# Time vs distance & the value of corner-exit speed

> A constant time gap looks like a growing distance gap on a straight. Exit speed gains carry down the following straight but decay — by how much depends on whether the car is traction-, power- or drag-limited.

Area: Gearbox & Gearing · Starting level: L1 Foundation · Updated 2026-11-16
Source page: https://race-engineer-academy.vercel.app/lesson/gearbox-time-distance-exit-speed

## L1 · Foundation

**Time versus distance (Segers §4.4).** "I catch him in the corners but he pulls away on the straights" is often an illusion. Two identical cars 0.4 s apart: at 65 km/h (18 m/s) that's 7 m; at 265 km/h (74 m/s) it's 29 m. The time gap never changed — the **distance gap scales with speed**, $\Delta x = v\,\Delta t$. Always compare in time (delta-time channel), never by eye in metres.

**Corner exit speed (Segers §4.5).** Straights can be 70–80 % of a lap's length. A higher exit speed is carried down the following straight, so time is gained over a long distance. Segers' Spa example: an aero change that cured high-speed understeer at Raidillon gave **+2.9 km/h** exit speed. If that delta had been held over the 990 m straight, the gain would have been 0.17 s; in reality it was **0.11 s**, because the speed difference shrinks as the car accelerates — the faster car has more drag and less excess power, so the slower car catches up in speed.

Lessons:
- Exit speed before a long straight is valuable — but less than the naive "constant delta" calculation says.
- The decay is fastest when the car is **power-limited** and especially near terminal speed, where drag dominates.
- That is why "sacrifice entry for exit" is a nuanced rule — it is very true out of slow, traction-limited corners onto long straights, much less true for fast corners where the car is already power-limited.

Gearing is linked: the gear used on exit and the rpm where full throttle is accepted determine how quickly the car builds speed in the first 50–100 m — where most of the gain is made.

## L2 · Practitioner

**How to analyse it in data.**
- Use the **time-difference (delta) channel** vs distance between two laps; its slope tells you where time is gained: $d(\Delta t)/dx = 1/v_B - 1/v_A ≈ -\Delta v/v^2$.
- Because of the $1/v^2$ weighting, **1 km/h at 80 km/h is worth ≈ 9× more per metre than 1 km/h at 240 km/h**. A small speed gain in a slow section is a big time gain per metre.
- Check the speed difference at fixed points (+50, +150, +300 m after the apex, and at the braking point). A shrinking difference is the normal exit-gain decay: identical cars show $\Delta v \propto 1/v$ when traction-limited and $\propto 1/v^2$ when power-limited, at fixed distance points. A difference that stays **constant** therefore already implies a small power/drag/tow advantage for the faster car. A **growing** difference means a large one (BoP, tow, wing level).

**Prioritising work.** Combine (a) the speed gain available at each corner exit (from driver comparison or theoretical optimum) with (b) its carry-over value. Slow corners onto long straights are top of the list. Segers' Raidillon case shows even a fast corner can be worth 0.1 s when followed by a long straight.

**Common misconception**: "the driver who is fastest on the straight has the better engine". If his exit speed is higher, he will be faster all the way down the straight even with identical power. Compare the **speed gradient** (acceleration at the same speed) instead: identical $a_x$ at the same speed = identical power/drag.

**Gearing & exit**: an exit gear that forces an upshift in the first 50 m costs a shift interruption in the most valuable part of the straight. A gear that leaves the engine below its torque band on pickup costs the same region. Gear choice is part of exit-speed optimisation.

## L3 · Advanced

**How fast does an exit-speed gain decay?** Compare two runs from the same point, one with $v_0 + \Delta v_0$.

- **Traction-limited (constant acceleration $a$):** $v^2 = v_0^2 + 2ax$ → at equal $x$: $v\,\Delta v = v_0\,\Delta v_0$, so $\Delta v = \Delta v_0\, v_0/v$.
- **Power-limited, drag-free ($m v\, dv/dt = P$):** $v^3 = v_0^3 + 3Px/m$ → $v^2\Delta v = v_0^2 \Delta v_0$, so $\Delta v = \Delta v_0 (v_0/v)^2$.
- **With drag**, the decay is faster still near terminal speed.

