Estimate your stakes

1The idea: your share of the draw

Divi makes a block every 60 seconds, and each block is a draw among all the coins that are staking at that moment. Your odds in one draw are your staking balance divided by the network staking supply, the total DIVI that is actually staking (not the total that exists). Nothing else matters: not how long you have been staking, not how many coins your balance is split into, not which wallet you use.

Three quantities follow from that one share, and every tool on this page reports them:

• How often you stake. If your share is one block in 1,500, you win about once every 1,500 blocks, which is about 25 hours.
• The chance of a stake in a window (30 days, a year). Each draw is independent, so the chance of no stake in B blocks is e^{−B/blocks-per-stake}, and the chance of at least one is one minus that.
• What it averages a year. 525,600 blocks a year × your share × the reward per stake. The reward is about 415 DIVI on mainnet today.

2The explorer's Staking Intelligence tab

Open any address in the explorer and click the Staking Intelligence tab, or add ?tab=staking to the address URL:

https://services.divi.domains/explorer2/address/<your address>?tab=staking

Below the "What the chain shows" tiles (your real history) is What to expect, the forecast. For a mainnet address holding 1,663,230 DIVI on 2026-10-21 it read:

• A verdict line: ✅ On track. Last stake 2 hours ago. 32 stakes in the last 30 days (about 28 expected).
• Stakes about · once every 26 hours · For a wallet this size, on average
• Each stake pays · 415 DIVI · Typical reward over the last 7 days
• Chance of a stake · Near certain · More than 99% within 30 days
• Averages a year · 142,185 DIVI · 8.5% of the balance, over the long run
• A footer: "Staking is random, so these are odds, not a schedule. A wallet stakes only while it is open, synced and unlocked for staking; a vault stakes on its staking node. Network staking supply measured at 2,551,539,080 DIVI at block 4,269,178."

The verdict can be one of five: ✅ On track, 🟡 A slow patch, 🟡 Staking only part of the time, ⚠️ Overdue, or ⏳ Waiting for its first stake. Section 6 says exactly where the lines are. When the Chance of a stake within 30 days is under 99%, the tile also gives the chance within a year.

3The sum by hand

You need two numbers: your staking balance and the network staking supply. Both nodes report the first in getstakingstatus. On a Divi node it is the coins old enough to stake, after the reserve balance; IronDivi reports the same field:

# Divi node (divid 3.0.0, testnet node used for this page)
divi-cli getstakingstatus
{
    "validtime" : true,
    "haveconnections" : true,
    "walletunlocked" : false,
    "mintablecoins" : false,
    "staking_balance" : 12999.99986980,
    "enoughcoins" : true,
    "mnsync" : true,
    "staking status" : false,
    "stake split threshold" : 20000.00000000
}

# IronDivi 0.2.6
irondivi-cli getstakingstatus
{
  "staking_balance": 12000.0,
  "staking status text": "Staking Active",
  ...
}

Divi Desktop and BoxWallet users: the balance on screen is close enough, as long as the coins are older than an hour. Then fetch the supply and the reward:

curl -s https://services.divi.domains/api/rpc/network.json
{"measured_height":4269178, "staking_supply_divi":2551539080,
 "stake_reward_satoshis":41500000000, "sampled_wallets":5, "sampled_stakes":1818, ...}

Now the four lines of arithmetic, done for the mainnet address from section 2 (1,663,230 DIVI) and for the 13,000 tDIVI testnet wallet above (testnet supply 436,516,454, reward 166 tDIVI):

Compare with the tab: 1,534 blocks is 25.6 hours, which it shows rounded as "once every 26 hours"; 142,185 DIVI and 8.5% match to the digit. The testnet rate is far higher than mainnet's because testnet has a tiny staking supply for its reward. Here is the same sum as one line you can paste, with your own balance and the two numbers from the JSON:

python3 -c 'import math; b,s,r=1663230,2551539080,415; bps=s/b; print(f"one stake per {bps:,.0f} blocks = {bps/60:.1f} h; 30-day chance {100*(1-math.exp(-43200/bps)):.1f}%; {525600*b/s*r:,.0f} DIVI/yr = {100*525600*r/s:.2f}%")'
one stake per 1,534 blocks = 25.6 h; 30-day chance 100.0%; 142,185 DIVI/yr = 8.55%

4What different balances can expect

The same sum at today's mainnet supply (2,551,539,080 DIVI, 415 DIVI a stake). The yearly rate is 8.55% on every row; the rows differ only in how often that arrives.

Read the small rows honestly: 10,000 DIVI expects one stake every six months or so, and a one-in-eight chance of none at all in a year. That is not a fault, it is what a 0.0004% share of the draw looks like. The tab and the bot both add a note to such wallets: "Rewards come in rare, large payments rather than a steady trickle; a bigger balance stakes more often." Staking returns walks through what a year of this adds up to.

5Ask the Stake Watch bot

The Stake Watch bot on Telegram does the same sum in a private chat. Send /check D... with any mainnet address (it accepts addresses that start with D; testnet ones are refused with "That doesn't look like a Divi address"), or a bare /check to check up to three addresses you have put on /watch. The reply is built from these lines, from top to bottom (wording from the bot's source, version 0.1.0):

The bot measures the staking supply itself, also every 6 hours, from its own sample of the chain, so its figures and the explorer's can differ by a few percent on the same day. Both are estimates of a number that nobody can read exactly, because the chain does not record which coins are staking. The bot's other commands (/watch, /alert, /history, /forkwatch) are covered in Stake alerts.

6When a gap is "slow" or "overdue"

Both the tab and the bot judge your current dry spell by one question: what fraction of wallets this size would still be waiting after this long? That is the no-stake chance from section 1, e^{−gap ÷ blocks per stake}. The lines are drawn at two values of it:

Because the no-stake chance falls as e^−x, the thresholds are fixed multiples of your average wait: a gap is "slow" after about 2.3× the average (ln 10) and "overdue" after about 4.6× (ln 100). So:

The third verdict, Staking only part of the time, catches a wallet whose last stake is recent but whose 30-day count is far too low: it asks how likely it is to see this few stakes or fewer when "about N expected" (a Poisson probability), and fires under 1%. A wallet that is open only in the evenings looks exactly like this.

7What the estimates can't see

• Whether you are actually staking. Every figure assumes the wallet is open, synced and unlocked for staking the whole time. The web wallet, the browser extension and the Telegram wallet don't stake at all, so for an address held in one of those the forecast is what it would earn in a staking wallet. Divi Desktop, BoxWallet, a Divi node and IronDivi all stake; see Stake in a wallet.
• Vaults. A vault's coins stake on its staking node, so for a vault address the tab reports the owner's balance and the footer says so. The sum is the same.
• Very busy addresses. The explorer gives no summary or forecast for an address with more than 20,000 transactions; exchange hot wallets and the largest stakers fall over that line. Use the sum by hand with the balance from your own node.
• Old-style stake outputs. Coinstake outputs paid to a bare public key rather than an address can't be found by address, so an old wallet's stake count can be short.
• The supply itself. It is measured from the busiest stakers of the last 30 days, not read from the chain, so two tools can disagree by a few percent, and all of them move when big wallets start or stop staking.
• Masternodes. None of this covers masternode rewards, which are a separate system.

Next

• Why am I not staking? when the verdict is "overdue".
• Stake alerts: let the Stake Watch bot message you instead of checking.
• Staking returns: what a year of 8.5% looks like in the explorer's "Plan ahead" section.
• Split and combine coins: why the number of coins doesn't change your odds, and what it does change.
• Explorer tour and the glossary.