MerisBot review
Our verdict
The most completely documented strategy in this index — MerisBot publishes the actual numbers its engine runs on, down to the entry band and the DCA steps, which is more than any hosted competitor here offers and more than enough to judge the approach before risking anything on it. What it does not have is a record. The 78% and the 130,468 trades belong to earlier builds, and the 130,468 is the same figure another bot in this index publishes as its own, so it corroborates nothing. The site itself is new, shows no live statistics, names no price and documents no key handling. It ranks fifth on the strength of its disclosure of method; it will move on the strength of a disclosed record, and it does not have one yet.
How it works
MerisBot works Polymarket's rolling 5-minute crypto up/down rounds across six correlated markets — BTC, ETH, SOL, XRP, DOGE and BNB, which it puts in a 73–85% correlation band. Its default Constellation mode runs an agreement scan at T+90 seconds, counting how many of the six have moved beyond ±0.03%, and requires at least five pointing the same way before it will act. It then isolates the one asset still flat (delta under 0.02%) and buys that laggard only while its Polymarket price sits between $0.49 and $0.55, on the premise that the straggler catches up before the round closes. A separate BTC & ETH Rules mode trades single nominated markets on user-set thresholds, bands and hours instead. Position sizing is a three-step DCA — $5 open, $3 added on a $0.09 slide, $2.50 on a further slide — capped at $10.50 per trade with buying halted 110 seconds before the close, and a BTC mean-reversion filter relaxes the opposing threshold after a 2%+ hourly run. Exposure is limited to three concurrent positions and $31.50 in total. Every position runs to resolution, and realised P&L is read back from Polymarket's closed-positions endpoint rather than estimated from the bot's own logs.
Strengths and weaknesses
What works
- Risk limits are structural rather than advisory: three concurrent positions, $10.50 a trade and $31.50 of total exposure are stated as hard numbers and as a percentage of bankroll.
- The entire strategy is published to the threshold — the ±0.03% agreement band, the five-asset consensus requirement, the 0.02% laggard test, the $0.49–$0.55 entry window, each DCA step and the 110-second cutoff. Most competitors describe their edge in adjectives.
- P&L is taken from Polymarket's settled closed-positions data, not projected from the bot's own fill log — an unusually specific commitment about where a number comes from.
- Non-custodial: orders settle on Polymarket through the user's own wallet.
- States plainly and prominently that it is an independent third-party tool with no Polymarket affiliation or endorsement.
- Six long-form strategy explainers cover the engine, the settings and how rounds resolve, so the product can be evaluated before connecting anything.
What doesn't
- The headline 78% hit rate and 130,468 trades are both attributed to "earlier builds", so neither describes the software currently shipping.
- That 130,468 figure is the same trade count 5MinuteBot (#3) publishes for its own prior versions — compare the two pages directly. Whatever the shared provenance, it is one dataset appearing under two brands, not two independent records.
- merisbot.com is days old and publishes no live record of its own — no current win rate, no running P&L, no dated stats.
- No active user or trader count published anywhere on the site.
- No price is stated on any public page. The site's structured data declares $0, which is not the same as the operator saying so where users will read it.
- No security or key-handling documentation: the non-custodial claim is asserted, not evidenced, and there is nothing describing how wallet credentials are held or exported.
- 5-minute binaries are the highest-frequency, highest-variance format on Polymarket, and the fee and slippage drag that decides whether a 78% hit rate is actually profitable is not addressed.