Ask ten trollers what the most important variable is and most of them will say depth. They are wrong, or at least they are answering a question about the tool rather than the fish.
Depth decides whether the lure is in front of anything. Speed decides whether the thing in front of it is worth eating. You can be at exactly the right depth all day with a lure that is not working, and the water will look empty.
Your GPS is measuring the wrong thing
Start here, because it is the mistake that costs the most and almost nobody makes it deliberately.
A GPS measures speed over ground. It tells you how fast the boat is moving relative to the bottom of the lake. That is a genuinely useful number for getting home and a nearly useless one for a lure.
The lure does not know about the ground. The lure knows about the water it is sitting in, and the water is moving too. If there is a current running with you at half a mile an hour, your lure is going half a mile an hour slower than your GPS says. Turn around and troll the other way and the same GPS reading is now half a mile an hour too slow. Same number on the screen, two completely different presentations.
On big water this is not a small correction. The Great Lakes are famous for subsurface currents that run in a different direction from the surface, so a boat can be pushing into a light surface breeze while the water eighty feet down is going somewhere else entirely. Two rods in the same spread, one shallow and one deep, can be fishing meaningfully different speeds at the same time.
There are three speeds and it is worth knowing which one you are looking at. Speed over ground, from the GPS. Speed through the water at the surface, from a paddle wheel or knot meter. And speed at the lure, which needs a probe sent down on a downrigger cable to measure directly. Serious tournament boats carry the probe. Everyone else has to infer it.
Correct Lake Troll Speed: Luhr-Jensen Tech TipsHow to infer it for free
Put the lure in the water beside the boat at trolling speed and look at it.
That is the whole method, and it is better than most electronics. Drop your spoon over the gunwale on a short line, hold it a foot under the surface, and watch what it does at the speed you are actually going. A spoon that is working has a rhythm to it. A spoon that is too slow hangs and rolls. A spoon that is too fast stops wobbling and spins like a propeller, which is a dead giveaway and also the fastest way to put line twist into your reel.
Do this every time you change lures and every time conditions change. It takes fifteen seconds and it answers a question that no gauge on your dash can answer, because it accounts for the current, the line, the knot and the lure all at once.
The limitation is honest and worth stating: what you see beside the boat is the surface. A lure at ninety feet on a downrigger is in different water. But it establishes the relationship between your gauge reading and that lure's behavior, and that is most of what you need.
What each kind of lure wants
Every lure has a speed window. Inside the window it works. Outside it, it is decoration.
Spoons are the most forgiving and the most common. A thin light spoon has a wide slow wobble and works down at the slow end. A thick heavy spoon needs more water flow to get moving and comes alive higher up. This is why a box of spoons that all look similar can behave completely differently at the same speed, and why people who troll spoons well sort them by thickness rather than by color.
Spinner blades are the most speed-sensitive things most anglers use. Blade shape is the control. A round Colorado blade has a lot of surface facing the water, so it turns at very low speed, throws a lot of thump, and lifts. A long narrow willow leaf blade cuts through instead, needs more speed to spin, and lifts much less. That difference is why a worm harness is a speed-specific tool and why changing blades is often more useful than changing speed.
Diving crankbaits have a speed where they run deepest and best, and a speed above which they blow out, meaning they lose their track and start skating sideways. Both edges matter. Under-speed and they wander without action, over-speed and they leave the depth you wanted them at.
Dodgers and flashers are attractors rather than lures, and they behave differently from each other. A dodger swings side to side without going all the way around, which is a slower-speed action. A flasher rotates completely, which needs enough flow to keep it turning. Whatever hangs behind them takes most of its motion from them rather than from itself, which means a plastic squid on a flasher is running at whatever speed the flasher wants and not at whatever speed the squid wants.
Trolling Speed: How to Choose ItLine twist, the tax you pay for getting it wrong
There is a specific and expensive failure mode at the fast end and it is worth naming because people rarely connect it to speed.
A spoon that is going too fast stops wobbling and begins to rotate. A rotating lure winds your line, one turn at a time, for as long as it is out there. Troll a blown-out spoon for two hours and you have put a very large number of twists into a line that had none, and you find out about it the next time you let anything out and it comes off the spool in coils.
The same thing happens with a spinner running above its blade's window, and with any bait that has begun to spin because it picked up a weed or got hooked crookedly.
The fixes are ordinary. Slow down into the lure's window. Use a good ball bearing swivel ahead of anything that rotates by design. And when you do get twist, let the bare line out behind a moving boat with nothing on the end and let it unwind itself, which takes a few minutes and saves respooling.
The reason to mention it in a speed article is that twisted line is often the first symptom people notice of a speed problem they had not otherwise detected. If your lines keep twisting, you are probably trolling faster than your lures want.
