ルアー · カーペンター · ヒラマサ

BF caught fish on 16 trips and not one over 20kg — what 378 days of lure records say

2026-08-08 · fishingmap.fun

First, let me write down what this article will absolutely not do. It will not give hit rates for lures. Because they cannot be produced.

While counting ten years of lure records I noticed this. I write the name of the lure only when I caught something. There is no motive at all to record "today I threw the γ250" on a day I caught nothing. So if I take "trips with a lure name" as the denominator, the hit rate of every lure jumps.

The numbers show it directly.

A gap of 23 points. This is not a difference in lure performance, it is a difference in how the records were kept. In the first draft I wrote I missed this and wrote "this lure has a hit rate of 100%". Before counting, I should have doubted what I was counting.

The background of these records

There were 161 lure casting trips, 86% of them expeditions, and 378 actual fishing days in total. It began in December 2015, in the Torres Strait, Australia (8 GT (giant trevally), 5–15kg). So this article is a story about ten years. A little over 7 out of 10 destinations were Nagasaki and Okinawa. It is fishing where I fly to a remote island and have three days on which I put a rod out (travel days are not included).

Most of the lure records are from expeditions, while in the Kanto area it is mainly komase bait fishing and mid-depth bottom fishing. So in practice this article is a story about lure selection on remote-island expeditions.

And — one expedition carries a flight, a boat, lodging and three actual fishing days on it. Choosing a lure becomes a judgment about whether or not to throw those three days away. That is the premise for everything below.

The tackle I take — a three-tier setup

Before counting the records, let me write down what I am actually taking with me in the first place. This is not statistics, it is the angler's own account as given. The setup is more or less fixed; I do not rethink it for each expedition.

That the main tier is six rods says a lot about the character of this fishing. They are not spares. Six rods of the same class are carried because breakage and lines parting high up are taken as given. In fact, the angler himself summed up 2025 as "the year of high line breaks".

For reels, the main tier is all Calcutta Conquest 401XG. I wrote about bringing the line retrieve per crank to 101cm in the article on line retrieve per crank. The 201XG in the light tier is smaller in diameter still, so it is a choice that takes that same logic one step further.

From here on, it splits into two layers

As I was writing, the angler himself told me plainly.

These are the basic lure choices, and the lures I take with me. Using them differently by sea area is a separate matter.

So there are two layers. What goes into the box, and what gets thrown out of the box.

This difference is the limit of this article itself. The loadout is a bet made before I know what the sea will do, and the selection is a judgment made after seeing what is in front of me. Only the results of the latter stay in the records, yet the range I can choose from is settled in advance by the former.

I will write about the loadout first. This part is fixed.

The box I pack changes with season and fish

Rather than rethinking it for every expedition, it felt more like there are four modes and one of them is chosen.

Hiramasa (yellowtail amberjack) in spring

Hiramasa in autumn

I hardly take anything 100g or under. For the same fish in the same sea, only the lower limit of the box goes up. It is the spring lineup with the small side taken out.

GT

Added to the spring lineup are the various Carpenter sinking lures and the Snecon and Gachi-sura. For poppers it is the Damsel and Utahime, and in between those sits the Garam.

When I am possessed by clear

Now and then I get possessed by the darkness of "they won't bite unless it's clear", and go with the Bocchu Hiramasa as the main lure,

in Shimano-type clear colors.

That this mode exists is, I think, the honest story. The box changes not on data but on whatever I am convinced of at the time.

When I do short-twitch fishing

Added are lures worked with short twitches, like a horizontal tairaba. They are the local standard for that sea area, and even the angler himself does not know the brand names. They are recognized in units of "what everyone locally uses", in other words.

Chosen by size and action rather than by brand

I still do not know the names of the short-twitch lures. Even for the GT popper, it took a moment of thinking after being asked before "it was the Garam" came out. The angler himself, after ten years of throwing them, cannot immediately say the names of the lures he actually throws.

This is a hole in the records, but at the same time I think it is also the answer. What is being chosen is not a product but a size and a type of action. Popper or sinking, how many grams, clear or not. The brand is no more than a means of meeting those conditions.

So this article cannot answer "which lure catches fish". It is not that it cannot answer, but that the angler himself is not looking there.

This lineup explains the tallies below exactly as they are

BF and γ are packed as a band from the 30g class up to 160g. The Hayabusa is a single lure, the 120.

That means BF and γ can be thrown in any situation. The Hayabusa appears only on the main tackle, and only in one specific scene.

Later in this article a table appears saying "BF caught on 16 trips with zero over 20kg, the Hayabusa 34.8kg on its one trip", but this is what that gap really is. Not a difference in performance, but a difference in how many are carried and in the band. It is not something I noticed by counting; the way of packing was like that from the start.

