Beware of Claims About Science and Algorithms With Solunar Fishing Scores

Let’s Start With the Big One - “Science-Washing”

We have been conditioned to trust scientific claims. Many don’t question the basis. In fact a study published in the Journal of Experimental Social Psychology found that " people who trust science are more likely to believe fake claims if those claims use scientific references." Study is here

That feels almost contradictory, using a science-backed claim to argue against scientific claims. Except, it isn’t. This is peer-reviewed 1 so it’s not just a claim made about a product. There’s real, established scientific method involved in arriving at the actual conclusion.

So when a fishing tool uses the phrases “Scientific”, “Science” and “Algorithm” to support their “forecast”, this is pseudoscience. Not science.

Ask yourself. Where’s the evidence? What controls were used to support the evidence (if any)? Was it “peer reviewed” or “pal reviewed”? The former is real scientists checking the data; the latter is your friends telling you “She’s great, mate!”

On Getfished I have avoided doing this and said so whenever I can. It bothers me so much that I’ve placed explanations and links on every fishing forecast page. Sure, I use solunar when I fish; no, I can’t prove it works. At this point, there is no scientific evidence to support it and much to show it depends on a myriad of factors that can’t be predicted. It is, at best, anecdotal and has remained that way for almost one hundred years.

Why a Single Fishing Score Can Look More Certain Than It Is

A single fishing score can be appealing.

It is quick to scan. A number such as 75, labelled Great, can make a complicated decision feel simple: conditions are good, so go fishing.

There is nothing inherently wrong with summarising conditions. A score can be a useful way to bring together weather, tides, daylight and other inputs.

The problem begins when a score looks more certain than the evidence behind it.

A fishing forecast or app may show something like:

That looks precise. Some throw in stars. But unless the formula has been tested against real fishing outcomes, the result is still a weighted opinion about conditions rather than a measured probability of catching fish.

So you need to ask - where is the data to support “Great” in the forecast? On Getfished, you will see percentages. Those represent the solunar scoring according to the tables and formula used by John Alden Knight as outlined in his book “Moon Up Moon Down.”

It’s a pattern, it’s an algorythm, it’s widely used, but it is not proven science. Getfished uses them to answer the question “What’s the solunar forecast for Melbourne today? " Or some other location you might be interested in, hypothetically. It’s not a guarantee you’ll catch fish. It’s more a statement of “Here is your answer according to solunar theory”, but you decide if it’s something you want to use or disregard.

This distinction matters because fishing is not one uniform activity. Conditions that help at one location, for one species using one technique, may be poor when the same technique is used at another time or location. Indeed, there are so many factors that vary across locations and time, it’s impossible to actually create a representative controlled experiment that won’t produce outliers and contradictions.

An Algorithm is Like Statistics, They Can Be Made To Say Anything

We have all heard the saying “Damn lies and statistics.” Algorithms are no different. They are just tools that use a pattern to achieve a desired outcome; statistics are similar. Of course, there’s some complex math used in both. Often very impressive. But the fact is, fish are living organisms. Fish are affected by other environmental factors that are not predictable. We cannot know where bait fish will congregate beyond very broad general terms. These kinds of “inputs” cannot be guaranteed and rely on historical data collection, which can change due to factors not in the model.

Pollution being a case in point. Good luck catching fish on a “Good or Excellent” day if a Blue Green Algae outbreak wiped out the fish population two weeks before and you had no idea about it.

A score is a summary, not proof

A fishing score is often created by assigning points to inputs such as:

• wind speed;
• tide movement;
• tide range;
• moon phase;
• barometric pressure;
• rainfall chance;
• time of day;
• swell;
• water temperature.

The score may then add those values together and produce a label such as Poor, Fair, Good or Great.

That can be useful as a broad summary. But the score itself does not prove that fish will bite, or even if there are fish present in the waterway you are fishing. What about structure? No structure means many species will be somewhere else, even if that somewhere else is 500 metres or kilometres from where you are fishing.

