What happened, and what is reported to have caused it

Zcash is a privacy coin launched in 2016 with a Bitcoin style supply schedule: a 21 million cap, regular halvings, and roughly 16.9 million coins in circulation. At Sunday's high its implied capitalisation was near $20 billion, which reporting places around ninth in the market.

Four dated events are given as the cause. We measured none of them and we are relaying them.

  • A roughly two year SEC investigation into the Zcash Foundation closed in January 2026 with no enforcement action.
  • The Ironwood upgrade addressed a network flaw in July 2026.
  • Grayscale listed ZCSH, the first US spot Zcash ETF, on NYSE Arca on August 25, 2026. Reporting puts its assets above $400 million within two weeks.
  • Short liquidations of a reported $34.5 million on the September 4 move through $1,000, and above $46 million on Sunday.

What we can measure is what the tape did. On Binance ZECUSDT, Sunday opened at $1,025.94, reached $1,256.98 and closed at $1,227.60. It traded $426 million, which was 60% of Bitcoin's volume on the same day. Today Zcash is the fourth largest USDT pair on Binance out of 740, ahead of Solana.

All of that is real, and none of it is the interesting part.

Twenty sessions out of 384

Take every session from the August 20, 2025 low to today and compound them one by one. The result is 3,335% across 384 sessions. Now remove the best ones, in order, and compound what is left.

Sessions removedShare of the runCompounded return
None0%+3,335%
Best 10.3%+2,014%
Best 51.3%+614%
Best 102.6%+167%
Best 205.2%50% loss

Five percent of the sessions are the entire difference between multiplying your money by thirty three and halving it.

This is not a quirk of Zcash. Return concentration like this shows up in most assets that go up a great deal, and it is the reason "it is up 3,549%" is a weaker statement than it sounds. The gain was not earned steadily. Of the 384 sessions, 197 closed higher and 187 closed lower. On any given day it was close to a coin flip.

Here are the ten sessions carrying most of it.

SessionGainCloseVolumeBuy share
Oct 1, 2025+62.53%$120.76$164M54.5%
Oct 8, 2025+39.53%$177.56$157M55.5%
Aug 21, 2026+29.07%$734.27$368M54.1%
Oct 11, 2025+28.92%$291.00$392M53.6%
Apr 7, 2026+27.53%$320.78$155M56.0%
Oct 4, 2025+23.12%$156.78$129M51.6%
May 5, 2026+22.31%$517.03$210M54.1%
Jun 7, 2026+21.58%$440.60$345M53.3%
Feb 14, 2026+21.04%$323.52$135M51.0%
Mar 16, 2026+20.54%$276.87$128M53.5%

Look at the last column before you look at anything else. Every one of these sessions is between 51.0% and 56.0%. We will come back to that, because it is the most uncomfortable measurement in this article.

The path cost 75.4%

The run is not a line. It contains a fall that would have removed most of the people who started it.

PointDatePrice
LowAug 20, 2025$34.45
First peakNov 7, 2025$750.00
Trough, 91 days laterFeb 6, 2026$184.57
Reclaimed the peakAug 22, 2026$802.11
HighSep 6, 2026$1,256.98

From $750.00 to $184.57 is a fall of 75.4%. Getting back to that old peak from the low required a gain of 306%, and it took until August 22, 2026, which is 288 days after the peak with 287 closes below it in between.

So somebody who bought at $750 in November 2025, entirely reasonably, on the same privacy thesis everyone is repeating today, spent nine and a half months underwater and watched three quarters of their position disappear first. They are now up 60%. Both facts are true and only one of them is being posted.

The February 6 low is worth one more line. It landed inside a market wide convulsion rather than a Zcash specific one: Bitcoin fell 14.02% on February 5 and gained 12.19% on February 6. Zcash bottomed on the second of those two days. When the whole market breaks, correlations go to one and a coin's own story stops mattering for a while.

From that trough to Sunday's high is 581% in 212 days. That is the number that produced the headlines. It exists because of the 75.4% fall that came before it, not in spite of it.

Our main metric cannot see moves like these, and here is the number

The taker buy share is the measurement this site is built on. It answers one question: of the volume that crossed the spread, what percentage was buyers hitting offers rather than sellers hitting bids. We have written repeatedly that it describes urgency rather than direction. Zcash is the hardest test of that claim we have run, and it fails the optimistic reading decisively.

