Table of Content
Empty miles are the cost that doesn’t show up on the load. They show up in the lane. Operators who optimize loads run more deadhead than operators who optimize lanes, and the difference shows up on every paid mile.
Deadhead is the most consistently underpriced cost in owner-op math. The truck runs empty between a delivery and the next pickup. Fuel burns, the engine runs, the operator’s clock ticks against HOS, the trailer wears at the same rate as if it were loaded. None of it produces revenue. All of it produces expense.
Most operators count it on a percentage basis , “I run about 12% deadhead”, and stop there. The percentage looks small. The actual cost looks small. The cost compounded across a year is two to three rate-increases worth of money, and most operators leave it on the table because they’re optimizing the wrong unit.
“12% deadhead is normal” is the assumption that hides the damage
The average owner-operator runs deadhead in the 10-14% range. The good ones run 6-9%. The difference is roughly four percentage points. On 110,000 miles a year, four percentage points is 4,400 empty miles.
| Deadhead % | Effective CPM impact | Annual hit (rough) |
|---|---|---|
| 5% | Small | Low |
| 10% | Material | Mid four figures |
| 15%+ | Significant | Five figures |
4,400 empty miles in 2026 dollars is roughly $2,600 in fuel at $0.62 per mile diesel cost, plus $480 in maintenance, plus 56 hours of HOS clock burned that could have been on paid runs. Total visible cost: $3,100. Total opportunity cost including the loads not run during those 56 hours: $4,800-$6,500.
The number doesn’t appear anywhere on the truck’s accounting. There’s no expense line called “deadhead opportunity cost.” But the truck running 8% deadhead nets $5,000-$7,000 more across a year than the same truck running 12% deadhead in the same lanes, and that gap is invisible because it’s a difference between what the truck did and what the truck could have done with the same effort.
Where deadhead actually comes from
Deadhead originates from three operational decisions, and most operators only see the third one.
- Lane selection. Some lane structures produce deadhead by their nature. A lane that delivers into a thin freight market, small Texas border towns, rural Pacific Northwest, parts of the Mountain West, produces 200-400 mile deadheads to get back to a freight zone. The operator running those lanes routinely is structurally absorbing higher deadhead than the operator running shuttle lanes between two freight-rich zones. The deadhead came from the lane, not from the load.
- Reload timing. An operator who delivers in Boston Tuesday afternoon and books the next pickup for Wednesday morning in Newark runs 220 miles of deadhead because the reload was booked on time but not on geography. An operator who books the same Wednesday Newark pickup but takes a Tuesday-evening Boston-to-Hartford backhaul first runs 200 paid miles where the deadhead would have been. Same delivery point, same next pickup, different reload pattern produces dramatically different deadhead percentage.
- Rate-chasing. An operator who delivered in Indianapolis and gets a call about a $2,800 load picking up in Memphis takes the load. The deadhead from Indianapolis to Memphis is 470 miles. The math the operator runs: “$2,800 gross is good.” The math they don’t run: “$2,800 gross minus 470 miles deadhead at $1.10 per mile loaded-equivalent cost is effectively a $2,283 net-revenue load.” That’s still maybe a fine load if the alternative was sitting. It’s a worse load than the operator priced. The operator who runs that math wrong consistently is the one whose deadhead is structurally high.
Most operators see only the third one. They optimize each load in isolation. The other two, lane selection and reload geography, produce more deadhead damage across a year than rate-chasing does, and they’re invisible at the load level because they only show up in the season totals.
The lane structures that actually hold deadhead under 8%
Operators running deadhead under 8% across a year are usually doing one of three things consistently.
The first pattern is shuttle lanes between two freight-rich zones. Atlanta-Chicago, Dallas-Chicago, Atlanta-Northeast metro, anywhere there’s reliable load volume in both directions. The operator picks two corridors and runs them repeatedly. Deadhead drops because the next load is almost always within fifty miles of the delivery, and the broker book is short and tight.
The second pattern is regional density. Northeast metro shuttle work, Texas triangle, California regional. The operator never runs more than 350 miles from home base, and the freight is dense enough that the next load is usually under 100 miles deadhead. This pattern produces lower RPM but very low deadhead and very high reload speed, and the net usually beats long-haul.
The third pattern is broker concentration. The operator runs with 8-12 brokers across the year, each of whom rebooks the truck regularly. The broker calls before the truck is even empty , “we have a load near you Wednesday, can you get there?”, and the deadhead drops because the booking is happening from inside a relationship, not from a board search. This is the pattern dispatched operators with mature desks run, and it’s the highest-margin pattern of the three because the rate also rises with broker concentration.
Pull the last ten loads and look at where the empty miles are actually coming from. The lanes are rarely the whole story.
Where high deadhead is not actually a problem
Some operations run high deadhead and the math is fine. Specialty freight, heavy haul, oversize, hazmat, produces deadhead in the 18-25% range routinely because the lanes are specialized and the return freight isn’t always there. The rate on those loads carries the deadhead because the rate is multiples of dry van.
Owner-operators running specialty freight who measure themselves against dry van deadhead percentages misread their own performance. 22% deadhead at $4.80 RPM is structurally healthier than 10% deadhead at $2.30 RPM on most operating cost structures. The percentage isn’t the metric; the percentage in context of the rate is the metric.
The same applies to seasonal freight. Operators running produce season May through July may run 14-16% deadhead during that window because the corridor pulls one direction harder than the other. The premium on the loaded miles carries it. The same operator running 14-16% deadhead in October on dry van is structurally bleeding.
The metric most operators should be running and aren’t
Deadhead percentage is one metric. Revenue per total mile run, including the empty ones, is the metric that actually predicts net.
An operator at $2.45 RPM with 12% deadhead is running revenue at $2.16 per total mile. An operator at $2.32 RPM with 7% deadhead is running revenue at $2.16 per total mile. The same. The first one looks better on the rate sheet. The second one looks better on the lane structure. The bank account doesn’t care which one looks better on which sheet.
The operator who runs revenue per total mile as the actual decision metric instead of RPM stops chasing rate as a primary lever. Lane discipline beats rate hunting on revenue per total mile by a clear margin in most owner-op operations, and the operator who reframes their math accordingly stops taking $2,800 Memphis loads with 470-mile deadheads in front of them.
Where dispatch earns the line on deadhead
The desk’s role on deadhead is mostly structural. Lane selection happens at the conversation level, what corridors does the truck want to run, what brokers does the desk have memory with, what reload patterns hold up week to week. That’s the kind of work a desk can do for a truck and an operator largely can’t do alone, because it requires broker book breadth that takes years to build.
The desk’s job inside the booking decision is to refuse the high-deadhead load when a paid backhaul is available. An operator on the phone alone in a cab is biased toward the load that’s offered now, not toward the load that might post in three hours that’s only 80 miles deadhead. A desk has the visibility to know whether to wait. Across a year, that judgment shaves two to four percentage points of deadhead off a truck that was running 12% solo.
The standard you should hold yourself to in 2026 is deadhead under 9% on a dry van one-truck operation, under 11% on reefer or specialty, and under 7% on regional shuttle work. Anything above that line is a structural lane problem, not a load-by-load decision problem, and the structural fix usually involves the broker book and the lane choice, not the rate negotiation. The dispatch fee usually pays back inside the deadhead reduction alone.
