Where is my block?
Staking on Algorand lets users participate in securing the network and earn rewards for doing so. At its core, Algorand’s consensus protocol relies on randomness to select which accounts produce new blocks, as outlined in one of our previous blog posts. This means even when you’re actively staking, you might not see consistent block production. In this post, we break down what to expect in terms of block production on Algorand during staking and how you can check your own performance.
Expected Time Between Blocks
The Algorand protocol selects a block producer randomly on each round, and the probability of selection increases with the size of your stake. Owning a larger portion of the total online stake increases your odds of producing blocks. We can estimate the expected time between produced blocks using a simple relationship:
For example, assuming the average block time is 3 seconds, and you have 150k ALGO staked out of a total 1.5B ALGO online, then you can expect to produce a block approximately once every 8 hours on average.
Variation of Block Production
In practice, you won’t produce a block exactly every 8 hours. The random selection in Algorand’s consensus causes the actual time between produced blocks to vary significantly. The below figure illustrates how often an account with 150k ALGO needs to wait a certain amount of time to produce a block.
From the data, block intervals ranged from just 4 hours to over 40 hours, demonstrating that the time between consecutive produced blocks can be significantly different from the expectation, i.e., the average time between blocks when observing block production over a long time duration.
Occasional Long Times Are OK
Long stretches without a produced block can feel frustrating — but they’re not a bug. They’re just how probability works. The below figure shows what the likelihood of not producing a block within a certain time looks like.
While the likelihood of going a certain amount of time without producing a new block decreases rapidly in the beginning, the slope gradually flattens and this probability never goes to zero. For example, the 150k ALGO account from the previous section has a 35 % probability of not producing a block within the first 8 hours and a 1 % probability of not producing one for 40 h. In other words, it’s OK to have to wait five times longer than the average 8 hours for one block out of a hundred.
Check for Yourself
During the recent Algorand Developer Retreat, we brought together input from multiple ecosystem experts to model and validate this behavior. The result is a prototype of a simple tool, built to help anyone better understand and evaluate their own block production history. You can find the tool and in an open-source repository at: https://github.com/Algorand-Developer-Retreat/block-anxiety
You are welcome to also explore other tools and case studies that were developed as part of the Algorand Developer Retreat at: https://github.com/Algorand-Developer-Retreat
Disclaimer
This article does not constitute financial advice. All information provided is for general purposes only. Readers should conduct their own research and fully understand the risks before participating in any staking or other blockchain activities. The information provided does not address all potential risks or other relevant considerations of staking, blockchain, and financial activities. The discussed data and models are of indicative nature only and no guarantee is given that these fit or reflect the actual observations of blockchain users.
