Cross Weight Is a Lie You Scale For

Column #15: Cross Weight Is a Lie You Scale For

Cross weight is the most cited, most misunderstood, and most dangerously trusted number in dirt racing. Every Saturday night, somebody rolls off the scales with 50.0% cross, punches the air like they just won the damn lottery, and then gets driven around for 20 laps by a car scaled at 46.3%. The 50.0% car was perfect — on the scales. The 46.3% car was perfect — on lap 20. That is the difference between scaling for a number and scaling for a racetrack.

What Cross Weight Actually Is

Cross weight — also called wedge — is the combined weight on two diagonally opposite corners of the car, expressed as a percentage of total weight. The formula is simple:

The diagonal relationship is what makes cross weight powerful and deceptive at the same time. When the RF and LR carry more weight than the LF and RR, the car has positive wedge. It wants to resist rotation — tighter on entry, more planted on exit. When you pull cross out — lower that percentage — you free the car to rotate. The nose dives into the corner easier. The rear gets lazier about following.

That is the static truth. Emphasis on static.

Why 50% Cross Is Not Neutral

Here is where the lie starts. People hear "cross weight" and think 50% is neutral. Balanced. Even. It is not. 50% cross means equal diagonal loading — in the shop, on flat concrete, at ambient temperature, with cold tires, zero fuel burn, and no lateral or longitudinal g-forces acting on the chassis. The moment the car fires and rolls onto the track, every single one of those conditions changes.

Think about what a dirt car actually does. It turns left. Only left. The entire weight distribution is biased left — 52% to 57% depending on class — to counteract the centrifugal force pushing the car to the right. The rear percentage is biased rearward — 54% to 60% — to keep the driven wheels planted. The car is already asymmetric before you ever touch the cross weight adjuster.

Now add the track surface. A freshly worked dirt track at hot laps has moisture content somewhere around 18-25%. Heavy, tacky, forgiving. By lap 20 of a feature, that same surface is 8-12% moisture in the racing groove, maybe less on a dry night in west Texas. The mechanical grip available from the tire-to-surface interface drops by 30-40% over the course of a feature. The car that was "perfect" on the scales has to survive that entire transition, and 50% cross does not account for any of it.

The Dynamic Cross Weight Problem

Here is the physics that matters and nobody writes down in a textbook. When a dirt car enters a corner at speed, weight transfers in three directions simultaneously: forward under braking, laterally to the right from centrifugal loading, and vertically through suspension compression. That transfer does not distribute evenly across the four corners. It follows the spring rates, the shock valving, the torsion bar preloads, the ride heights, and the tire sidewall deflection — all of which are different at each corner by design.

On a winged 410 sprint car running 1,450 lbs total with 54% left and 56% rear, the wing is generating 400-800 lbs of downforce depending on angle. That downforce loads the rear axle disproportionately — roughly 65-70% of wing load hits the rear. It also shifts the center of pressure depending on yaw angle. When the car rotates sideways through the corner — standard on dirt — the wing's effective angle of attack changes by 3-8 degrees. That is a 50-150 lb swing in downforce happening dynamically, every single corner, and it does not show up on the scale pads.

This is why winged sprint cars run 43-47% static cross weight. They look "loose" on the scales. They are not loose on the track because the wing adds downforce that effectively raises the dynamic cross weight by 3-6 percentage points at speed. A sprint car at 45% static cross is running closer to 49-51% dynamic cross in the corner. The scales told you one thing. The tire contact patches experienced something completely different.

Look at that spread. A winged 410 at 43% cross and a non-wing 410 at 54% cross are not even in the same zip code on the scales, yet both cars can be perfectly balanced in the corner. The wing is the variable. It always is. And it changes every lap as the track surface changes, because the car's speed changes, and aero load is a function of velocity squared. Two mph slower through the middle of turns three and four costs 30-50 lbs of downforce on a winged sprint car. That is the equivalent of half a percentage point of cross weight — gone — because the track got a little slick.

The Moisture Curve and Why Your Scales Lie by Lap 20

This is the part nobody talks about, and it is the whole reason this column exists.

You scale the car at 5:30 PM. The track was just worked. It is tacky, heavy, beautiful. You set the cross at 52% in your late model. Hot laps feel great. Heat race, you run second — car is hooked up. You are confident.

Feature time. Lap 1 through 8, the car is solid. Lap 9, the track starts to take rubber in turns one and two. By lap 14, the top three inches of moisture are gone in the primary groove. Lap 18, you are sliding the right rear, fighting a push off the corner, and the car that was "perfect" at 52% cross is now effectively running at 54-55% dynamic cross because the reduced surface grip has changed how weight transfers through the suspension.

