Dogecoin Weekly Data Report — September 14–20, 2026

The dashboard at a glance

Price rose while usage fell for the second time in four weeks — the fourth consecutive weekly decline in transactions. As of Sunday, September 20, 2026:

Price is the Gate.io DOGE/USDT Sunday close (cross-checked against CoinEx, maximum discrepancy $0.000166). Transactions and active addresses are 7-day averages of CoinMetrics TxCnt and AdrActCnt for September 14–20; the Sunday print alone was 22,266 transactions and 33,395 active addresses. Hashrate, difficulty, fees, issuance and holder counts are Blockchair network stats pulled September 21. Market cap is computed from on-chain supply times spot price. Every week-over-week change is calculated on unrounded values.

The week in price

The week was a two-stage test. DOGE opened at $0.0824, sagged through Tuesday and Wednesday as the Fed decision and the CLARITY Act's failure hit the tape, printed its intraday low of $0.07835 on Wednesday, then reversed hard on Friday once event risk cleared. The Friday and Saturday closes at $0.08746 and $0.08770 were the week's best, and Sunday gave back only a little to close at $0.08731.

Four weeks of closes

Zooming out, September 14–20 recovered roughly two-thirds of the prior week's decline and put the asset back into the range it has occupied since August. The four-week span:

Net over the four weeks: +6.4% from the August 30 close to the September 20 close. Month to date (August 31 close → September 20 close), DOGE is up 5.4%. Weekly changes are computed on unrounded closes.

On-chain usage: the fourth decline

Here is the number that matters most in this edition. Price-and-usage lockstep had already broken three weeks ago; this week the usage decline steepened. DOGE rose 6.0% while daily transactions fell 7.2% to a 23,660 average and daily active addresses fell 5.8% to 36,377. The Sunday print was weaker still: 22,266 transactions and 33,395 active addresses, the lowest active-address count of the week.

We have flagged this comparison every week, and consistency requires us to be exact about what it is and is not. Four weeks is still not a trend, but it is no longer a single print. A 25% contraction spread across a month could still be a slow drift rather than a break — Friday's rally was a macro beta move, and on-chain counts often lag a sharp two-day price jump. What it does establish is that the two series can separate, which is the precondition for the adoption thesis to show up in data at all. The test for next week: does the transaction series recover toward 25,000 as the price holds, or does it stay near 23,000 while price sits above $0.085? The former re-couples; the latter starts a trend.

Context that matters more than any single print: a $13.62B market capitalization supported by about 23,660 daily transactions works out to roughly $576,000 of market cap per daily transaction, or about $374,000 per daily active address. That ratio is the cleanest single statement of where Dogecoin stands — a large asset by market value, a modest one by on-chain workload.

Mining and the chain's clock

The security layer cooled slightly. Hashrate eased to 3.81 PH/s (24-hour average) and network difficulty retargeted down to 55,873,021, about 2.5% below the prior week's 57,320,342 — a real but unremarkable adjustment after August's hashrate highs. The chain's clock ran close to target: the week averaged about 1,362 blocks per day, an implied block time near 63.4 seconds against the 60-second design target. That is ordinary proof-of-work variance, and we flag it so a slightly slow week never gets mistaken for a signal.

There is no staking rate to report, because Dogecoin is pure proof-of-work: there is nothing to stake and no yield to quote. Any dashboard showing a DOGE 'staking APY' is describing a centralized wrapper, not the chain.

One sourcing note we repeat from last week: cross-source hashrate estimates differ materially for Dogecoin (a function of merge-mining attribution), so we quote the Blockchair snapshot and label it as such rather than blending sources.

Fees: what a transfer actually costs

The median transaction fee printed at 0.0590 DOGE, about $0.0052 at the weekly close. In dollar terms a DOGE transfer remains one of the cheapest things in crypto; the median fee did rise from last week's 0.0334 DOGE, though we would not read a single-snapshot move in the median as a congestion signal — no congestion event occurred this week and the mempool stayed small (824 transactions, 185 KB at the time of the snapshot).

Two derived facts sharpen the picture. At ~23,660 transactions per day, total fee income for miners is on the order of $122 per day — about 0.01% of the roughly $1.19M of daily issuance. And that is fine, because it means Dogecoin's security budget is a deliberate, subsidy-funded choice rather than a market outcome.

For USDT movements specifically, renting energy on TRON cuts the cost by roughly 90% versus paying bandwidth directly — which is why we send readers comparing stablecoin transfer costs to Tronsell's energy rental rather than pretending DOGE's fee is the only number in the comparison.

