Home
/
Digital wallets
/
Wallet security
/

Evaluating dice rolls for enhanced seedphrase security

Fresh Insights on Dice Rolls | Seeds of Security in Crypto Controversy

By

Javier Morales

Aug 31, 2026, 12:48 AM

Edited By

Elena Ivanova

Updated

Aug 31, 2026, 07:12 AM

2 minutes to read

A hand rolling multiple dice next to a Keystone Pro 3 device, illustrating the relationship between dice rolls and seedphrase safety

A recent test with dice to generate crypto seed phrases has ignited discussions across various forums. Users are questioning the reported 167 bits of entropy from 100 rolls, prompting debates about the necessity of additional rolls for security.

What Sparked the Debate?

The user rolled 100 dice on Keystone Pro 3 to create a seed phrase, which was verified using Ian Coleman’s BIP39 tool, leading to confusion around entropy calculations. The reported entropy is lower than the expected maximum of 256 bits. This experiment reflects concerns around the tools used for generating secure seed phrases.

Key Clarifications from Comments

  • Entropy Calculation Issues: One contributor explained that the 167-bit number stems from Ian Coleman’s Raw Entropy mode, where four of six dice outcomes yield only 2 bits and the others yield 1 bit. This averages out the actual bits collected, which does not directly affect the efficacy in selecting 24 words. "100 rolls already give you about 258 bits of input entropy," highlighted an informed user, emphasizing the importance of understanding the calculation differences.

  • Bias Considerations: There were discussions about the potential bias in dice outcomes. Participants noted that even with a slightly biased die, approximately 2.5 bits of entropy can still be collected per roll. The mapping process used in conversions can create a perception of decreased entropy but does not directly affect overall security when fed into SHA-256 hashes.

  • Adjusting Methodology: Users debated whether adjusting the methodology could yield better results. Some questioned the effects of using Base 1-10 in the Ian Coleman tool and its relation to manually generated seed phrases, comparing it with ColdCard's manual rolling methods. Consensus suggests that the conversion processes are similar, confirming that proper input should yield matching results.

"Rolling the dice just a few more times than 99 is sufficient to achieve 256 bits of entropy," noted one commenter, emphasizing the industry standards.

User Sentiments

Overall, contributors shared mixed views on the number of rolls required for a secure seed phrase, with many advocating for the standard of 99 or 100.

A mixture of skepticism and practical insights permeated the discussions, with some comments reflecting the tiring nature of generating seed phrases manually. One contributor humorously noted, "I did it a few days back with coin flips god does it get tiring after a few minutes!"

Key Points to Note

  • πŸ”Ή 167 bits result from bias calculations in techniques.

  • πŸ”Ή Participants agree that 99-100 rolls typically meet security standards.

  • πŸ”Ή "100 rolls already gives you about 258 bits of input entropy," explains a knowledgeable source.

  • πŸ”Ή Using dice adjustments effectively acknowledges bias impact.

As discussions around seed phrase creation progress, user experiments continue to influence best practices, stressing the need for reliable tools and community education. With heightened adoption of crypto in 2026, increased resources on how to create secure seed phrases are anticipated to become mainstream as forums adapt and evolve based on user feedback.