Methodology

Purpose

LMU BoP Tourism is an independent comparison site for Le Mans Ultimate. It turns public lap time, balance, race pace, and schedule signals into pages that help players compare cars, tracks, setup shops, and game versions.

The site is not an official BoP source. It is a data aid for sim racers, not a rulebook or guarantee of what will be fastest for every driver.

Lap Time Sources

Setup-shop benchmark laps are collected from GoSetups, BeAlien, and Hymo. Each source is handled separately, then converted into the same internal shape: source, car, class, track, game version, and lap time in milliseconds.

The website displays benchmark lap times only. It does not sell, host, or distribute setup files.

Normalization

Car, class, and track names are matched to canonical site records before data is stored. Game versions are normalized so different source formats can be compared consistently. Lap times are stored as integer milliseconds so sorting and gaps are deterministic.

If a source publishes a car, class, or track name that cannot be resolved with confidence, that row is held out of the public comparison tables until it can be mapped safely.

Fastest Lap Selection

For each source, the site keeps one current row per car, track, and version. If the same version appears again, the row is updated only when the newly seen lap is faster. Slower repeats are ignored for ranking purposes.

This keeps comparison pages focused on each source's best known benchmark for the selected car, track, and version.

BoP And Version Context

BoP pages compare lap times against the game version attached to each benchmark. Where version and balance context matters, the interface labels older data so users can judge whether a time is still useful after later game updates.

Patch and BoP history pages are derived from public update information and internal canonical car records. They are intended to explain context, not to replace official patch notes.

Reading A Benchmark Lap Time

A setup-shop benchmark is an optimized qualifying lap, not typical race pace. It is the ceiling of what a car can do at a circuit, set by a fast driver on a prepared setup, so treat it as the top of the range rather than a target you should be hitting.

For where you actually sit, the lap time pages measure the real driver field instead: every BoP page links through to the tiers for that class and track, and those tiers are percentiles of the people racing, not a fixed gap to the quickest lap on record.

Driver Lap Distributions And Tiers

The lap time pages are built from official Le Mans Ultimate race sessions: one qualifying lap and one race lap per driver, per car, per track, per setup mode. That population is what the curves plot and what the tier cutoffs are cut from.

Tiers are percentiles of that field, so a badge means the same thing on every circuit. Alien is the top 0.5%, gold the top 10%, silver the top 30%, bronze the top 50%, rookie the top 75%, and offline is everyone behind that. The earlier approach, a fixed percentage off the quickest recorded lap, could not do this: field spread grows with lap length, so one band that described an average driver at Monza left most of the field off the scale at Le Mans. Nobody falls off a percentile scale by construction.

Three splits are never pooled, because pooling them would describe a field that does not exist. Fixed and open setup are separate fields and are shown separately, with a switch only where a combination has both. Race pace is split by race length, since fuel burn-down makes a sprint lap and an enduro lap different measurements. And every lap is scoped to the car's current BoP period, so a balance change starts the measurement over rather than blending two sets of rules into one curve.

Qualifying is the exception to the length split: a hot lap does not care how long the race after it runs, so those laps pool across every length.

Race pace is a sustained figure, each driver's median lap across their best race at that combination, not their single quickest lap. A driver's race laps carry a slow tail from lap one, traffic, pit in and out, and mistakes, which is why the median is used rather than the average. You will beat your own marker on individual laps, and so will everyone else on the curve, which is why a handful of quick laps does not move you a tier.

Driver Rating is shown as an overlay on the curve rather than as the scale itself. The top of the rating ladder holds only around a hundred drivers site-wide, far too few to set a meaningful fast-end boundary, so the boundaries come from percentiles and the rating only marks where each stratum sits. A curve is drawn only once its field is large enough to mean something; below that the page shows the summary figures and says the chart is not ready.

What We Will Not Claim About A Car

The obvious chart, one curve per car ranked by median lap, is the one thing this data cannot support. Drivers choose their own cars, so a car's measured pace level is partly a measurement of who picked it. Tested at Le Mans, the car with the fastest median was eighth of ten once the same drivers were compared against themselves across cars, a seven place inversion. So no page here ranks cars by the pace of their own field.

Two per-car figures are published, and they are different things.

