Banks Lake is twenty-seven miles of open water in the Grand Coulee, and in late June it holds some of the best smallmouth fishing in Washington. Post-spawn fish are aggressive, the basalt talus is stacked with crawfish, and the rock-to-depth transitions are about as textbook as smallmouth structure gets.
None of that changed between Sunday and Monday. The verdict did.
The call
Eighty-nine is the highest score the engine has returned on home water. It clears the Outside Edge threshold, which is reserved for days worth rearranging a schedule around.
Three things stacked. Fish availability and catch history both scored near the top — Banks is a renowned smallmouth fishery and this is its season. Run timing came in at nine of ten for peak post-spawn. And critically, the calm wind left launchability and boat handling almost untouched, which is what lets a big open lake score at all.
The day before was a different story.
Here is the part worth sitting with. On the numbers alone, Sunday still scored 75 — a comfortable Go. The fish were just as available, the water just as proven, the timing just as good. Degrading the weather, launch, and boat-handling factors wasn't enough to pull it under the line.
It was the safety rule that stopped it.
That asymmetry is deliberate. A scoring system that merely docks points for dangerous conditions will always, eventually, produce a high enough score on a day you shouldn't launch. Making safety a separate veto is the difference between a tool that optimises for fishing and one that optimises for getting home.
The plan for the good day
Steamboat Rock's points and the island structure nearby, first — rocky points with quick access to deep water, and the sort of rock-to-sand edges smallmouth patrol at first light. Topwater early, then tubes and football jigs dragged along the west-shore talus as the sun climbs, then a drop-shot out on the mid-lake humps in twelve to twenty feet once fish slide off the bank.
The calm day matters for the finesse presentations specifically. Drop-shotting a hump requires boat control you simply don't have in a twenty-knot wind.
On the water early, and started hammering really large smallmouth right in front of Steamboat Rock — especially around the large submerged boulder toward the south end.
Ned rigs with green or blue worms were good, but bright chartreuse jig heads with about half a nightcrawler was the ticket.
Then we noticed the shade line. As the sun rose it moved closer to shore. If we were outside the shade line, no bites. Inside was gold.
As soon as the shade line hit the shore, everything buttoned up. No more bites.
We were able to find some smaller bass that would eat Senkos up in the coves to the north. By 12:30 the bite was dead.
Scoring the call
Eighty-nine earned it. The day was as good as the engine said it would be, and the spot was right — Steamboat Rock is where the plan sent us, and Steamboat Rock is where the fish were.
Two things it got wrong, and they are the same two things it keeps getting wrong.
The technique. The plan called for topwater early, then tubes and football jigs, then a drop-shot. What actually worked was a Ned rig, and then a chartreuse jig head with half a nightcrawler — bait, which the engine does not consider at all. It scored the day correctly and then handed over a lure list that missed.
The clock. The engine returns one number for a whole day. This day was over at 12:30. And it wasn't over because of the hour — it was over because the shade ran out. The bite tracked a moving line across the water and stopped when that line reached the bank.
The second one is the more interesting failure, because shade is not a mystery. Where the sun sits is arithmetic. A model that knows the solar angle, which way a bank faces, and how tall the rock standing behind it is could have said fish the shaded side until about half nine — which is the sentence that would actually have been worth reading at five in the morning.
That one is now on the list.
Scores from OLPNW-FIE™. Conditions from the National Weather Service. The engine currently scores from expert-assigned inputs rather than a live data feed — always confirm conditions and advisories yourself before you launch.