH before a BOS, then the heavier volume.
  • Columns - Symbol and LTP; Side; Break (BOS or CHoCH); Swing Broken; Broke On and Ago; Close vs Level (the latest close against the broken level, in %); Swings (HH or LH, HL or LL); Vol Ratio (the latest day's volume against its 20-day average); Chart, with the swing, the break and the order block.
  • Controls - Candles (Daily or Weekly), Universe (All, Futures, Options) and the session date.

What the numbers are built from, and what they cannot tell you

The daily candles are rolled up from Unfluke's one-minute spot archive (Binance data) into IST days, 00:00 to 23:59. A "session" on these pages is one such day; crypto has no weekends or holidays, so ten sessions are ten calendar days.

  • Which pairs. The live USDT pairs, about 33 on 8 November 2026, with stablecoins left out. The Universe box narrows them to the pairs with a perpetual future (Futures) or with options (Options) in the archive.
  • The liquidity gate. A pair is listed only when its 20-day average turnover is above $500,000 and the day's volume is above 0.8 times its 20-day average. There is no circuit rule; crypto has no price bands.
  • Weekends. Crypto volume drops on Saturdays and Sundays, and the 0.8× test is made against the average of the 20 days before, most of them weekdays. Applying the gate to Binance candles for 30 pairs: 29 passed on Friday 30 October, 26 on Thursday 5 November, but only 1 on Saturday 7 November and 1 on Sunday 8 November (BTC's volume that Sunday was 0.35 times its average). A row that disappears at the weekend has usually not failed; it is just not shown. Read these pages on a weekday close, or pick one with the session date.
  • The newest day can be unfinished. The daily bars are topped up during the day, so the newest day's high, low and volume are partial until 23:59 IST.
  • History. The scan pages read about 400 days of daily bars, so a weekly row can rest on at most about 57 weeks. Swings are five days each side (three on Equal Highs & Lows), ATR is 14 days, and zones and swings are followed for 120 days: about four months, not the six months the same 120 sessions span on NSE.
  • One exchange, spot only. A wick on one venue may not exist on another, and liquidations happen in perpetual futures, whose volume is often several times spot volume and is not counted here.
  • Visitors. Without signing in you see the newest day on daily candles; past dates and the Weekly setting need a free account.

Smart Money Concepts are a way of reading a chart, popularised by the trader known as ICT. There is no official rulebook and no published evidence that they work better than other methods; the pages apply one fixed reading so a row can be checked by hand. They describe what price did, not what it will do.

Where to go next

Open the page at Smart Money → Market Structure (switch the market to Crypto), or sign up free to open the Market Terminal. The NSE version, with the full list of columns, is explained in the Market Structure guide.

The other crypto Smart Money guides: SMC Setups, SMT Divergence, Retracement / OTE, Equal Highs & Lows. The building blocks (fair value gaps, order blocks and liquidity sweeps) are explained in the Fair Value Gaps, Order Blocks and Liquidity Sweeps guides.

More from the blog

  • Strategy Backtesting on Unfluke

    Backtest trading strategies on real Indian market data with Unfluke. Test equity, Nifty, Bank Nifty and options strategies with realistic costs and key metrics.

  • Live Market Scanners

    Explore Unfluke's real-time stock scanner for NSE markets. Scan 1,800+ stocks with 40+ technical indicators, fundamental filters, F&O data and custom rules.

  • Option Strategy Simulator

    Build and simulate multi-leg options strategies with Unfluke. Visualize payoffs, Greeks, max profit and loss, and replay historical option data from 2023.