The accountable party behind each vault, with its whole cohort, the denominator, not the survivor.Note 18
Concentration
Operators known only by an address are folded into one slice rather than listed. That an address holds three hundred million is true and tells a reader nothing they can act on, and it would take a slice from a firm whose name means something.
| # | ||||||||
|---|---|---|---|---|---|---|---|---|
| 1 | curator firm | 51 | — | 45 | $2.39B | autonomous | steakhouse.financial | |
| 2 | 2 linked records | curator firm | 11 | — | 5 | $1.98B | autonomous | — |
| 3 | 2 linked records | curator firm | 67 | — | 43 | $1.52B | autonomous | gauntlet.xyz |
| 4 | protocol | 32 | — | — | $1.24B | unknown | app.concrete.xyz | |
| 5 | protocol | 4 | — | — | $380.98M | unknown | ether.fi | |
| 6 | 2 linked records | curator firm | 64 | — | 11 | $210.11M | autonomous | — |
| 7 | 3 linked records | protocol | 3 | — | — | $134.21M | unknown | app.avantprotocol.com/products |
| 8 | curator firm | 6 | — | — | $131.30M | model-assisted | — | |
| 9 | curator firm | 8 | — | 2 | $127.31M | rules-automated | — | |
| 10 | protocol | 56 | — | — | $114.40M | unknown | lagoon.finance | |
| 11 | curator firm | 5 | — | 2 | $110.12M | autonomous | — | |
| 12 | protocol | 4 | — | — | $109.35M | unknown | veda.tech | |
| 13 | curator firm | 10 | — | 6 | $106.13M | autonomous | — | |
| 14 | curator firm | 2 | — | — | $97.02M | unknown | — | |
| 15 | curator firm | 16 | — | 1 | $84.32M | rules-automated | clearstar.xyz | |
| 16 | 2 linked records | curator firm | 12 | 1 | 9 | $83.82M | autonomous | hyperithm.com |
| 17 | curator firm | 3 | — | — | $82.62M | model-assisted | — | |
| 18 | unknown | 12 | — | — | $82.01M | unknown | — | |
| 19 | protocol | 25 | — | — | $74.02M | unknown | ember.so | |
| 20 | protocol | 8 | — | — | $59.21M | unknown | app.upshift.finance | |
| 21 | KPK✓ | curator firm | 10 | — | 6 | $57.74M | rules-automated | — |
| 22 | protocol | 35 | — | — | $55.76M | unknown | app.ipor.io/fusion | |
| 23 | protocol | 10 | — | — | $45.75M | unknown | app.makina.finance | |
| 24 | curator firm | 11 | — | — | $45.60M | unknown | — | |
| 25 | 2 linked records | unknown | 8 | — | 3 | $41.21M | rules-automated | — |
The same firms measured twice: across the strategies still running, which is the figure anyone would quote, and across every strategy they launched. The gap is survivorship, the part of a record made of vaults that are no longer there. It needs nobody’s cooperation to compute, which is why it is not published anywhere a firm controls.
55 of 450 firms have closed at least one strategy, and 5 can be measured this way. The rest are firms too small for the difference to mean anything, or firms whose closures we cannot price; the largest being one with 48 closed strategies. That is a limit of this reading, not a finding about them.
| Firm | Launched | Closed | Survivors say | All of them say | Gap | Capital in the closed |
|---|---|---|---|---|---|---|
| HyperTwin | 4 | 3 | 121.0% | 32.3% | +88.8% up, survivorship gap | $0 |
| 0x146d…a1de | 19 | 3 | -1.0% | -17.5% |
| $208 |
| Hyperithm | 12 | 1 | 0.2% | 0.3% | −0.0% down, survivorship gap | $1 |
| Snipe trading Bear market | 10 | 9 | -43.7% | -27.2% | −16.5% down, survivorship gap | $0 |
| 0xf3ca…c606 | 6 | 3 | -54.4% | -31.2% | −23.2% down, survivorship gap | $355 |
50 firms have closed strategies that could not be measured this way: Hyperland (only 2 of 4 closed strategies can be priced, the unpriced ones would flatter this), Larp Capital_LLC (only 2 launched, too few for the difference to mean anything), Pulse@Evo (only 2 launched, too few for the difference to mean anything), VaultBot (only 2 launched, too few for the difference to mean anything), Dragon Portfolio (only 1 launched, too few for the difference to mean anything), XJ Systematic Dual (only 0 of 3 closed strategies can be priced, the unpriced ones would flatter this), and others. Listed because a firm whose closures cannot be priced is the case where this gap would be widest, and omitting it would repeat the deletion being measured.