Analysis of Bitcoin Transaction Size Trends

Follow:  

The analysis below examines factors contributing to the increase in the average size of bitcoin transactions over the past couple of years. In particular, the average transaction size has increased from 450 bytes in Jan 2013 to almost 600 bytes in Oct 2015. Meanwhile, the prevalence of P2SH and other non-typical transaction outputs has also risen. To carry out the analysis, ~86 million transactions were categorized into one of five buckets, based the output scripts of each transaction.¹

Avg Tx Size

Categorization of Transaction Type
Included in parenthesis “()” below, is the proportion of 2015 transactions within each category.

    • ‘Basic’ (89%): Standard bitcoin transaction from Party A to Party B’s bitcoin address. Most within are Pay-to-Public-Key-Hash, also known as P2PKH transactions. Transaction example (see ‘output scripts’).
    • ‘P2SH’ or Pay-to-Script-Hash (7%): P2SH allow for transactions to be sent to a script hash (addresses starting with ‘3’) instead of a public key hash. Used commonly for multi-signature scripts. Transaction example (see ‘output scripts’).
    • ‘Multisig(1%): Multi-signature scripts set a condition where N public keys are recorded in the script and at least M of those must provide signatures in order to release the encumbrance. Transaction example (see ‘output scripts’).
    • ‘RETURN’ (1%): Transactions that provide a mechanism for inserting metadata into the blockchain. OP_RETURN allows senders to add 40 bytes of non-payment related data to a transaction output. Transaction example (see ‘output scripts’).
    • ‘Non-standard’ (2%): Output script does not fall into any of the categories above. Transaction example (see ‘output scripts’).

 

All non-basic categories are collectively labelled as ‘atypical’ transactions for purposes of the analysis. Summarized below are the results from the analysis.

 

1) Historical Transaction Type Trends: 2009-2015

Fig 1

    • Non-basic or atypical transaction types began appearing on the Bitcoin blockchain in 2013.
    • Atypical transactions types comprised 0.03%, 1.23%, and 10.15% of the total transaction volume in 2013, 2014, and 2015 (to-date), respectively.

 

2) Transaction Type Trends vs. Average Transaction Size: 2014-2015

Fig 2

    • Thus far in 2015, P2SH transactions have reached as high as 15% of the daily transaction volumes; meanwhile, multisig transactions have remained under 2%, partly owing to the efficiency of P2SH over the latter.
    • Non-standard transactions appear to coincide with recent attempts to saturate the network with spam transactions, also known as the ‘stress tests.’
    • Note: Chart represents 30-day moving averages.

 

3) Daily Transaction Sizes: 2015

Fig 3

    • The chart above summarizes the mean and median daily transaction sizes for the five categories described.
    • ‘Basic’ bitcoin transactions with 1 input and 2 outputs are typically ~250 bytes of data.
    • ‘P2SH and ‘non-standard’ transaction categories, which together comprise 87% of atypical transactions, have average byte sizes that are 52% and 80% higher than ‘basic’ transactions, respectively.

 

4) Blockchain Data Size vs. Transaction Volume: 2015

Fig 4

    • The somewhat muted effect of atypical transactions on the blockchain size is illustrated in the two charts above.
    • Of note, P2SH transactions have accounted for 7% of total transactions thus far 2015, while the overall data size is disproportionately higher at 10% of the total.
    • The use of atypical transaction scripts likely had an impact on the increasing global average transaction size; albeit, the effect is limited given that such transactions comprise only 10% of the total volume.

 

Shifting the focus back to ‘basic’ transactions, the average daily size of such transactions was 566 bytes in 2015, per Figure 3, more than double its median figure of 274 bytes. Part of this variation is explained by several instances larger transactions, such as this 999,657 byte single transaction that occurred during the July stress test. Interestingly, the aforementioned transaction had over 5,000 inputs, and requires the same byte space that ~4,000 ‘basic’ transactions would.

 

5) Inputs/Outputs within ‘Basic’ Transactions

Fig 5

    • The chart above highlights the high degree of correlation between the daily average number of inputs and outputs for ‘basic’ bitcoin transactions relative to the average daily transaction sizes.
    • Examination of the data reveals a significant spike the average daily inputs & outputs during the stress tests in the July-Sep period.
    • Note: Chart represents 30-day moving averages for both series of data.

 

Final Thoughts

Based on the analysis, there are two primary factors that have contributed to 33% larger average transaction sizes since 2013:

    • The proportion of atypical transactions, which include P2SH, multisig, RETURN, and non-standard outputs, has been increasing steadily. Majority of these transactions, while comprising only 10% of the total volume today (up from 1% in 2014), tend to be 50-80% larger than basic/standard bitcoin transactions, on average.
    • However, the more pronounced effect on transaction size stems from the incorporation of a greater number of inputs and outputs within ‘basic’ transactions, as depicted in Figure 5, above. This trend was amplified during the recent network capacity tests.

 

¹This analysis uses output scripts to categorize transactions to ensure consistency. An analysis of transaction inputs revealed similar trends, especially with regard to P2SH scripts.

This analysis has been prepared in good faith on the basis of information available at the date of publication without any independent verification. Schvey, Inc. does not guarantee or warrant the accuracy, reliability, completeness or currency of the information in this presentation nor its usefulness in achieving any purpose. Readers are responsible for assessing the relevance and accuracy of the content of this publication. Schvey, Inc. will not be liable for any loss, damage, cost or expense incurred or arising by reason of any person using or relying on information in this publication. This analysis may not be duplicated, shared, or reproduced in its entirety or in part for any reason without the expressed written consent of Schvey, Inc.

Subscribe

For the latest analysis and updates