How to use Stout Racing Lab

Getting started

1. Record telemetry in iRacing. In the car, press Alt+L to start logging. iRacing writes an .ibt file to Documents\iRacing\telemetry when you leave the car.
2. Upload it on the Upload page (or let the companion do it, below). Car, track, weather and the setup are read from the file.
3. Read the session page: laps, corners, tires and setup. Open Traces to compare laps sample by sample.
4. Ask the engineer for up to three setup changes and three driving tips, each with its evidence.
5. Try a change, drive again, compare. Mark the change as applied, upload the next session and use Compare with… to see whether it worked.

Units: switch between Metric (km/h, m, °C, kPa) and Imperial (mph, ft, °F, psi) in the header; your browser remembers it. This page follows your choice where it quotes numbers (metric now). Five to ten clean laps at a track you know give the most useful results.

Uploading sessions

Telemetry (.ibt)

Drop it on the Upload page. Large files (80 MB is normal) upload with a progress bar and are processed in the background: laps, corners, tires, balance and driving metrics are usually ready within a minute.

Setup sheet (.htm, optional)

In the garage, export or print the setup to HTML and drop it in together with the .ibt. The app checks it is the same car and the same settings as in the telemetry. If you saved it right after the run, its garage readouts (pyrometer temperatures, last hot pressures, tread wear) are added to the tire data. To add a sheet later, upload the same .ibt again together with the sheet.

Automatic upload

The Windows companion app runs on the sim PC, watches the telemetry folder and uploads every finished session on its own. It connects to your account through the browser, with no token to copy.

Same file twice

Uploading the same telemetry again opens the existing session instead of creating a copy.

Laps

One row per complete lap: lap time, the gap to your best lap, a strip of the lap's segments, and flags.

Segment strip

The lap is cut into segments from corner to corner. Each block is colored by the time that lap lost in that segment against the fastest clean lap there: the fastest, almost nothing lost, a medium loss, the biggest losses of the session. The steps are sized to each session, and the legend under the table gives the times. Hover a block for the exact value. Columns of red across many laps are where your time goes consistently; a single red block is a one-off mistake.

Flags

out lap, in lap, incident (incident points gained), off track, slow (more than 7% slower than your best clean lap), incomplete (recording started or stopped mid-lap). Flagged laps are greyed and left out of every average, so one spin doesn't distort the results.

Clean laps

Complete laps with no flags. Every metric below is computed over clean laps only.

Corners

Corners are found automatically from the path the car drove (how tight it turned), so they match the track without manual setup. Each is named C1, C2… along the lap, plus the track's turn name when the app knows it. All values are the mean over clean laps.

Type

Direction (L/R) and how tight: hairpin (radius under 35 m), slow (under 80 m), medium (under 150 m) or fast.

Entry / Min / Exit

Speed where the corner starts, the slowest point, and where it ends. Minimum speed is the number that most often decides the corner time.

± (spread)

Lap-to-lap spread (standard deviation) of the minimum speed. Large values mean you drive the corner differently each lap: a consistency problem before a setup problem.

Brake at

Where braking starts, as distance from the start/finish line, with its lap-to-lap spread. A large ± means an inconsistent brake point.

Pickup

Where you get back on the throttle, relative to the slowest point of the corner. Negative: before the slowest point (usually good: you carry the car through); large positive: late pickup, often after coasting.

Time lost

Average time this corner costs per lap compared to your best lap's time in the same corner. The corners with the largest values are where to look first.

Balance mid

Mid-corner understeer or oversteer in steering-wheel degrees, compared to the car's own average; see Balance. Greyed values are low confidence (hairpins, or a large lap-to-lap spread).

Balance (understeer and oversteer)

The app learns how much steering this car needs for a given curve and lateral grip on this session, then measures every corner against it. The result is in steering-wheel degrees:

• Positive: you needed more lock than the car normally needs: understeer (the front is running out of grip).
• Negative: less lock, or counter-steer: oversteer (the rear is stepping out or the car rotates easily).
• Split into entry, mid and exit, because the fixes differ: entry problems often come from braking and brake bias, mid-corner from the mechanical balance (anti-roll bars, springs), exit from traction (rear bar, differential, traction control, throttle).

It is a relative indicator: compare corners and phases with each other and with other sessions of the same car, not with other cars. A shift in one corner with a large spread usually means driving; the same shift in many corners with a small spread usually means the car.

Tires

One card per wheel, laid out like the car seen from behind (outside edges on the outside). Values are averaged over the last three clean laps, when the tires are up to temperature; the laps are listed in the heading.

Three color blocks

Surface temperature across the tread: inner (I), middle (M), outer (O). The color scale is shared by all four wheels, blue cool to orange hot.

I−O (inner minus outer)

Positive: the inner edge is hotter. A few degrees is normal with negative camber; a large value suggests too much negative camber, a negative value too little.

M−edge (middle minus the average of the edges)

Positive: the middle is hotter, a sign of too much pressure. Negative: the edges are hotter, too little pressure.