Zcash's own daily distribution over 400 sessions: median 50.19%, tenth percentile 45.60%, ninetieth percentile 53.84%.

Now the test. Zcash had 44 sessions gaining 10% or more during the run. How many of them showed a buy share above its own 90th percentile?

Thirteen. That is 29.5%. Seven in ten of the biggest up days looked completely ordinary on the tape.

The extreme case is the clearest one to state.

SessionAssetGainBuy share
Oct 1, 2025Zcash+62.53%54.50%
Feb 6, 2026Bitcoin+12.19%53.61%

Zcash's largest session in 400 days moved five times as far as Bitcoin's largest session in 400 days, and its buy share was 0.89 percentage points higher. If you had been handed those two readings with the price changes hidden, you could not have told them apart.

Sunday makes the same point on fresh data. Zcash gained 19.66% at a buy share of 52.89%, which is below its own 90th percentile. A twenty percent day that does not even register as unusual.

The correlation between a session's gain and its buy share is 0.487 for Zcash, almost exactly the 0.505 we measured for Bitcoin. So the relationship is real and it is the same strength on both. It is simply nowhere near strong enough to convert a buy share reading into a size prediction, and Zcash shows you what that looks like at the extreme.

Why the tape stays calm while the price does not

This is the maker repricing blind spot, and Zcash is the loudest demonstration of it we have found. When resting sell orders are cancelled and replaced higher rather than being bought, the price moves and no trade occurs. Nothing enters the tape. A market with a thin resting book can travel a very long way on modest aggressive volume, and every taker based metric on every venue stays quiet while it happens.

That is not a defect in our data. It is the difference between what price does and what trades, and no amount of better querying closes it.

Almost three quarters of accounts are short it

The Binance global long/short account ratio for Zcash reads 0.3701. That means 27.01% of accounts are long and 72.99% are short a coin sitting at an eight year high.

DateLong/short ratioAccounts shortZEC close
Sep 20.525665.6%$815.53
Sep 30.465268.3%$952.69
Sep 40.540664.9%$1,022.94
Sep 50.421570.4%$1,025.95
Sep 60.429470.0%$1,227.60
Sep 70.370173.0%partial

Six consecutive readings below parity, and the price rose 47.1% across them. The crowd has been positioned against this move for at least a week and has kept adding to the position as it went wrong.

We have written three times that a sub parity reading on Bitcoin describes the present rather than forecasting the future. On Bitcoin the most extreme reading we ever recorded was 0.7816, which is 56% short. Zcash at 0.3701 is far past that, and the honest thing to say is that we have no comparable history for a reading this skewed, so we are reporting it as a fact about how people are positioned and not as a signal about what comes next.

Two things it is fair to note. Perpetual open interest stands near 577,900 ZEC, worth roughly $709 million, so this is a real market rather than a rounding error. And funding has stayed close to flat, between 0.00625% and negative 0.00545% across the last nine settlements, which is odd company for a crowd this one sided and suggests the short accounts are individually small.

What we cannot see here

Zcash has one blind spot that Bitcoin does not, and it is the point of the coin.

Shielded transactions. The whole design of Zcash is that transfers can be encrypted, so the on chain analysis that people run on Bitcoin, tracking wallets and exchange balances and holder cohorts, does not work in the same way here. A large part of the standard analyst toolkit simply has no input on this asset. Anyone telling you what Zcash holders are doing should be asked how they know.

ETF creations. ZCSH listed on August 25 and reporting puts it above $400 million. Those creations settle between the issuer and its authorised participants, away from the public order book, exactly as they do for the Bitcoin funds. Our buy share never sees them.

Maker repricing. Covered above, and on this asset it is not a footnote. It is probably the main mechanism behind the numbers in the buy share section.

Other venues. We measure Binance ZECUSDT. Reported highs elsewhere differ from ours by around $30, which tells you the price is venue specific at these speeds.

Four blind spots is more than we carry on Bitcoin. That is worth saying out loud in an article about a coin that has gone up thirty three fold, because the temptation with a number like that is to sound more certain, not less.

What this article is not

It is not a view on where Zcash goes next, and nothing above should be read as one. We have published the run, the concentration, the drawdown, the tape and the positioning because all five are measurable. Where the price goes from $1,200 is not measurable, and the four other coin studies on this site end the same way.

If you want the single sentence a reader should take from the data: the return is real, it lives in about twenty days out of 384, and the path to it included losing three quarters of the position.