For a number to become a meaningful prediction, it needs to be tested against actual catch outcomes over time. That means comparing the forecast conditions with what was really caught, where it was caught, when it was caught, and how often similar conditions produced similar results.

Without that validation, exact-looking weights and multipliers are still choices made by the person building the model. Opinion, perhaps, but not fact and not scientific.

That’s not “science”; that is the very definition of “pseudoscience”. Why? Because, using that method, you can claim the world is flat, the sun revolves around the earth, or that elves and pixies are real.

In programming, we call this “GIGO”, which stands for “Garbage in, Garbage Out."2

The false precision problem

There is a big difference between saying:

Light wind, a building tide and a dawn bite window may be worth considering.

and saying:

Conditions are 82 out of 100. Rating: “Good Fishing.”

The first statement makes room for uncertainty. It presents useful context without pretending that a particular combination of inputs guarantees a result.

The second can imply a precision that may not exist. it is puffery 3 at best and can mislead at worst. It’s fair to say nobody believed it when a well-known energy drink “…gave you wings!” Not literally.

A score of 82 rather than 78 suggests that the model has found a measurable difference between those situations. But that is only true if the numbers were derived from real-world testing and repeatedly shown to correlate with better fishing outcomes.

Otherwise, the difference becomes arbitrary scoring choices.

Indeed, the actual peer-reviewed studies that have been done on solunar have been inconclusive at a minimum 4.

Overlapping inputs can be counted twice

One of the easiest problems to create in a scoring system is double-counting.

Moon phase is a good example. In tidal waters, the moon can influence tidal range. New and full moons are often associated with larger spring tides, while quarter moons often produce smaller neap tides.

If a model says moon phase matters mainly because it affects tidal amplitude, then separately scoring:

• moon phase;
• tidal range;
• tide direction;

can give the same underlying effect more than one influence on the final result.

That does not mean moon phase, tide range or tide movement are useless. It means the model needs to be clear about whether it is measuring separate effects or repeatedly rewarding the same one. Otherwise it tends to imply a relationship beyond the actual data.

The same issue can appear with weather factors. For example, pressure trend and proximity to a weather front may be closely related. If both are scored heavily, the model should explain what separate information each one contributes.

There are species, like Australasian Snapper that will often come closer inshore during a strong weather event like a storm to feed. Making it possible to catch snapper bigger than you usually find from a bayside pier. Whereas species like Garfish seem to disappear and Calamari move into deeper water.

Clock time is not the same as daylight

A fixed rule such as:

Dawn: 4:00 am–8:00 am
Dusk: 5:00 pm–8:00 pm

is easy to program, but it is not a reliable substitute for real daylight conditions.

Sunrise, sunset and twilight move substantially through the year. What counts as dawn in summer is not the same as dawn in winter. The useful period may also change with:

• species;
• water clarity;
• habitat;
• cloud cover;
• depth;
• whether the angler is fishing an open beach, deep pier, shaded river bend or shallow flat.

A better approach is to show actual sunrise and sunset times, then present them alongside other relevant conditions rather than assigning every day the same fixed clock bands.

Time Zones and Daylight Saving

This one is a huge factor that seems to throw a lot of websites and apps completely off. Most of us know that the world and indeed some countries are divided up into time zones. There is anything from an hour to several hours difference between them. Some states make this more complex by having Daylight Saving Time. It surprises me how many people fail to take this into account when publishing “times” like sunrise/set, moonrise/set, tides, etc. Indeed, I have seen these problems begin to compound over successive days.

Finally - the track of the moon across hemispheres. While it’s reasonable to say the moon’s phase is seen globally at the same, equivalent times. The track of the moon is different between the northern and southern hemispheres. It doesn’t orbit the equator. The Earth is not a smooth, even sphere. Some of these factors modify moon times slightly and need to be taken into account.

Wind Direction and Fishing Rules of Thumb

These “rules of thumb” suffer badly in the Southern Hemisphere. In the Northern Hemisphere, a northerly frequently brings cool or even freezing conditions. In Australia, for example, it’s the opposite. So who is right when the saying says “Wind from the North, do not go forth” when some of the best fishing in Victoria can be had during the summer, northerly wind prevalent months?

