Splash Signal: Runners logic enhancement💧
The primary focus lately has been one thing: higher quality calls, fewer rugs, and more control in your hands. Here's everything that changed recently, in plain English.
Runner grades (A, B, C) now actually mean something. They're a live read on the odds that a call really runs. A is highest conviction, B is solid, C is momentum-only and low odds. Higher grade, better chance it works out.
How a token earns its grade is the important part. It's not just about pumping. A token climbs toward A by SURVIVING, by staying alive over time without collapsing. Going up doesn't automatically raise its grade, and a normal dip doesn't tank it. These coins are volatile, we expect them to wobble.
What we punish is a fast, deep collapse. That goes straight at the worst calls: the fast rugs. If a token craters 50, 70, 90 percent within minutes of signaling, the system now sees it dying and drops the grade hard. The strong ones that hold their ground keep climbing instead. Survival is the tell.
Tokens can also get upgraded. If a runner starts as a B and then proves itself by holding and strengthening, it gets bumped up to A and you get a short follow-up ping that it leveled up.
We didn't just guess at this. We backtested the new grading on two weeks of real signals. A-grade calls hit winners at roughly double the rate of the old blended average, and the C-grade calls (the ones we now flag or filter out) were almost all losers.
We're continuing to monitor the data, backtest, and iterate regularly. With more time and a larger sample to work from, accuracy and quality only get sharper from here. We’re just getting started.