Here is the cause-and-effect chain:

Less surface grip → less lateral g-force in the corner (car cannot generate the same cornering speed) → less weight transfer to the right side → the static cross weight percentage becomes a larger portion of the total cornering load → car tightens.

On a high-grip surface, a 52% cross car might see dynamic cross rise to 53-54% because the aggressive weight transfer partially counterbalances the static bias. On a slick surface, that same 52% car sees dynamic cross stay at 52-53% because less weight transfers laterally. The static number dominates. The car pushes.

The fix is to scale the car for lap 20, not for hot laps. That means setting static cross 1-2 percentage points lower than what feels "right" on a fresh track. The car will feel slightly free early — maybe a half-tenth loose on entry during the heat — and come alive as the track slicks off. I have run this approach on late models for 15 years. It costs you positions early sometimes. It wins you features.

Wedge vs. Cross: Same Coin, Different Slang

In dirt racing, "wedge" and "cross weight" are used interchangeably by 90% of racers. They are related but not identical.

Cross weight is the measurement — the percentage. Wedge is the adjustment — the act of changing the preload on one or more corners to alter the cross weight percentage. On a torsion bar car like a sprint car or micro sprint, you add wedge by tightening the RR torsion bar stop (adding preload) or loosening the LR stop (removing preload). Both moves increase RF + LR diagonal loading, raising cross percentage.

On a coilover car like a late model or modified, wedge adjustments happen via spring rubbers, weight jacks (screw jacks on the spring perch), or physically changing spring rates. Turning the RR weight jack up one full turn on a typical late model moves approximately 8-12 lbs of cross weight, which shifts cross percentage by 0.3-0.5% on a 2,400 lb car.

The common mistake: racers adjust wedge without knowing what their actual cross percentage is. They add "a round of wedge" because the car was loose, without realizing they are already at 54% cross and the looseness is coming from the rear percentage being too low, or the left-side percentage being insufficient, or — and this is the one that burns me — the RF tire is 3 psi higher than the LF because nobody checked pressures after the heat race. A 3 psi increase in the RF tire raises that corner's effective spring rate by 8-15% on a dirt tire at 10-14 psi operating range. That is a cross weight change of 10-20 lbs that does not show on your scale sheet because you did not re-scale after the heat.

Tire Pressure: The Invisible Cross Weight Adjuster

Every 1 psi of tire pressure change on a Hoosier dirt tire alters the effective spring rate of that corner by approximately 3-6 lb/in on a late model tire, 2-4 lb/in on a sprint car tire, and 1-2 lb/in on a micro sprint tire. That spring rate change moves weight on the scales. Not a lot — 3-8 lbs per corner per psi — but enough to shift cross weight by 0.2-0.4% per psi on a 1,400 lb sprint car.

Over a 25-lap feature, tire pressures climb 2-4 psi from cold. If the RF and LR gain pressure faster than the LF and RR — common when the car is rotating well and loading those corners harder — cross weight increases dynamically without anyone touching a wrench. The car gets tighter as the race goes on, and the crew chief is standing on the trailer scratching his head because "we did not change anything."

You did not change anything. The tires did. Every lap.

The Bite Number: Cross Weight's Quieter Cousin

Bite — also called left-rear bias — is the weight difference between LR and RR. A car with 640 lbs on the LR and 580 lbs on the RR has 60 lbs of bite. This number governs exit traction independent of cross weight. You can have identical cross percentages and vastly different bite numbers depending on how the rear weight is distributed.

On dirt, bite matters more than on pavement because the rear tires are fighting a surface that changes grip level every lap. Sprint cars typically run 5-15 lbs of bite. Late models run 10-30 lbs. Modifieds run 10-25 lbs. More bite means more forward drive off the corner from the LR tire, which is the inside rear on a left-turning oval. That tire has less vertical load from cornering forces, so it needs the static preload to stay planted.

The mistake I see most often: a racer adjusts cross weight to fix an exit problem, when the actual problem is bite distribution. You can be at 51% cross and still be loose on exit if your bite is at zero or negative. The cross percentage can be "right" while the individual corner weights are wrong. This is why I tell every racer who comes through my pit: do not chase a percentage. Read all four corners.

How Each Adjustment Moves the Cross

Weight moves diagonally. This is the fundamental law of scaling. If you add weight to the LF corner — via torsion bar preload, spring rate, ballast, or ride height — the car behaves as if you simultaneously removed weight from the RR. The RF and LR diagonal gets lighter. Cross goes down. Car loosens on entry.

If you add weight to the RF corner, cross goes up. Car tightens on entry. Every adjustment at one corner has a diagonal echo at the opposite corner. This is not theory. It is measurable on the scales to within 2-3 lbs.

