Crowding

Nominally independent vaults are often in the same trade. When the market moves against them they do not leave one by one — they leave together, through the same liquidity, turning a controlled loss into a cascade. This page measures how much of that is happening, and is honest about how much of it we can see. Click an instrument to see exactly which vaults hold it.

Who trades after whom

7 days to 2026-08-25 00:27Z · 2,600,795 fills · 245 pairs

Holding the same instrument is not the same as having the same idea. This looks at timing instead: pairs where one vault’s trades are repeatedly followed by another’s, sooner than the two series produce when shifted past one another. Fills are collapsed into decisions first, because an execution algorithm slicing one order into fifty is one intent, not fifty.

Leads	Follows	On	Together	Shifted	Excess	One-way
Growi HF	Satori Quantum HF Vault	CRV	71	0.5	70.5σ	78%
Growi HF	Satori Quantum HF Vault	INJ	118	2.4	62.1σ	79%
Growi HF	HyperTwin - Growi HF 2x	INJ	86	1.7	60.3σ	77%
Growi HF	Satori Quantum HF Vault	BCH	60	0.3	59.7σ	85%
Quant Mirror	V3 momentum	ETH	56	0.8	55.2σ	84%
Growi HF	HyperTwin - Growi HF 2x	BCH	53	0.4	52.6σ	71%
Growi HF	Satori Quantum HF Vault	LINK	53	0.5	52.5σ	89%
Growi HF	Satori Quantum HF Vault	NEAR	95	1.8	51.8σ	76%
Growi HF	Satori Quantum HF Vault	AAVE	111	3.1	51.6σ	80%
Growi HF	HyperTwin - Growi HF 2x	CRV	51	0.3	50.7σ	65%
Growi HF	Satori Quantum HF Vault	SOL	51	0.5	50.5σ	85%
Satori Quantum HF Vault	HyperTwin - Growi HF 2x	BCH	50	0.5	49.5σ	64%
Quant Mirror	V3 momentum	LINK	73	2.1	47.8σ	90%
Satori Quantum HF Vault	HyperTwin - Growi HF 2x	INJ	84	2.0	47.7σ	73%
Satori Quantum HF Vault	HyperTwin - Growi HF 2x	AAVE	81	2.1	47.1σ	62%
Growi HF	HyperTwin - Growi HF 2x	LINK	47	0.3	46.7σ	62%
Growi HF	HyperTwin - Growi HF 2x	ETH	47	0.4	46.6σ	65%
Growi HF	HyperTwin - Growi HF 2x	AAVE	81	2.2	44.5σ	64%
HyperTwin - Growi HF 2x	Quant Mirror	BCH	45	0.5	44.5σ	36%
HyperTwin - Growi HF 2x	Quant Mirror	CRV	45	0.6	44.4σ	38%
Satori Quantum HF Vault	HyperTwin - Growi HF 2x	CRV	51	0.5	44.2σ	55%
Growi HF	HyperTwin - Growi HF 2x	SOL	44	0.6	43.4σ	63%
Quant Mirror	V3 momentum	INJ	131	4.5	43.1σ	88%
Growi HF	HyperTwin - Growi HF 2x	XRP	90	1.7	43.0σ	67%
Growi HF	Satori Quantum HF Vault	ETH	43	0.4	42.6σ	69%
Growi HF	HyperTwin - Growi HF 2x	NEAR	68	1.0	42.5σ	62%
Satori Quantum HF Vault	HyperTwin - Growi HF 2x	LINK	42	0.6	41.4σ	56%
HyperTwin - Growi HF 2x	V3 momentum	INJ	78	2.8	41.4σ	42%
Quant Mirror	V3 momentum	BCH	70	1.0	40.7σ	89%
HyperTwin - Growi HF 2x	Quant Mirror	INJ	72	2.5	40.5σ	41%
Quant Mirror	V3 momentum	NEAR	117	3.7	40.4σ	83%
Growi HF	Satori Quantum HF Vault	UNI	89	2.1	40.0σ	85%
HyperTwin - Growi HF 2x	V3 momentum	BCH	43	0.5	40.0σ	39%
Quant Mirror	V3 momentum	CRV	81	1.5	38.9σ	98%
Growi HF	Quant Mirror	BCH	39	0.4	38.6σ	42%
Growi HF	Satori Quantum HF Vault	BTC	58	0.7	37.8σ	81%
HyperTwin - Growi HF 2x	V3 momentum	CRV	41	0.7	37.2σ	32%
Quant Mirror	V3 momentum	BTC	63	1.0	36.9σ	83%
Growi HF	Satori Quantum HF Vault	kSHIB	71	1.1	36.6σ	75%
Quant Mirror	V3 momentum	UNI	116	4.0	36.5σ	87%

This does not say copying. A copied signal, a shared data vendor, one operator running both books, or the same obvious level on the same chart all produce this pattern, and none of them can be told apart from a timestamp.

Does being in a crowd cost anything?

121 events across 112 vaults

Vaults entering a crowd went on to underperform matched vaults that did not by -135281.76% at the worst point of the following 20 days. Shuffling the event dates produces a path that extreme 26% of the time, so this is not yet distinguishable from noise.

Day	Entered a crowd	Matched controls	Difference	Events
-5	1347.03%	887.77%	459.26%	119
-4	1071.39%	205.99%	865.40%	119
-3	450.37%	71.20%	379.17%	119
-2	292.07%	28.83%	263.24%	119
-1	36.11%	6.62%	29.49%	119
+1	62.24%	0.66%	61.58%	119
+2	3663.23%	12.43%	3650.79%	119
+3	15637.06%	3359.97%	12277.09%	119
+4	44341.31%	28647.83%	15693.48%	120
+5	81299.32%	111480.06%	-30180.74%	120
+6	125772.40%	261054.16%	-135281.76%	120
+7	940495.95%	448420.52%	492075.43%	121
+8	90033927.05%	734160.48%	89299766.57%	121
+9	11896849066.26%	1870513.00%	11894978553.26%	121
+10	12111414679.95%	4240060.65%	12107174619.29%	121
+11	13141975834.85%	5300688.32%	13136675146.53%	121
+12	13141975834.85%	5300688.32%	13136675146.53%	121
+13	13141975834.85%	5300688.32%	13136675146.53%	121
+14	13141975834.85%	5300688.32%	13136675146.53%	121
+15	13141975834.85%	5300688.32%	13136675146.53%	121
+16	13141975834.85%	5300688.32%	13136675146.53%	121
+17	13141975834.85%	5300688.32%	13136675146.53%	121
+18	13141975834.85%	5300688.32%	13136675146.53%	121
+19	13141975834.85%	5300688.32%	13136675146.53%	121
+20	13141975834.85%	5300688.32%	13136675146.53%	121

A difference is not a cause. Vaults enter crowds because they took a popular position, and a popular position going wrong is one explanation among several. What this rules out is the version where the crowd is a coincidence: that one predicts a difference of zero.

What this cannot tell you

We see on-chain and venue-side only. Anything held on a centralised exchange is invisible, so true crowding is at least as bad as shown, never better.
Coverage is usually low. Most instruments here show us holding a low single-digit share of open interest. That is a statement about our reach, not about safety.
Book clusters use single-linkage grouping. Members are chained by pairwise similarity, so a cluster average can sit below the joining threshold. It is transparent and arguable, which we prefer to a clever method nobody can interrogate.
Positions are snapshots. A strategy that flips intraday can look uncrowded at every observation and still be crowded in between.