Speed as a trigger, not just a presentation
Everything so far treats speed as a setting to be optimized. There is a second way to use it that is worth knowing.
Fish that are following without committing are a common and maddening situation, and they show up on sonar as marks that rise to the gear and then fade. The standard response is a change in speed rather than a change in lure: a brief surge, a sudden drop, a hard turn that makes everything in the spread accelerate on one side and stall on the other.
The reasoning is that a bait behaving predictably is easy to inspect, and a bait that suddenly changes is either escaping or vulnerable, both of which prompt a decision. Whether that is what the fish is thinking is not something anyone can say, but the practice of deliberately breaking rhythm is close to universal among people who troll for a living, and it costs nothing to try on a day when everything is following and nothing is eating.
The practical version is simple: if the spread is being inspected and refused, stop optimizing your speed and start varying it.
The numbers people start from
These are starting points, not rules, and they exist because a lot of people converged on similar answers rather than because anyone measured them for your lake.
Kokanee salmon get trolled very slowly, often around a mile an hour or a little more, with small gear.
Walleye cover a wide range depending on method and season, roughly a mile an hour up to two and a half, with cold water pushing toward the slow end and summer crankbait trolling toward the fast.
Chinook salmon and the other Pacific salmon in the Great Lakes generally get trolled in the two to three range, faster than most freshwater fish and slower than saltwater pelagics.
Muskellunge trollers routinely run faster than anything else in freshwater, well up past three and sometimes considerably higher, pulling large baits.
Take all of them as the middle of a range that shifts with water temperature. Cold water generally means slower, both because fish are less inclined to chase and because the same lure has a different feel in denser water. That is a rule of thumb with a real basis and plenty of exceptions.
Speed is not one number all day
Two things make holding a speed harder than it sounds.
Wind and waves change your speed constantly. Running downwind in a chop, the boat surges on every wave, and a lure that is right on the average is blowing out on the peaks. Into the wind, the opposite. A trolling plate, a drift sock or a kicker motor exist largely to make speed steadier rather than lower, and steady is the part that matters.
Turns change your speed differently on each side. In a turn, gear on the inside slows down and drops, gear on the outside speeds up and rises. This is a nuisance if you are trying to hold one presentation and a gift if you are trying to find one, which is the argument for the S-turn covered in flatlining and longlining. Every curve is a free experiment across a small range of speeds, and the side the fish comes from tells you which way to adjust.
What changing speed does to your depth
You cannot change speed without changing depth, and this catches people out constantly.
Speed up on a downrigger and blowback increases, so the ball rises even though the counter reads the same. Speed up on lead core, and the line planes rather than sinking, so every color is worth less depth than it was. Speed up with a diver and it runs shallower. Speed up with planer boards and the boards pull harder and ride higher while any diving lure behind them may blow out.
So the sequence matters. Set the speed first, based on what the lure needs, then set the depth. Setting depth first and then adjusting speed means undoing the depth you just set, and the counter will not tell you it happened.
The gear
Almost none of this is expensive, and the cheapest item on the list is the most useful.
A GPS speed readout is what nearly everyone uses and it is the baseline. Just know what it is measuring.
A paddle wheel or knot meter gives you speed through the water at the surface, which is a genuinely different and often more useful number than speed over ground. On a boat that fishes moving water this is the upgrade that pays.
A subsurface speed and temperature probe is the real answer and the expensive one. It rides down on the downrigger cable and reports what the lure is actually experiencing. Not a first purchase, but it explains days that otherwise make no sense.
A kicker motor or a trolling plate, for holding a slow speed steadily. Most main engines idle faster than a lot of freshwater trolling wants.
Drift socks, for the days when wind is doing the driving and you need to take speed out of the boat without taking control out of it.
A tackle box sorted by action rather than by color, because thickness and blade shape determine speed window and color does not.
And a rod holder position where you can actually see the rod tips. A working lure puts a steady rhythmic pulse in the tip. A dead one does not. That pulse is a free speed gauge on every rod you have out, and it is the first thing to go when a lure fouls.
The habit worth building
Change one thing at a time and write down what happened.
Speed is the variable most people never systematically test, because it feels like a setting rather than a choice. Depth gets changed constantly, lures get changed constantly, and the throttle sits where it sat last time.
Try moving it deliberately. Half a mile an hour faster for twenty minutes. Then half slower. Watch which rods go off. The information is sitting there in every trip and almost nobody collects it, which is why the boats that do tend to catch more than the boats that do not.
Keep the notes crude and keep them anyway. Date, water temperature, the speed on the gauge, which direction you were going, what was on each rod and what took a fish. Four lines in a notebook. After a season of that you will have something no amount of reading produces, which is a record of what your water does rather than what somebody else's water does, and speed will have stopped being the thing you guess at and become the thing you look up.
