Bitcoin transactions per second (TPS) is the number of qualifying on-chain Bitcoin transactions counted during a stated time window, divided by the window’s duration in seconds. It describes transaction-record activity under a specified method—not a fixed capacity guarantee or a complete measure of payment performance. To interpret a reported rate, check what was counted, which records were included, and how long the measurement period lasted.

What Bitcoin transactions per second means

A Bitcoin transaction uses inputs to spend specific prior transaction outputs and creates new outputs. Those outputs remain unspent until a later transaction spends them. The underlying record therefore represents a transaction in Bitcoin’s on-chain model; it does not, by itself, establish how many people, businesses, or broader payment events are involved. source

For an on-chain TPS measure, the numerator is the number of qualifying transaction records in the selected data over the chosen period. The denominator is the duration of that period in seconds. The result is expressed as transactions per second.

The counting rule is essential. A published figure should identify which records count as transactions and what data is included. If two reports use different rules or data scopes, their numbers may not be directly comparable even if both use the label “Bitcoin TPS.” The metric should remain focused on on-chain activity unless its methodology explicitly defines a broader scope.

How to calculate a Bitcoin transaction-per-second rate

The basic calculation is:

TPS = qualifying transaction count ÷ elapsed seconds

If a defined dataset contains N qualifying transactions during a window lasting S seconds, the rate for that window is N ÷ S. This illustrates the arithmetic only; it is not an observed Bitcoin rate. A numerical result requires an actual transaction count, a stated counting rule, a data source, and the duration measured.

A measured rate reports a count from specified data over a past interval. An estimate may rely on assumptions or incomplete information. A theoretical or protocol-level figure is a different type of claim and should not be presented as though it were a measurement of observed activity. Each requires its own stated method and scope.

The available evidence here supports the basic transaction structure, but does not provide a particular dataset, reproducible count, or numerical TPS value. A precise figure cannot be calculated from the definition alone. Anyone publishing one should identify the records used and explain how the count and time period were determined.

Scope also affects the calculation. A rate derived from Bitcoin’s on-chain transaction records is an on-chain rate. A figure that purports to include activity beyond those records needs a method that defines and measures that activity; it cannot be inferred from an on-chain count alone.

Why the rate varies over time

The number of qualifying transactions recorded during one interval can differ from the number recorded during another. Dividing each count by its own duration produces a rate for that specific interval. A value observed in one window should not be treated as a constant simply because it is expressed per second.

Window length changes what a figure summarizes. A short-window result describes activity during that brief period. A longer-window average summarizes the combined period. Those measurements answer different questions and can produce different rates without either being inherently inconsistent. A published figure should state its period so readers know which activity it represents.

The numerator can vary with the counting rule and dataset, too. If sources count different records or apply different inclusion criteria, their transaction counts—and resulting rates—may differ. The unit alone does not make two figures comparable. Readers need enough methodological detail to understand both the records counted and the time over which they were counted.

What transactions per second does not tell you

TPS is not the same as a block interval. TPS counts qualifying transactions during a stated period; a block interval concerns the spacing of blocks. Nor is TPS the time a particular transaction takes to receive confirmations. Bitcoin confirmations indicate increasing confidence in a transaction’s place in the chain, with the degree of confidence depending on confirmations and operating conditions. That transaction-specific consideration is distinct from a rate of counted transactions.

A transaction count alone does not establish transaction size, value, how many users or organizations are represented, or whether a particular payment workflow meets its operating requirements. Payment-network suitability should be assessed against the required transaction flow, confirmation and finality expectations, capacity, fee behavior, application integration, dependencies, and operational requirements. A TPS figure addresses only one part of that assessment. source

An on-chain figure also does not automatically include off-chain payment activity. If a report claims broader coverage, its methodology must explain what additional activity it counts and how it combines that activity with on-chain records. Without such a definition, interpret the number as an on-chain measure only when its stated data supports that scope; do not assume it captures all Bitcoin-related payment activity.

How to assess a published Bitcoin TPS figure

Before interpreting a reported number, check the following:

  • Counting rule: What qualifies as a transaction, and which records are included in the count?
  • Time window: What period does the figure cover? Is it a short observation or an average over a longer interval?
  • Method: Is the number measured from identified data, estimated using stated assumptions, or presented as theoretical or protocol-level?
  • Scope: Does it cover on-chain transaction records only, or does the report define and document additional activity?
  • Reproducibility: Does the source provide enough information about its data, count, and duration for someone to understand how the rate was obtained?

If a report omits these details, the figure is not sufficiently specified to assess as a reproducible rate. Treat it as unverified rather than as an authoritative statement of capacity. Even a fully specified TPS number answers a narrow question: how many transactions the stated method counted per second over the stated interval. It does not, on its own, describe confirmation expectations, payment experience, or whether Bitcoin meets a particular organization’s requirements.

Sources and references