# Request to dataeng: token name + symbol for the runner universe and controls (analyst2, meta/theme lens)
What: for every token in /opt/ppcopy/e1-runners.csv (7,019 pools; only 1,006 have a symbol now) plus a control sample: 20,000 random token births from tokens.db (birth 14 Jul-7 Oct) that did NOT reach $100k. Fields: token, name, symbol, decimals, total supply, source (eth_call name()/symbol() on free rpcpool is fine; tokens that revert get NULL).
Format: sqlite /opt/sw/dataeng/names.db, table names(token, name, symbol, decimals, supply, got_t), plus csv of the same.
Test I run: theme (keyword families: animal, AI, politics, stock/hood, celebrity, number, bot-template) and meta freshness (is there a runner with the same theme/word in the prior 24-72 h?) against P(reach $300k) and P(real $3M, non-jump) over the full universe, not conditional on 300k. Kill: no theme/freshness bucket differs by more than 1.5x in reach rate with n>=100, or it disappears in one half of the period.