Here is the map, burned into my brain from 40 years of standing next to scale pads:

LF preload up: Left% up, Cross% down. Car loosens entry, can tighten center.
RF preload up: Left% down, Cross% up. Car tightens entry, loosens center.
LR preload up: Left% up, Cross% up, Rear% up. Tightens entry AND exit.
RR preload up: Left% down, Cross% down, Rear% up. Loosens entry AND exit.

The LR is the most powerful single corner on a dirt car. Touching it moves left percentage, cross percentage, rear percentage, and bite — all four key metrics — simultaneously. This is why experienced crew chiefs start with the LR torsion bar or spring when making cross weight changes, not the RF. The RF is the easy button. The LR is the right button.

Sprint Cars vs. Late Models: Why the Numbers Diverge

A winged 410 sprint car at 45% cross and a super late model at 52% cross are both "correct." The reasons are mechanical and aerodynamic, and understanding them is the difference between a crew chief and a guy with a scale sheet.

The sprint car has a top wing generating 400-800 lbs of downforce. It has no front brakes — only a left rear inboard brake. It uses torsion bars at all four corners, birdcages that allow rear steer, and a Jacobs ladder that controls lateral location of the rear axle. The wing downforce supplements the mechanical cross weight so aggressively that high static cross creates an over-tight condition in the corner that the single rear brake cannot overcome on entry. Low static cross gives the car freedom to rotate on entry, and the wing catches it on exit by adding load to the RF-LR diagonal dynamically.

The late model has no wing. It has a pull bar or lift arm that controls rear load transfer under throttle. It has coilover shocks with independently adjustable spring rates. It has front and rear brakes. Without aero downforce to supplement mechanical grip, the late model needs higher static cross to prevent rotation — the car does not have a wing to "save" a loose exit. The pull bar becomes the dynamic wedge adjuster: under acceleration, the pull bar loads the left rear and right front, functionally adding cross weight as the car drives off the corner. A properly set pull bar angle (35-42 degrees from horizontal on most cars) adds 1-3% effective cross weight under throttle application.

The modified lives between these worlds. No wing, but a torque link rear suspension that creates its own dynamic wedge through geometric anti-squat. The Harris torque link, specifically, generates rear load transfer differently than a 4-link — it ties the housing rotation to chassis pitch, effectively adding wedge under acceleration proportional to driveshaft torque. This is why modified cross weight targets (50-54%) overlap with late models despite the modified being 100-200 lbs heavier.

The Street Stock Reality

Street stocks deserve their own paragraph because the scaling situation is fundamentally different. You are working with a GM metric chassis — stock front suspension, stock-ish rear, factory mounting points. Your cross weight adjustment tools are limited to: fuel cell placement (left side of trunk, always), ballast location, spring rate selection (which at this budget level means cutting coils or buying one spring from the speed shop), and tire pressure.

Most street stock racers I have worked with do not own scales. They borrow them once, set the car, and never re-scale. That one session is still worth more than zero sessions. At 3,000-3,400 lbs total, a street stock that is 2% off on cross has 60-70 lbs of diagonal imbalance. That is a full handling condition — not a nuance. Get on scales once. Write down all four corners. Tape it inside the door bar. That number is your baseline forever until you change springs or move weight.

Common Mistakes — The Wrong Numbers People Chase

Mistake #1: Targeting 50% cross in a winged sprint car. I have watched people put 50% cross on a 410 and wonder why the car plows through the corner like a freight train. The wing is already adding 3-6% dynamically. At 50% static, you are running 53-56% dynamic. That is late-model territory on a car with no front brakes. The car cannot rotate.

Mistake #2: Chasing cross without checking total weight distribution. Cross weight is one of four numbers — left%, rear%, cross%, and bite. A car with "perfect" 51% cross but only 51% left will be loose all night because there is not enough static left-side weight to counteract centrifugal force. I see this constantly on late models where someone added ballast to "fix cross" without realizing they diluted the left percentage.

Mistake #3: Scaling with inconsistent tire pressures. If the LR is at 14 psi and the RR is at 10 psi from sitting overnight, your cross weight reading is garbage. Every tire must be at the same cold pressure — or your target hot pressure, if you are scaling hot. I use 12 psi all four corners as my scale-pad standard for late models. Consistent starting point.

Mistake #4: Ignoring ride heights during scaling. Ride height determines the static cross weight as much as spring preload does. If the RF ride height is 1/2" higher than intended because a jack stand was in the wrong spot during setup, you have added 15-25 lbs to that corner. Cross jumps 0.6-1.0%. The car tightens on entry and you have no idea why because "the numbers were right."

Mistake #5: Scaling once and racing all year. Every