> ## Documentation Index
> Fetch the complete documentation index at: https://offthepace.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Which teams find more in qualifying than in the race?

> The gap between each constructor's single-lap qualifying pace and their sustained race pace, separating qualifying specialists from race-spec cars.

"Party mode" was the paddock nickname for the engine maps teams could unleash for a single qualifying lap. This chart measures the effect structurally: it differences each constructor's **qualifying structural pace** against their **race structural pace**, both estimated from the same fixed-effects panel model. A team that extracts far more over one lap than they sustain across a stint shows a negative delta they run a party-mode profile.

## What it shows

Each constructor is placed on a single delta:

* **Negative delta (party mode)**: the team extracts more pace over a single lap than they sustain across a race stint a qualifying-biased car or engine map.
* **Positive delta (race spec)**: the car comes into its own over race distance; the team may lose time in qualifying but manage tyres or fuel better than their grid slot suggests.
* **Balanced**: teams within ±0.15s of zero are even across the two sessions.

Both pace estimates come from the fixed-effects panel that isolates the constructor's car contribution after controlling for driver, tyre, fuel, and conditions so the delta is a property of the car, not of who drove it.

<Note>
  The effect was most pronounced in the turbo-hybrid era, when manufacturers could publish separate qualifying and race engine maps. It persists today as an aerodynamic signature: a high-downforce qualifying setup versus lower-drag race trim.
</Note>

## How to use it

1. **Select a season** to see that year's constructor deltas.
2. **Sort by delta** to rank the field from most qualifying-biased to most race-biased.
3. **Read the ±0.15s balanced band** teams inside it have no meaningful session bias.
4. **Pair with grid-vs-finish results** to see how often a party-mode car's qualifying advantage survived the race.

## Reading the signal correctly

The delta is a **difference of two model estimates**, so it inherits both estimates' uncertainty a team with few clean teammate-pair observations in either session has a noisier delta. A negative delta is a *bias*, not a flaw: a strong qualifier still scores from the front.

<Warning>
  Qualifying and race pace are estimated on different lap populations (single-lap Q runs versus race stints). A team that barely runs in a given session a rain-shortened qualifying, for instance can show a misleading delta for that round.
</Warning>

## Data source

The chart queries the constructor structural-pace panel for both sessions via DuckDB-Wasm in the browser: `int_constructor_structural_pace` (race) and `int_constructor_structural_pace_qualifying` (qualifying).

| Column                           | Meaning                                                |
| -------------------------------- | ------------------------------------------------------ |
| `structural_pace_s` (race)       | Constructor car contribution to race lap time          |
| `structural_pace_s` (qualifying) | Constructor car contribution to qualifying lap time    |
| `party_mode_delta_s`             | Qualifying pace minus race pace; negative = party mode |

The data window covers **2018–2024**.

For the full model definitions, see the [`int_constructor_structural_pace` reference](/reference/models/int/int_constructor_structural_pace) and the [`int_constructor_structural_pace_qualifying` reference](/reference/models/int/int_constructor_structural_pace_qualifying).