Largemouth bass in cold clear water (13 °C): which lure gets eaten
The largemouth bass model is shown 11 generic lure archetypes, each tried in all eight colours, under one fixed set of conditions. 2 of 11 get eaten. The wakebait ranks first, eaten in 4 of 8 colours. The lipless rattle is outside what the Tank models and is listed unranked.
- Species
- Largemouth bass
- Water temp
- 13 °C · cold water
- Water
- Clear water
- Mood
- Neutral fish
- Depth · light
- 2 m · noon
- Retrieve
- 0.6 m/s
- Metabolism
- 25%
- Eaten
- 2 / 11
The ranking
Ordered by how far through detect → approach → track → commit each lure got, then by how strongly. Each lure is shown in its best-scoring colour. The verdict, reason and fix are the model's own words; the same text you get in The Fish.
- 01Wakebait120 mm · wobble 8.7
Eaten.
EatenTracked it, closed, and ate it. 31% commitment at 13°C on a 120mm profile.
Size and action are both matched to the forage base. This is the presentation.
- 02Deep diver110 mm · wobble 9.0
Eaten.
EatenTracked it, closed, and ate it. 30% commitment at 13°C on a 110mm profile.
Size and action are both matched to the forage base. This is the presentation.
- 03Pike bait150 mm · wobble 6.2
Refused at close range.
RefusedClose inspection, no commitment. Nothing is badly wrong; nothing is quite right either.
Loosen the action slightly, or present it where the fish has less time to look — cover, low light, dirtier water.
- 04Square-bill80 mm · wobble 5.4
Refused at close range.
RefusedClose inspection, no commitment. Nothing is badly wrong; nothing is quite right either.
Loosen the action slightly, or present it where the fish has less time to look — cover, low light, dirtier water.
- 05Medium diver80 mm · wobble 3.8
Refused at close range.
RefusedClose inspection, no commitment. Nothing is badly wrong; nothing is quite right either.
Loosen the action slightly, or present it where the fish has less time to look — cover, low light, dirtier water.
- 06Jerkbait120 mm · wobble 0.8
Refused at close range.
RefusedClose inspection, no commitment. Nothing is badly wrong; nothing is quite right either.
Loosen the action slightly, or present it where the fish has less time to look — cover, low light, dirtier water.
- 07Flat-side80 mm · wobble 1.0
Refused at close range.
RefusedIt got close enough to inspect, and the bait does not hold up to inspection — the action is stiff and the fall looks wrong.
Loosen the action slightly, or present it where the fish has less time to look — cover, low light, dirtier water.
- 08Trout minnow60 mm · wobble 1.0
Followed, then refused.
FollowedToo little action. At 1.0 it barely moves, and a bait that does not move reads as dead.
Loosen it: more lip area, or take out some low ballast.
- 09Micro crank30 mm · wobble 0.1
Saw it. Did not care.
IgnoredAt 30mm this is smaller than anything a largemouth bass is built to eat (60–180mm).
Go bigger — this reads as not worth the swim.
- 10Top-heavy fail80 mm · wobble 0.0
It never got a look.
Rolled overThe bait is lying on its side and skating. Fish ignore it because nothing in the water moves like that.
Move ballast below the centreline until the Tank says it swims.
- 11No-lip control80 mm · wobble 0.0
No driven action.
—This lure has no working lip and line tie pair to make it swim on a steady retrieve.
Add a lip below the line tie, then check the swim test in the Tank.
- —Lipless rattle100 mm
Not ranked
Outside the modelIts action comes from body shape, which the Tank does not model: it only scores wobble from lip drag, so it reads this bait as having no action. The fish model is never shown a fair version of it, so it is listed here and not ranked.
Why it came out this way
At 13 °C the model runs a largemouth bass at 25% of peak metabolism (optimum 27 °C, feeds between 7 and 34 °C), which caps its chase at 0.6 m/s. Every lure here is retrieved at 1.0 m/s, or just under the fish's top chase speed if that is lower: 0.6 m/s in this scenario. The page tests the lure rather than the reel.
- 5 of 11 got all the way to the fish and were refused on inspection: close enough to look, not compelling enough to eat.
The commit stage → - 1 of 11 never swam properly, so the fish model never got a look at them. That is a Tank verdict, the same in every scenario.
How the Tank decides a bait swims → - 1 of 11 were seen and ignored because they sit outside the 60–180 mm prey window the model gives a largemouth bass.
The approach stage and prey size → - 1 of 11 were followed and refused on action: outside the 4–9 wobble range the model gives this species, or too stiff to read as alive.
What decides a lure’s action →
Change one thing
The same largemouth bass, with a single condition changed.
This page is model output, not field data. The species values behind it, including temperatures, prey sizes, preferred action and reaction distance, are working assumptions that have not been calibrated against observations. The lures are generic archetypes, not products. Read where this model is wrong before trusting a ranking, and treat it as a hypothesis to test on the water.