**Time gained** over the straight $\Delta t = \int_0^X \Delta v / v^2\, dx$. For the power-limited, drag-free case this integrates exactly:

$$\Delta t = \frac{m\, v_0\, \Delta v_0}{P}\left(1 - \frac{v_0}{v_X}\right)$$

Physical interpretation: $m v_0 \Delta v_0$ is the **extra kinetic energy** carried out of the corner; the gain is (almost) the time the engine would need to produce it. **But this formula is only valid where the car is really power-limited**, i.e. above $v^* = P/(m\,a_{trac})$. For a GT3 (1300 kg, 360 kW at the wheels, ≈ 0.75 g of traction-limited acceleration) $v^*$ ≈ 38 m/s ≈ 135–140 km/h. Below that the engine could deliver more force than the rear tyres can carry (360 kW at 80 km/h would be 16 kN, ≈ 1.3 g).

**Two-phase model for a slow exit.** Traction phase at constant $a$ from $v_0$ to $v_1 = v^*$, then power phase to the end of the straight $v_X$:

$$\Delta t_1 = \frac{\Delta v_0}{a}\left(1 - \frac{v_0}{v_1}\right),\qquad \Delta v_1 = \Delta v_0\frac{v_0}{v_1},\qquad \Delta t_2 = \frac{m\, v_1\, \Delta v_1}{P}\left(1 - \frac{v_1}{v_X}\right)$$

Example: GT3 as above, exit 75 km/h (20.8 m/s) with +2 km/h (0.556 m/s), $v_1$ = 140 km/h, straight ending at 270 km/h. $\Delta t_1 = 0.0755 × 0.464 ≈ 0.035$ s; $\Delta v_1$ = 0.30 m/s; $\Delta t_2 = 0.042 × 0.48 ≈ 0.020$ s; total ≈ **0.055 s**. The power-only formula from 75 km/h would give only 0.030 s, and the naive constant-delta estimate ($\Delta v$ held at 0.556 m/s over the same ≈ 500 m speed trace) ≈ 0.12 s. Drag makes $\Delta v$ decay faster at a given distance, but it also lowers the net acceleration. Level 4 shows that for the same exit and end speed the total gain depends only on the net acceleration at the exit, $\Delta t = \Delta v_0/a_{net}(v_0)\,(1 - v_0/v_X)$: here ≈ 0.0755 × 0.72 ≈ 0.055 s, the same as the two-phase sum. At 75 km/h drag lowers $a_{net}$ by only a few %, so the result barely changes; from a fast, power-limited exit drag lowers $a_{net}$ noticeably and the true gain is larger than the drag-free formula says. That is why you should read $a_x$ from the logged long-g at the exit speed rather than compute it from $P/v$.

**Unifying view:** the gain is roughly the time the car needs to *generate* $\Delta v_0$ at its exit acceleration, $\Delta v_0 / a_x(v_0)$, times a decay factor.

Consequences:
- **Slow, traction-limited exits:** $a_x$ is set by grip and is the highest of the lap. The speed delta decays only as $1/v$ until $v^*$, so the gain persists down the straight. Per km/h, the value is roughly flat with $v_0$ below $v^*$.
- **Fast, power-limited exits:** the per-km/h value $m v_0/P$ rises with $v_0$, but the delta decays as $1/v^2$ and the $(1 - v_0/v_X)$ factor shrinks it when the exit is already close to the end speed.
- Per km/h, slow and fast exits are worth similar amounts ($1/a$ rises with $v_0$ but $(1 - v_0/v_X)$ falls). Slow corners onto long straights usually come out ahead mainly because much larger $\Delta v$ gains are available there; traction improvements in the first 50 m carry the most time.
- Value ∝ $m/P$ in the power phase: exit speed is worth more in low-power cars (F4, GT4), where $v^*$ is also lower and most of the straight is power-limited.

Use this to rank corners quickly before a sim: gain ≈ $\Delta v_0/a$ for traction-limited exits, ≈ $m v_0 \Delta v_0/P$ for power-limited ones, each with its decay factor.

## L4 · Expert

**General form with drag.** For any longitudinal model $m_{eq}\,v\,dv/dx = F(v) - D(v)$ (with $F = \min(F_{trac}, P\eta/v)$, $D = \tfrac12\rho C_dA v^2 + C_{rr}mg$), linearise around the reference run. With $a(v) = (F - D)/m_{eq}$ the perturbation obeys

$$\frac{d(\Delta v)}{dx} = \Delta v\,\frac{d}{dv}\!\left(\frac{a(v)}{v}\right) \quad\Rightarrow\quad \Delta v(x) = \Delta v_0\, \frac{a(v)/v}{a(v_0)/v_0}$$

So the speed delta at a fixed point scales with the **net acceleration per unit speed**. The time gain then is

$$\Delta t = \int_0^X \frac{\Delta v}{v^2}dx = \frac{\Delta v_0\, v_0}{a(v_0)} \int_{v_0}^{v_X}\frac{dv}{v^2} = \frac{\Delta v_0}{a(v_0)}\left(1 - \frac{v_0}{v_X}\right)$$

using $dx = v\,dv/a$. This exact first-order result holds for *any* force model, including drag and the traction/power switch: **the time gain equals the time the car needs to produce $\Delta v_0$ at the exit acceleration, times $(1 - v_0/v_X)$**, where $v_X$ is the reference speed at the end of the section. Check: with $a = P/(m v)$ it gives the power-limited formula; with constant $a$ and $v_X = v_1$ it gives the traction-phase term. (The two-phase sum at Level 3 is the same thing split in two.)