Hot / Cold / Rise

Running (hot) pressure, the cold pressure set in the garage, and the difference. rising means the hot pressure was still climbing at the end of the session (by 0.5 kPa per lap or more), so it hasn't settled yet.

Pit stall, end of run

When the session ends with a stop in your pit stall: the pyrometer temperatures (inner, middle, outer), the pressure and the tread remaining that iRacing recorded there. These are probe temperatures taken after the in-lap, so they are lower than on track, but the inner/middle/outer spread is the classic camber and pressure reading. See Stints and pit readings.

Garage after run

Without a pit-stall reading, the same values from an uploaded setup sheet. Greyed and marked other run when the sheet was saved after a different run: shown for reference but not used.

Stints and pit readings

A stint is the running between two pit visits. The section appears when a session has more than one stint or ends in the pits, and the lap table marks where each new stint starts.

Pit reading

iRacing measures the tire carcass temperatures and tread only when the car reaches its pit stall, never on track. So each stop records one reading for the stint just driven: inner / middle / outer temperature per wheel, the pressure, and the tread left. Because it comes from the telemetry, it is always from this run (unlike a setup sheet saved later). To get one, finish your run by driving into your pit stall, not by exiting from the track.

Pace trend

How much the lap time changes per lap over the stint's clean laps (needs at least four). Positive (red): getting slower, usually tire wear or overheating; negative (green): getting faster, usually you improving or the track rubbering in. Grey: within the lap-to-lap noise. The fuel load drops every lap, which makes the car a little faster and can hide some tire wear.

First 3 → last 3

Average of the first and the last three clean laps of the stint (with six or more clean laps).

Fuel / lap

Median fuel used per lap, without in- and out-laps: useful for planning race stints.

Pressures settled

The lap from which each tire's hot pressure rose by less than 0.5 kPa per lap. “No” means it was still climbing at the end: set the cold pressures for the pressure you reach late in the stint, not on lap 3.

Tread used

Tread worn during the stint, per zone. Uneven wear across the tread tells the same story as the temperatures: more wear on the inside, too much negative camber; in the middle, too much pressure.

Tire averages on the session page use the last stint with at least three clean laps, so laps on different tire sets or fuel loads aren't mixed.

Setup and setup sheets

The setup is read from the telemetry file, grouped as in the garage. Greyed parameters are garage readouts (computed values like corner weights): they can't be changed directly and the engineer never recommends them.

When you uploaded a setup sheet, a badge next to the heading shows the check:

• setup sheet matches: same car and every setting agrees with the telemetry (within the sheet's rounding, e.g. toe printed as 1/16").
• setup sheet: N settings differ: the sheet has other settings (listed below the heading); the telemetry's setup is used and the sheet's readouts are ignored.
• setup sheet: other car: ignored.

Legal values. Every change the engineer suggests is checked against the garage: ✓ in the garage range (limits entered for this car), ✓ seen in a legal setup (that exact value appeared in a setup that passed iRacing's tech check), ✗ outside the legal range (with the nearest legal value), or not checked while the limits for that car aren't known yet: check the value in the garage then. The setup list shows the known range next to each value.

Fixed setups. When the session was fixed-setup, a fixed setup tag appears: the garage was locked, so the engineer suggests only in-car adjustments (brake bias, TC, ABS) and fuel, and otherwise driving tips.

Trace viewer

Open it with Open traces on a session. Stacked charts show speed, throttle, brake, steering, gear and the gap to your best lap along the lap distance; every lap you tick is overlaid in its own color.

Reading the gap chart

Each line is that lap's time difference to your best lap at that point of the track. Where a line rises the lap is losing time; where it falls it is gaining. Look for where it starts rising, then check the speed, brake and throttle traces at the same spot to see why (braked earlier, lower minimum speed, later throttle).

Corners

Shaded bands mark the detected corners, labeled on the top chart.

Track map

The cursor position on the track, colored by speed, brake or throttle (t switches).

Cursor

Move the mouse over any chart: the cursor is linked across all charts and the map, and the table at the top shows every lap's values at that point.

Choosing channels

Channels opens the list of charts: tick the ones you want, change their order, or pick a preset (driving, braking, balance…). Beyond the basics there are RPM, G forces, yaw rate, ABS and, per wheel, wheel speed, ride height, shock travel and tire surface temperature. Anything the file of that car doesn't carry is greyed out. Your choice is saved in this browser.

Moving around the lap

The strip above the charts shows the whole lap and highlights the part you are looking at: drag it to pan, or drag outside it to choose another stretch. The arrows and the corner list jump from corner to corner, and a click on the track map zooms to the corner you clicked.

Driving report

Open Driving report on a session page. It turns the session's driving data into a coach-style report: where your time goes and what to do about it. It is computed from your telemetry (free and instant); every point comes with the numbers and laps behind it. Print / save PDF gives a clean printout.

Headline

Best lap, your theoretical best (your best time in every part of the track, combined) and the time available between them, plus your typical lap. The verdict says whether the gain is mostly in consistency (your typical lap is far from your best) or in pace (you're consistent, but your best sectors aren't on the same lap).