The best lap reached in each car is a target, not a ranking. It is the quickest lap anyone has managed, so it improves with how many people have raced the car: a car 2,000 drivers have run will show a better lap than an equally quick car 200 have run. It is sorted fastest first so you can find your own lap in the list, and that ordering carries no claim that the car at the top is the quicker car. Cars under 30 drivers are held back, and the tiers pages say how many, because a balance change restarts that count and can remove most of a grid at once.

Driveability is the one per-car figure here that does carry an ordering claim. It measures how far a mid-pack driver falls behind the quick drivers in the same car, so each car is judged against its own quick drivers and never against another car. Narrow means forgiving, wide means demanding. This survives the problem above because it is a measure of spread rather than of level: the bias that wrecks a pace comparison largely cancels, and what is left is far smaller than the differences between cars.

It still says nothing about speed. A car can be demanding and quick, and the clearest case on record is a car whose fastest drivers were mid-pack while its field trailed furthest behind them: its ceiling was fine, most of its drivers could not reach it.

Naming a most forgiving car has two guards. Cars need at least 100 drivers before they are measured at all, and the count that cleared that bar is always printed beside the claim, because cars below it are left out and some of them would have placed well. And where the leaders are within a tenth of each other they are named together rather than ranked, since that gap is close enough to measurement noise that picking a winner from it would be inventing a result.

Car Strength Rankings

Car strength pages aggregate benchmark pace across selected tracks. The ranking favors cars that are consistently close to the fastest benchmark rather than cars that are strong at only one circuit.

These rankings are best read as pace indicators. Driver comfort, stint length, setup style, tyre behavior, traffic, and race format can all change the practical answer.

Best Cars At A Track

The best-cars block at the bottom of each lap time page answers "what should I drive", where the curves above it answer "am I fast". It deliberately shows four signals side by side rather than one score, because they disagree, and the disagreement is the useful part. At Sebring the GT3 signals point in three directions: the Lamborghini holds the quickest setup lap, the Porsche has the highest win rate and sits fifth of the six cars measured for driveability, and the cars a mid-pack driver gets the most out of are the Mustang, the Ferrari and the Lexus, level with each other.

Win rate is the strongest signal that a car is genuinely competitive, and it appears only once at least 12 races have been run at that layout, so a car cannot be crowned off a handful of results. Podium rate carries the same floor. Pick rate is listed separately and labelled as popularity, since it measures what people choose rather than what works. All three come from official race results on periods where the balance was stable, and from whichever setup mode has run the most races there, which is not always the mode the curves above are drawn on, so each panel names its own.

The per-car lap lists and the driveability figure are covered above under what we will not claim about a car. Race figures there are split by race length, one list per length, since a car that suits a sprint is not always the car that suits a long race. At Laguna Seca the GT3 that leads the 60 minute list does not place at all in the shortest one.

Layout variants of the same circuit (different chicane or gate configurations) are tracked as separate tracks because setup-shop benchmark laps and pace genuinely differ between layouts, even when official race results are still too sparse to name a win-rate leader on the less-used variants.

Race Result And Schedule Data

Lap distributions, race statistics, leaderboards, and the weekly schedule all come from game-service and public-facing race data processed on a schedule. The site groups and summarizes it into track, class, car, setup mode, and race length views.

The collection details for those feeds are intentionally summarized here. The important user-facing rule is simple: these pages are generated from stored data, refreshed by background jobs, and not by live provider calls during page rendering. Pages are cached at the edge and say so, so a figure can be a few hours behind the most recent race.

Update Cadence

Source refreshes run automatically. Different data families update on different schedules depending on how often the underlying source changes and how expensive the refresh is. Public pages show the latest stored data available to the application.

Limitations

• - Benchmark laps reflect the source's driver, conditions, setup, and publication timing.
• - Not every source publishes every car and track at the same time.
• - A newer patch can change the relevance of older laps even when those laps remain visible for comparison.
• - Race pace targets are guidance, not a promise that a specific time is achievable for every driver.
• - Any "best lap reached" figure improves with how many people have raced that car, so it is a target to aim at rather than a comparison between cars.
• - A curve can fall back one balance period when the current one is too thin to draw. The page labels it when that happens: the laps are real and the spread holds, but the cars' positions relative to each other may not.
• - Only the circuits the daily race rotation has reached carry a driver field. Everywhere else the pages fall back to setup-shop benchmarks and say so.

Corrections

If a car, track, version, source, or lap time appears wrong, send details through the Contact page. Corrections are handled manually when source evidence is clear.