**What this says for engineering:**
- The quantity to know at each corner is $a_x$ at the exit speed, which you read directly from the logged long-g at that speed. No power or drag model is needed.
- At a corner exit where the car is still at high combined lateral g, the *available* $a_x$ is lower. A small gain there is worth more, which is why "straightening the exit" pays.
- Near terminal speed $a \to 0$ and the formula diverges. That is not a contradiction: there the gain is cut off by the braking point, through $v_X$ approaching $v_0$.

**Limits of the model.**
- It is first-order. Large exit gains (≥ 5 km/h) change the shift points and possibly the gear, so they must be simulated.
- It integrates the gain up to the reference braking point. The faster car must brake about $v_X\,\Delta v_X/a_{brake}$ earlier, but that changes the result only at second order (≈ $\Delta v_X^2/(2\,a_{brake}\,v_X)$), so it is negligible for small gains.
- Tow, traffic and DRS change $D(v)$ between the two runs, which breaks the "same car" assumption. In a race the exit-speed advantage often matters more as *time in the tow* than as lap time.
- Fuel or energy limits (hybrid deployment, lift-and-coast) mean that higher exit speed may be "spent" differently. In energy-limited cars the correct metric is time per unit energy, not time alone.

## Key formulas

**Distance gap from time gap**

$$\Delta x = v\, \Delta t$$


**Time gain from speed difference**

$$\Delta t = \int_0^X \frac{\Delta v(x)}{v(x)^2}\, dx$$


**Speed-delta decay, power-limited**

$$\Delta v = \Delta v_0 \left(\frac{v_0}{v}\right)^2$$


**Exit-speed time gain (power-limited, no drag)**

$$\Delta t = \frac{m\, v_0\, \Delta v_0}{P}\left(1 - \frac{v_0}{v_X}\right)$$


**Exit-speed time gain (traction-limited phase, constant a)**

$$\Delta t_1 = \frac{\Delta v_0}{a}\left(1 - \frac{v_0}{v_1}\right)$$


**Traction/power crossover speed**

$$v^* = \frac{P}{m\, a_{trac}}$$


## Trackside takeaways

- Constant time gap = growing distance gap at speed; compare drivers in time, not metres.
- Exit-speed gains decay along the straight: ∝ 1/v when traction-limited, ∝ 1/v² when power-limited.
- Exit gain ≈ Δv₀ / a_x at the exit: ≈ Δv₀/a when traction-limited, ≈ m·v₀·Δv₀/P (extra kinetic energy / power) when power-limited.
- Check which regime applies: below v* = P/(m·a_trac), about 135–140 km/h for a GT3, the exit is traction-limited.
- Segers’ Raidillon case: 2.9 km/h exit gain → 0.11 s, not the 0.17 s of the naive estimate.
- Exit speed is worth relatively more in low-power cars (F4, GT4) — momentum matters.
- Compare speed gradients at equal speed to judge power/drag, not top speed alone.

## By car category

### Formula 4

Low power-to-weight makes exit (and minimum) speed extremely valuable: m/P is large, so each km/h carried out of a corner is worth more time than in faster cars.

### GT4

Heavy and modestly powered: corner exit speed and momentum dominate; drivers who over-slow for a "better exit" often lose more than they gain.

### GT3

Traction-limited exits from slow corners onto long straights are top priority; diff, TC and exit gear choice all feed into the first 50–100 m.

### Hypercar (LMH/LMDh)

Power capped by BoP: exit gains decay like any power-limited car; in traffic the time-vs-distance illusion is acute (closing speeds on GT cars).

### FIA F2

High power and no TC: exit traction in the first 50 m (throttle application, diff) is where drivers differ most.

## References

- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §4.4 Time versus Distance
- Segers, Analysis Techniques for Racecar Data Acquisition, 2nd ed. (SAE 2014): §4.5 The Importance of Corner Exiting Speed (Raidillon example: 2.9 km/h, 0.11 s vs 0.17 s)
- Milliken & Milliken, Race Car Vehicle Dynamics (SAE 1995): Ch. 20 Driving and Braking
