While running backtests recently, I found the data keeps “stuttering.” At first, I thought my own script was the problem. After troubleshooting for a long time, I realized it was Subgraph’s indexing speed that couldn’t keep up with the RPC requests. To put it simply: you push the requests over, the other side rate-limits and blocks you, you sit there waiting, and then the data ends up with gaps. It’s pretty annoying—especially for high-frequency strategies; being off by even a second can throw everything off.

Lately, I’ve been watching those new L1/L2 rollouts offering incentives. Even longtime users are complaining about “mine, withdraw, and sell.” People mine and run, and the data stream becomes even harder to track. Anyway, I’ve gotten more cautious: I’m going to strengthen the redundancy of my data sources first, so the backtesting environment doesn’t get misled by this kind of latency. Forget it—I don’t know if I’m just overthinking it. In any case, with this current market, staying quiet probably can’t hurt.