Blank Days Belong in the Log

Most logbooks are trophy shelves.

You photograph the fish. You enter the length. You keep the good days and quietly forget the rest. The app learns that you catch fish, because that is all you gave it.

That record is incomplete. The days you drove, fished, and caught nothing are the other half of the pattern.

What a blank actually is

A blank is not a failure of technique by default. Often the water was wrong before you left the house. Cold water after a warm spell. Dirty water after rain. A tide that moved bait off the mark. A species that had already shut down.

If those days never enter the log, the history is a highlight reel. Highlight reels make poor forecasts.

GILLI treats a blank day as a first-class session: same water, same species target, same conditions, no fish. That is the denominator. Several fish from one session still count as one productive day. Otherwise a lucky hour inflates the rate and a long blank disappears.

Why the score needs both

Personal history should only move a forecast when there is enough of it, and when the days are comparable.

Comparable means the same sort of water, the same sort of target, similar conditions. Ten carp from a stocked pond do not tell you much about a wind-blown pike lake. Three blanks on the same mark, on similar tides, do.

When the sample is thin, the forecast stays generic. That is not a bug. It is the honest state. The useful prompt is which day would actually change the picture: another blank on this water, or a catch in these conditions.

Apps that only celebrate fish cannot do that. They have no rate. They have a pile of trophies.

What we changed

In January we stopped treating sessions as catches with optional notes. The log is catches and blanks. Success is day-level. The forecast can use that history when enough matching days exist, and it stays quiet when they do not.

You can log a blank without a photo, without a fish, without pretending the day was something else. The conditions still attach. That is the point.

What this is not

It is not a promise that logging will make the forecast more accurate. Some waters never gather enough comparable days. Some species move too much. A blank caused by a snagged lure is not the same as a blank caused by a cold front.

It is also not a guilt mechanic. There is no streak. Nobody is counting how often you go. The blank is there so the next call on that water is less of a guess.

If you only log the fish, you are training the app to flatter you. Log the blanks too.