And it also matters that there is a separate slot for "250g as a change of pace". The 250g lures are not in the front-line lineup. It is a slot for throwing them because I felt like throwing them.

What I threw and what I landed

Hit rates cannot be produced, but the share of the trips that produced a catch does mean something. Trips where the lure can be identified number 44.

BF and γ account for 6 out of 10. I myself thought "the BF60 is the one I use most", and as a share that turned out to be exactly right.

This is the real point

Only "whether that trip produced a fish over 20kg" is a measure with relatively little recording bias. Because the force of "a big fish gets recorded" applies equally to every lure.

BF has caught the most fish, yet the number of trips with a fish over 20kg is zero. The Hayabusa has only one recorded catch, and that one time was 34.8kg and 26kg.

Reading this as "the Hayabusa is the better lure" is wrong. The scenes they are deployed in are simply different. BF gets thrown every time, so the numbers pile up. The Hayabusa appears only at the single point of "going for a big fish at Goto (Fukue Island) in spring" — in fact the two fish landed on the Hayabusa, 34.80kg and 26kg, came from the same expedition.

The way they are used differently has taken the shape of the numbers directly. Without realizing it I had separated "the everyday lure" from "the lure for the moment that counts", and ten years later the records told me so.

The single day that has a denominator

In ten years there is exactly one day on which "the lures I threw that did not get eaten" are written down. Goto, December 2024. I quote from the post at the time.

Matching the lure of the Iriomote skipper who drew the first bite, I used the SE150 single barbless to the full. I also threw the Hayabusa, BF and the γ250 and 350 a fair amount, but there was no response.

On the day the three ace series were all shut out, only the SE150 produced 28, 24 and 13kg (the total over the three days 2024-11-30 to 12-02).

It is the experience of this day that makes me think of lure rotation as "the work of changing the vibration". Rather than throwing them in order of preference, when they see through it, change the series. That is why I carry several series.

However — the only record supporting this is this one case. On other days "the lures that did not work" have not been kept, so there is no way to verify it. I believe it as a piece of reasoning, but as data it is one case.

The other layer — selection on site

From here it is what gets thrown out of the box. With the caveat "it varies from time to time" attached, the axes of judgment still came out in words.

Waves — chosen by how little they skip out

There is one axis: how little it skips out. When the waves are high the lure jumps out of the surface. I choose ones that do not jump out.

Here I corrected one point in my own understanding. In old notes on this site the same lineup was written under the condition "fast current, tide waves present", but when I checked with the angler himself: "as for the speed of the tide, high waves is actually the right way to put it". What he is watching is not the tide speed but the sea surface. He does not choose because the tide is fast, but because waves are standing up.

He did attach a proviso here, though. Since this is a matter of how you handle line slack, it varies from person to person, he said. The SE can be quite hard to work depending on the conditions, apparently. In other words, even for the same lure, where "hard to skip out" sits shifts with who is throwing it. Because of that one remark, please read the table above not as "the correct answer" but as "in this person's case".

Baitfish — matched only when they are big

When the fish finder can be watched, if the baitfish are up, they will eat anything. On top of that, the Maihime and SE are often chosen deliberately, apparently.

And this was the most unexpected part.

I don't pay much attention to baitfish size (when doing sasoidashi (topwater teasing for tuna)).

But when the baitfish are big I throw a big lure. Throwing a small lure just because the baitfish are small is rare (nabura is another matter).

The matching only goes one way. He matches upward, but not downward. If you did "match the bait" straightforwardly you would match in both directions, but that is not what is happening.

When there is no response — bigger and heavier

I often escape to 160g and above, to poppers, and to sinking lures. For GT especially, sinking.

In the loadout section there was a slot for "SE250, γ250 and Utahime250 as a change of pace", but that is less a change of pace than a bolt-hole for when things are stuck, I think.

Sinking lures are dropped two at a time, in alternation

Of the three — Pandora, Zeus and Darting Rapid (proper name unknown) — I tie on two and drop them in alternation.

When that still does not work, I drop the Gachi-sura.

Rather than trying one at a time, two are run in parallel and put in alternately. This is the only place where the procedure is concrete. And the Gachi-sura comes last — the order in which something other than a Carpenter comes out is fixed.

Not overlapping with the others on board — making the boat a single experiment

This was the most interesting axis.

When there are others on board, I deliberately avoid that size. When it is all pencils, a popper;

when it is all small lures, a big lure. If someone gets a response, I move over to that for the time being.

The boat is being used as a single piece of experimental apparatus. If everyone throws the same thing, only one piece of information comes out of that day. Spread them out, and the hit can be searched for as a division of labor across the boat. When it comes, I move over to it. (Whether the others move over is unknown. The angler himself said "?" to that.)