The “East, Fishing is Least.” To generalized. It fails to take into consideration sheltered areas, protected from wind, as so many of these “rules” do.

What matters more is the location you are fishing in vs the prevailing wind. Can you fish it in comfort? Are the fish present? Are they feeding?

Rain chance is not rainfall intensity

Rain probability is often easy to obtain from weather APIs, but it should not be confused with how much rain may actually fall.

An 80% chance of a few drops is not the same thing as heavy rain.

Likewise, a 30% chance of a severe thunderstorm can matter more to safety and comfort than a higher chance of drizzle.

Rainfall amount, storm risk, wind, lightning, river flows and local access can all matter. A simple rain-chance score may be convenient, but it can hide the difference between a wet afternoon and a dangerous weather event.

Local assumptions do not travel well

A model calibrated around one estuary system may be useful there. It should not automatically be treated as universal.

A tidal range that suits a sheltered Sydney estuary may not suit:

• Port Phillip;
• Western Port;
• Victorian surf beaches;
• freshwater rivers;
• impoundments;
• alpine streams;
• Murray Cod water;
• a jetty in deep water;
• an estuary entrance with different flow and structure.

The same applies to wind.

A gentle onshore breeze may improve a surf beach by stirring water and moving bait. That same breeze can make a kayak trip uncomfortable or unsafe. A stronger wind may help cover an angler’s presence in one protected estuary but make casting impossible from an exposed pier.

There is no universal wind score that understands all of those situations.

Important factors are often missing

Most fishing-condition models are built around fields that are easy to collect from public weather, tide and moon data.

But some of the most important influences are often harder to measure consistently:

• water clarity and turbidity;
• freshwater flow and river height;
• bait presence;
• water temperature at depth;
• local structure;
• access conditions;
• seasonal closures;
• crowding;
• recent fishing pressure;
• location-specific safety issues.

A score may look complete while omitting factors that matter greatly at the location being considered.

That is not an argument against forecast data. It is an argument against treating a single number as the whole answer.

What a more useful fishing forecast looks like

Fishing forecasts are most useful when they help anglers make their own informed decision.

That means showing the underlying context:

• current weather and wind;
• tides and tide movement where relevant;
• solunar activity and bite windows;
• actual sunrise and sunset;
• pressure and rainfall context;
• species guidance;
• bait or lure context;
• recent fishing report patterns;
• historical report patterns;
• access, safety and location notes.

A rising tide may be useful at one location and irrelevant at another. A rough-looking day may be productive for one species and poor for another. A calm, sunny day may score highly but still produce little if bait is absent, water is dirty, or the fish are simply elsewhere.

Recent reports do not guarantee a catch either. But they provide something a generic score cannot: evidence that anglers have actually been catching particular species in comparable places.

Forecasts should assist, not command

The honest role of a fishing forecast is not to say:

The score is 82. Go now.

It is to say:

These are the conditions. These are the likely opportunities and limitations. Here is what recent local evidence suggests. Use that information to decide whether the trip is worthwhile.

Fishing remains uncertain. That is part of the appeal.

The best fishing information does not remove that uncertainty with a bright number and a confident label. It helps anglers understand the conditions, compare them with local evidence, and make a better decision about where to fish, what to target and when to go.

Final Word

The biggest factor leading to fishing disappointment isn’t the moon. It isn’t the tide or the weather. It’s a lack of knowledge on what species to target, when to target them and what the most effective methods are to target them. Even the fishing celebrities and fishing influencers on YouTube have “doughnut days” where they don’t catch anything. The difference is they have developed enough skills to reduce those days. Whether it be the appropriate tackle for a species, the bait or lure they are confident in using, or the location and conditions they have learned produce fish.

For myself, I have caught fish time and time again at a pier when the people fishing before me said there was “nothing there today.” Berley, bait, hook size. Things you adjust through trial and error. Anybody can do that; it’s a matter of putting in the time.

Share this post