TD Sequential Indicator

Get the historical TD Sequential (Tom DeMark Sequential) counts of a futures trading pair on an exchange, including buy counts and sell counts. Used to track trend continuation and identify potential reversal signals.

Recent Requests
Log in to see full request history
TimeStatusUser Agent
Retrieving recent requests…
LoadingLoading…

Available on these API plans: Startup Standard Professional Enterprise

Update frequencyReal-time

Response Data

data is an array of TD counts, one record per candle; time is the candle start time in UTC milliseconds. A count may be null or omitted when it is not formed. The first four candles do not have enough comparison history. With fewer than five candles, data may be null or omitted.

📘

Valid counts and response compatibility

Valid counts are integers from 1 to 9. While the condition continues, the count stays at 9 until the direction changes or the compared closes are equal. Some responses may still contain 10–13; treat these points as unavailable and retain the raw value for investigation. Do not clamp them to 9 or interpret them as a Countdown phase. Null or missing counts are not 0.

200 · application/json · See the response example in the right panel.

FieldTypeAlways ReturnedDescription
codestringYesBusiness status code; "0" means success
msgstringNoResult message; may be omitted on success
dataarray of object or nullNoTD count array; may be null or omitted when there are fewer than five candles
data[].timeinteger (int64)YesInterval start time, timestamp in milliseconds (UTC); the indicator value is calculated from candles up to and including this interval
data[].td_buy_countinteger or nullNoTD buy count, 1–9; stays at 9 while the condition continues, until the direction changes or the compared closes are equal; null or omitted when no count is formed
data[].td_sell_countinteger or nullNoTD sell count, 1–9; stays at 9 while the condition continues, until the direction changes or the compared closes are equal; null or omitted when no count is formed

API Plans

Endpoint access and available intervals are listed below. See API Plans for pricing.

API PlanAccessAvailable intervals (interval)
Hobbyist❌ No—
Startup✅ Yes30m, 1h, 2h, 4h, 6h, 8h, 12h, 1d, 1w
Standard✅ YesAll intervals supported by this endpoint, starting at 1m
Professional✅ YesAll intervals supported by this endpoint, starting at 1m
EnterpriseAs agreed in your authorizationAs agreed in your authorization

Choose only values listed in this page’s interval parameter; an arbitrary value above the minimum is not necessarily supported. These restrictions concern sampling or calculation intervals, not update frequency or available history length.

The length of history you can query differs by API plan; if start_time is earlier than the range you can query, it is automatically changed to the earliest time available. For how far back each interval's history can be queried, see Intervals.

Use This Endpoint with AI

Pick a task and click Open in ChatGPT or Open in Claude: the AI reads this page first and then gives you the result. You can also click Copy and paste the full prompt into the AI tool you use.

Build TD 9 alerts
Have AI write a script that alerts you when the 4-hour TD count reaches 9.
Mark counts on a price chart
Have AI write code that marks the TD counts on the candle close price line.
Understand this endpoint
Have AI explain TD Sequential and the response fields in plain language.

This endpoint provides data through the REST API. Use the prompts above to have AI write the request code. For the tools available through MCP, see CoinGlass MCP.

Use Cases

ScenarioDescription
Identify Potential ReversalsWhen the buy or sell count approaches or reaches 9, the price has moved in the same direction for a relatively long time, and a reversal or pause may occur.
Track Trend ContinuationA steadily increasing count indicates that the trend is still continuing; a count that breaks off and starts over indicates that the previous rhythm has been broken.
Multi-Interval ConfirmationRequest the 1-hour, 4-hour and daily counts separately for the same trading pair; if you only need the current values, Coin TD Sequential List returns multiple intervals for all coins in one call.
Strategy BacktestingUse historical counts to backtest how rules such as “trade in the opposite direction when the count reaches 9” perform on different intervals.

Related Endpoints

EndpointHow It Differs
Coin TD Sequential ListCurrent TD counts of all coins across multiple intervals, with the sign distinguishing buy from sell
Pair RSIUses a value from 0 to 100 to identify overbought and oversold conditions
Moving Average Convergence Divergence (MACD)Gauges trend momentum and identifies golden crosses and death crosses
Pair Price OHLC HistoryPrice candles for the same trading pair; can be aligned by time for comparison

FAQ

Why do some records have only time?

No count is formed for that candle. The count fields may be null or omitted; do not replace them with 0. The first four candles lack the comparison history, and fewer than five candles may produce null or omitted data.

What is the maximum value of td_buy_count?

Valid counts are 1–9. A count stays at 9 while the condition continues and resets when the direction changes or the compared closes are equal. Treat 10–13 in an older response as unavailable, not as another stage.

What should I pass as symbol?

A trading pair, not a coin; for example, the BTC perpetual contract on Binance is BTCUSDT. Exchanges write trading pairs differently; go by what the Supported Exchanges and Pairs endpoint returns.

Why does the last record change?

The data is updated in real time. The last record corresponds to an interval that has not yet closed; its count changes with the price.

Learn More: TD Sequential

What Is TD Sequential?

This endpoint compares the current candle close with the close four candles earlier. One direction counts consecutively while the other resets; equal closes reset both. The count stays at 9 while the condition continues. The returned count fields do not represent a complete TD Sequential/Countdown algorithm or a separate trading signal.

What Does TD 9 Mean?

9 is the valid count ceiling and may repeat while the condition continues. It is not a new signal on each candle and does not guarantee a reversal.

How Do I View the TD Counts of All Coins at Once?

Use Coin TD Sequential List, which returns the current counts of all coins across multiple intervals from 1 minute to 1 week in one call; positive numbers are buy counts and negative numbers are sell counts.

Query Params
string
required
Defaults to Binance

Exchange name, e.g., Binance, OKX. For the exchanges you can query, see the Supported Exchanges and Pairs endpoint.

string
required
Defaults to BTCUSDT

Trading pair symbol, e.g., BTCUSDT. Exchanges write trading pairs differently; for the pairs you can query, see the Supported Exchanges and Pairs endpoint.

string
required
Defaults to 1d

Candlestick interval; the indicator is calculated from price candles of this interval, e.g., 1d. Options: 1m, 3m, 5m, 15m, 30m, 1h, 2h, 4h, 6h, 8h, 12h, 1d, 1w.

Plan interval limits: this endpoint is not available on Hobbyist; Startup starts at 30m, and Standard and Professional at 1m; Enterprise follows your actual authorization. Only intervals supported by this endpoint are available.

integer
1 to 1000
Defaults to 1000

Number of candles used in the calculation, default 1000, range 1–1000. One record is returned for each candle, but the first few records may have no count. When both start_time and end_time are provided, the number of records returned is not guaranteed to correspond to limit.

int64

Start time, timestamp in milliseconds (UTC), e.g., 1790812800000 (2026-10-01 00:00 UTC). The candles used in the calculation satisfy start_time ≤ time < end_time: the start time is included and the end time is excluded. A start time that is not on an interval boundary is aligned to the next interval; if the start time is earlier than the history range your API plan can query, or the time span exceeds the maximum allowed for a single query, it is automatically moved later into the allowed range. If only start_time is provided, limit candles are taken forward from the start time.

int64

End time, timestamp in milliseconds (UTC), e.g., 1792022400000 (2026-10-15 00:00 UTC). The interval at the end time is not included; an end time later than the current time is treated as the current time. If both start and end times are omitted, limit candles are taken backward with the current time as the end point.

Response

Language
Credentials
Header
LoadingLoading…
Response
Click Try It! to start a request and see the response here! Or choose an example:
application/json