However, this is the theory for when there are a lot of people. Normally about three people are casting at the same time, so there is no overlap to begin with. It is a story that takes effect on a boat where 7 to 8 people are throwing at once — a Korean expedition, for example.

And on such a boat, the choice of lure depends on "what the person next to me is throwing". The choices are not independent. The impossibility of measurement in this article comes not only from how the records are kept (writing only when I catch), but also from the fact that on expeditions with many people the choice itself is settled from outside.

This does not extend to all the records. On most expeditions, with three people on board, this effect is not at work — let me write that accurately.

The combinations that remain in the records

The above are the axes of judgment; below is what remained in the records as a result. It is not "what worked".

Only the Damsel and γ have produced fish over 20kg for both hiramasa and GT. The Damsel: 30kg at Iriomote, about 20kg in Sagami Bay, 16kg at Tsushima. It is a lure that crosses both sea areas and species. Being light and easy to work, it can be thrown continuously for long periods, so its coverage ends up being wide — that is how I read it.

And both of the 13kg kue were landed on the BF60. For fishing that lifts a bottom fish up from 20m of water on a topwater lure, no other lure appears. I cannot explain the reason, but as a fact that is how it is.

What I will record from now on

With the way I do it now, lures can never be compared. Because only the ones that caught fish are kept. So I am changing the form of the records.

Thrown: Hayabusa 120 / BF140 / γ250
Ate it: SE150

If these two lines are written separately, then for the first time "thrown but not eaten" can be counted. The record of December 2024 is special only because that happened to be how it was written in the original post. If I keep it up for one season, by next year a real comparison should be possible.

---

Catches for which the lure can be identified come from 44 trips, 27% of the 161 casting trips. Every figure given can be recalculated with data/author/fishing-profile/verify_stats.py. What is counted is not the number of fish but "the number of trips on which that lure name remains" (several fish on one expedition still count as one case). Fish landed by companions are excluded from the tallies (the (他者) marker in the catch field). Blank days are unlikely to be recorded, and when several lures were used on the same trip it is in principle unknown which fish came on which lure (the March 2024 case is an exception where the original post made rod, lure and fish clear one to one: 34.8kg and 26kg on the Hayabusa 120, 18kg on the Bocchu Hiramasa 190). Casting began in December 2015 (Torres Strait, Australia; 8 GT) and the records before that are bait fishing and mid-depth bottom fishing. Up to 2018 the lure entries are especially thin, so this article is in practice the picture from 2019 onward. Some of the denominators for fish over 20kg are single digit, so the order can be read but the ratios cannot.

For the general characteristics of the products, please see each manufacturer's material. What is written in this article is only the range that appears in the operator's own records; it does not include the results of others or results published by manufacturers.

2026-08-08 addition. Based on the operator's own account, the three-tier tackle setup and the four ways of packing that change with season and species were added. Without "what is being taken along" before the tallies, the share figures end up looking like differences in performance. At the same time it was made explicit that "the loadout (what is taken along)" and "the selection by sea area" are separate layers — the former is a bet decided before leaving home, the latter a judgment made on site. Only the brand names of the lures for short-twitch fishing are still unknown.

2026-08-08 further addition. The axes of selection (how little the lure skips out on the waves／matching the baitfish only when they are big／when stuck, bigger and heavier／sinking lures two at a time in alternation／not overlapping in size with the others on board) were added based on the operator's own account. With the axis of choosing depending on the others on board coming out, the reason why the relative merits of lures cannot be extracted from these records became one step clearer. The brand name of one sinking lure ("Darting Rapid") is unconfirmed. The lures for short-twitch fishing are recognized only in units of "the local standard for that sea area", so their brand names will not be pursued.

2026-08-08 correction (3). The "13 years" in the article was wrong. The 13 years applies to the fishing records as a whole (from July 2013; including bait fishing and mid-depth bottom fishing), and lure casting has been 10 years, from December 2015. The 5 relevant places were fixed. The source of the figures appears in the 【期間】 block of verify_stats.py.

2026-08-08 further correction. ① The description in the old notes, that a small lure is matched to extremely small baitfish, was wrong (confirmed with the angler himself; in fact it is not done). ② The condition "fast current, tide waves present" is, in fact, correctly read as "high waves". What is being watched is not the tide speed but the sea surface. ③ The Gen'ei is sometimes used when the waves are high. ④ Not overlapping with the others on board is a theory for a boat with many people, where 7 to 8 are throwing at once; with the usual three or so there is no overlap. ⑤ Only I move over to a hit; what the others do is unknown.

TideThis month's tide — read in Korean 물때 and a 3-day window→