Risks

Most of what the bot buys dies

97.3% of the 9,221 tokens bought no longer have a live pool. This is not a failure mode — it is the normal case, and the strategy is built around it. It becomes a problem only if the survivors stop paying for the rest.

The result depends on a few positions

1,764 of 9,119 closed positions made money. A book this concentrated can look very different after one bad stretch, and a short observation window cannot tell you how deep a bad stretch goes.

The window is short

26.86 days is not a sample. Every figure here describes what happened between 2026-08-31 and 2026-09-27, and nothing about it forecasts the next period.

Most attempts cost money and land nothing

unknown of transactions revert, burning unknown ETH in gas over the window. If competition increases or block space gets more expensive, that cost rises without any change in strategy.

Open positions may not be sellable

37 positions are open and 9 have no indexed pool at all. A mark assumes a pool deep enough to exit into; on tokens this small that assumption is usually wrong. Treat marked value as an upper bound.

Capital does not scale the result

Covered in full on the ceiling. More capital cannot simply mean a bigger stake.

The token adds risks of its own

No contract exists yet, so nothing is audited. The payout rule is undecided. Paying profit to holders is treated as a security in many jurisdictions and has not been legally reviewed. A losing period pays nothing.

The record is not audited

Every figure is reconstructed from public chain data by our own pipeline. The method is documented and the raw responses are kept, but no third party has verified it. The balance reconciliation currently reports not run, because public RPC on this chain refuses historical balance queries.