Creating automated strategies for Polymarket, everyone is very cautious when deploying, but when disabling, hardly anyone talks about it — today I learned this lesson the hard way after 14 hours.

Three weeks ago, I stopped a strategy bot running on Polymarket. pm2 stop, saw the status turn to stopped, and thought the issue was resolved.
Today, I casually did a 10-minute server cleanup and restarted the process manager. That "dead" strategy from three weeks ago was revived and quietly ran for 14 hours in real money mode.
I checked the root cause — pm2 stop only marks the status as "stopped," it doesn't actually delete the process. As long as a batch start is triggered once, all "paused" processes are brought back.
This trap isn't unique to pm2; systemd, Docker, k8s all have the same problem: thinking they've stopped, but in fact, they haven't.
When building automated strategies for prediction markets, no matter what tools you use, stopping a strategy involves 5 steps:
1. Completely delete the service at the process manager level — (not just pause)
2. Remove its definition from the configuration file — (otherwise, it will resurrect itself during the next batch start)
3. Add a shutdown gate at the code entry point — (to prevent accidental wake-up)
4. Update your documentation or status table — (otherwise, in a couple of weeks, you'll forget whether this strategy was stopped)
5. If it involves a trading ledger, officially close the accounting cycle — (epoch), to prevent new data from mixing into the old strategy's records
Everyone is very careful when deploying, but often only do step 1 when disabling, considering it done. The real pitfalls are buried in steps 2 to 5.
14 hours of sample time didn't cause a loss, but this luck doesn't mean you'll be so fortunate next time.