Supply and holders

Circulating supply stood at 156.00 billion DOGE on September 20, with roughly 13.62M DOGE (about $1.19M) added per day by the block subsidy — about 3.19% annual supply growth, or 4.97B DOGE a year, with no maximum supply by design. The number of addresses holding DOGE reached 8,473,412 (Blockchair), up about 13,400 over the week. Supply growth is the fixed cost of the security model; it never pauses and never scales down.

Scenario math: what $0.075 and $0.10 do to the security budget

The dashboard above is a snapshot. Because almost everything in Dogecoin’s economics is a function of two inputs — the price and the block subsidy — it is worth showing how the numbers move when the price does. Everything below is arithmetic on the September 20 supply (156.00B DOGE) and the observed issuance rate (13.62M DOGE a day). Nothing in it is a forecast.

Two things fall out of it. First, the subsidy’s dollar value scales one-for-one with price: a 20% move in DOGE is a 20% move in what miners earn, whether or not a single extra transaction is processed. Second, the fee share of miner revenue does not move at all — it stays near 0.01% across a 43% price range, because numerator and denominator are both denominated in the same coin. That is the cleanest way to see why Dogecoin’s fee market is structurally irrelevant to its security budget, and why no plausible price scenario changes it.

Read the same table from a holder’s side and it becomes a cost: at the September 20 close the network issues about $434M of new DOGE a year against a $13.62B market cap — roughly 3.19% dilution, independent of price. A flat year of price and a flat year of supply growth still leaves existing holders with a smaller share of the network.

The per-transaction ratio, three ways

We used $576,000 of market cap per daily transaction as this edition’s headline ratio. It is worth seeing how sensitive that number is to its denominator, because ‘per transaction’ and ‘per active address’ tell slightly different stories and both get quoted in the wild.

The table makes the adoption test concrete. Nothing in Dogecoin’s price has to change for this ratio to improve — it improves if usage rises. Equally, nothing in usage has to change for it to worsen: a rally on flat transaction counts widens the gap. That asymmetry is why we treat the transaction series, not the price, as the series carrying information.

What changed this week

• Price: +6.0% to $0.08731, with a midweek low of $0.07835 and a Friday–Saturday recovery.
• Usage: transactions −7.2% (23,660/day), active addresses −5.8% (36,377/day) — a fourth consecutive weekly decline, 25% below the August 17–23 peak.
• Security: difficulty retargeted down 2.5% to 55.87M as hashrate eased to 3.81 PH/s; block time averaged ~63.4s.
• Fees: median 0.0590 DOGE (~$0.0052); fee income still about 0.01% of issuance, with no congestion event.
• Institutional: one ~$285,000 DOGE ETF inflow day (Sep 15, GDOG) against ~$595M of market-wide crypto ETF outflows the same day.
• Holders: +13,400 addresses holding DOGE, to 8,473,412.

Reading the numbers like an analyst

Three reads worth doing. First, the usage check. Price +6.0%, transactions −7.2%, addresses −5.8% — the fourth weekly decline in a row. A single print is noise, but a month of declines is not; the specific thing to watch is whether the transaction series re-couples to price or starts tracking its own level. Second, the per-unit math. $576,000 of market cap per daily transaction is not 'bad' by itself — it is the quantitative gap between Dogecoin's market value and its on-chain workload, and every adoption story that matters should eventually compress it. Third, the security-budget read. Difficulty falling 2.5% while price rose is the normal, healthy behaviour of a merge-mined chain: hash rate follows miner economics, not price sentiment, and a modest retarget after an August high is housekeeping, not capitulation.

What the numbers mean for you

If you hold DOGE as a position, the dashboard says the asset is still priced by macro narrative rather than usage — this week that meant +6.0% on a Fed decision and a short squeeze, not on adoption. If you use DOGE to move value, the price tape is irrelevant: a transfer still costs about $0.0052 and confirms in about a minute. If you are comparing chains for stablecoin payments, run the full comparison — DOGE's ~$0.0052 median fee versus TRON's ~$0.02 with rented energy — instead of quoting any single number out of context. And if you are watching for a real signal, watch the transaction series relative to price: it has now fallen four weeks running, and next week decides whether that is a trend or a drift.

Appendix: data snapshot

Note on market-cap discrepancies: some aggregators list a higher circulating supply than CoinMetrics' on-chain SplyCur, which makes their market caps read higher. We compute from on-chain supply times spot price and label the source on every figure — when trackers disagree, we would rather show our work than pick the biggest number.