Biggest opportunities

The parts of the lap (from one corner's turn-in to the next) that lose the most time on average against your own best time through them, biggest first. Each shows what the data shows there (coasting, a moving brake point, late throttle, lock-ups, wheelspin, steering corrections, a late apex), what your quickest pass through it did differently, and one thing to do. View in traces opens the trace viewer zoomed to that spot.

Consistency

One bar per lap, its time lost to your best lap; grey laps are excluded (in/out laps, incidents). Below: the corners where your minimum speed and brake point vary most.

What you do well

Corners you already drive close to your best every lap, with a repeatable minimum speed.

Coach's notes

Ask the coach has Claude turn the report into plain-language priorities, one drill per priority and a plan for your next session. It only sees the report's numbers, so it can't invent data. It costs 1 credit.

Ask the engineer

On a session page, optionally describe what the car does (“loose on exit at Stavelot”) and click Analyze. Claude reads the session's summaries through tools (it never sees raw telemetry), checks your description against the data, and answers in two separate parts: setup changes and driving tips.

An engineer run costs 3 credits and a coach run 1 credit (your balance is in the header; buy more on the Credits page). You get 6 free credits when you confirm your email. Credits never expire, and if a run fails, its credits come back automatically.

Data quality

good, limited or poor: how much the conclusions can be trusted (few clean laps, incidents, missing channels).

Setup changes (max 3)

Each card names the symptom, the parameter with its current and new value, the evidence (metrics with values), the expected effect, the side effects and a confidence. Change one thing at a time, then drive again.

Driving tips (max 3)

When the cause is the driver, not the car: the corner, what the data shows, one clear instruction and, when the data supports it, the time it may gain.

Checks

Every recommended value is checked against your setup and, where the app knows them, against the car's legal garage range (see Setup). A badge on each card shows the result; when it says not checked against the garage, confirm the value is allowed in the garage before using it.

Mark as applied

Mark the change you made. When you compare this session with the next one, the app checks whether the change is in the new setup and whether the metrics it should improve did improve.

Your description is a hint, not a verdict: if you say “loose on exit” but the data shows mid-corner understeer, the engineer should say so and fix what the data shows.

Comparing sessions

On a session page choose Compare with… (sessions at the same track). The earlier session is A (before), the later one B (after); all deltas are B − A.

You don't have to compare by hand: when you mark a recommendation applied and upload your next session with the same car and track, that session's page shows Did the change work? with the lap-time change, whether its setup has the change, and the verdict described below.

Header

Best and mean lap, lap-time spread, theoretical best, and the car's overall understeer gradient (steering degrees per g; lower means the car turns more easily). Notes warn when the conditions differ, e.g. track temperature.

Did the applied changes work?

For each change marked applied in session A: whether B's setup has it, the metrics that change should move (camber: that axle's inner−outer spread; pressures: middle−edges; bars, springs, wing, diff: corner balance; ride height: the lowest ride height per lap), and a verdict: improved, worse, mixed or no clear change.

Corners

Corners are matched between sessions by position. Only corners with a real change are listed (tick show all for the rest). US is the understeer angle; B is measured with A's steering model so a car-wide balance shift shows up. Negative = more rotation.

Tires and ride height

Changes per wheel. Temperature differences convert to °F without the +32 offset, so “no change” reads +0.0 °F.

Reference laps and progress

A reference in the trace viewer

Pick a Reference above the lap list: your best lap from another session in the same car on the same track, or the lap of a driver who shares theirs. It is drawn as a dashed line, and the gap trace and the Gap column are measured against it. The two laps are lined up by their position on track.

Sharing your laps

Off by default. On your account, switch on Share my best laps and pick a nickname if you like. Others see your nickname, the time, the date and the traces of your best lap of each session; never your email or your setup.

Progress

The Progress page shows, per car and track, your best and average lap in each session, your theoretical best, and which corners got better between your first and latest session. As when comparing, a change only counts when it beats your normal lap-to-lap variation.

Real change or noise?

Lap times and corner speeds vary from lap to lap even when nothing changed. A difference is only colored when it is larger than twice that lap-to-lap noise and big enough to matter (0.7 km/h, 3 m, 1 °C or 1 kPa):

Hover any value to see A, B and the noise band. Grey does not mean “no effect”: with only a few laps the noise is large, so drive more clean laps to detect smaller changes. The same applies to the lap table: one fast lap is not yet a faster car.

Limits and data quality

• Few laps: with fewer than three clean laps most comparisons can't be significant and the engineer reports limited data.
• Legal ranges: known only for some cars and values; check any suggested value marked not checked against the garage before you drive.
• Corner detection depends on the line driven: a flat-out kink may be found only on some laps (marked low confidence). Turn names come from the app's track files; tracks without one show C1, C2…
• Tire surface temperatures react quickly to the last corner; use trends over several laps rather than single values.
• Telemetry channels differ between cars; when one is missing, the affected metric says why instead of showing a number.