Lactate Test Record: Why Every Field Matters

Why document every field of a lactate test record: conditions, stages and measurement errors, with a worked example where 0.2 mmol/L moves LT by 3 W.

Why This Matters: The Note Behind Each Lactate Test Field

Kristian Hoffmann ยท SaaS founder and operator

Every field on a lactate test record exists because something in the analysis breaks without it. Stage workload and lactate value are the curve itself. Stage duration and sampling minute decide whether the stages are comparable with each other. Conditions such as room temperature, last meal, meter model and strip lot decide whether this test can be compared with the next one. Measurement notes explain gaps and repeats. Leave one of them out and an LT1 and LT2 calculator will still return a confident number, just from data you can no longer check.

This page goes through the fields of a home step-test record, gives the reason behind each, and shows with a fictional dataset how much one small measurement error moves the result. It covers data quality and workflow only. What a threshold means for your health or your training is a question for a qualified coach, sports physician or exercise physiologist. Which fields to type into a calculator is covered in our guide to the 12 data-entry fields and common mistakes; this page is about why you record them in the first place.

What each field protects

A step test produces pairs: a workload and a blood lactate value. Everything else on the sheet exists so that those pairs can be trusted and repeated. A step test here means an incremental protocol in which workload rises in fixed increments at fixed intervals, with a blood sample near the end of each stage.

FieldWhy you record itTypical failure
Stage workload (W or pace)Defines the x-axis; every interpolated value depends on itEntering target power instead of the average actually held
Stage durationDifferent durations produce differently shaped curvesShortening a stage "because it felt done"
Sampling minuteSamples taken at different points in a stage are not comparableSampling at 2:30 on one stage and 3:45 on the next
Lactate value (mmol/L)The y-axis; small errors move curve crossingsDecimal comma typed as thousands separator
Heart rate at sampleLets you map results to a second metric laterCopying the stage average from a different window
Resting baselineAnchors the left end of the curveSkipped to save a strip
Meter model and strip lotDifferent devices can read the same blood differentlySwitching meters between tests without noting it
Room temperatureIndoor heat changes conditions between sessionsUnrecorded change from garage to heated living room
Last meal and caffeinePre-test state differs between sessions"Normal breakfast" that was not normal
Notes on a failed stripExplains a gap or a repeatA repeated reading logged as a new stage

The table is not ranked. The first four fields decide whether a single test can be analyzed at all; the last six decide whether two tests can be compared.

Single-test fields

Workload, duration, sampling minute and lactate value are the minimum input for any curve fit. Most online calculators, including tools that report LT1, LT2 and VLamax, assume these four are correct and consistent. They cannot detect a stage that ran 3:20 instead of 4:00.

Comparison fields

Baseline, device, temperature and pre-test state rarely change the shape of one curve visibly. They matter when you put two curves side by side and want to know whether a shift is real or produced by the setup.

A worked example: Where a 0.2 mmol/L error moves the curve

The following numbers are fictional and chosen to make the arithmetic visible. A rider completes seven 20 W stages:

StagePower (W)Lactate (mmol/L)
11501.1
21701.2
31901.4
42101.8
52302.6
62503.8
72705.6

To keep the calculation transparent, use straight-line interpolation between neighbouring stages and two fixed reference lines at 2.0 and 4.0 mmol/L. These are arithmetic reference levels used in the literature as conventions, not a recommendation for how to define or use your own thresholds.

Crossing at 4.0 mmol/L. It falls between 250 W (3.8) and 270 W (5.6). The slope is 1.8 mmol/L over 20 W, or 0.09 per watt. The gap to 4.0 is 0.2 mmol/L, so the crossing sits 2.2 W above 250, at 252.2 W.

Now suppose the 250 W reading was off by 0.2 and should have been 3.6. The slope becomes 2.0 over 20 W (0.10 per watt), the gap is 0.4, and the crossing moves to 254.0 W. Shift: 1.8 W.

Crossing at 2.0 mmol/L. It falls between 210 W (1.8) and 230 W (2.6). Slope 0.04 per watt, gap 0.2, crossing at 215.0 W. If the 210 W reading should have been 1.6, the slope becomes 0.05 per watt, the gap 0.4, and the crossing moves to 218.0 W. Shift: 3.0 W.

Same size of error, two different consequences. The flatter the curve where an error lands, the further a reading error moves the crossing. That is why the low end of a lactate curve needs the most careful sampling, even though it feels like the easy part of the test.

What this means for the record sheet

  • Log every reading exactly as the meter shows it, without rounding. If you average two readings and get 1.75, write 1.75: Rounding it to 1.8 is already a quarter of the error above.
  • Mark repeats explicitly. If stage 4 was measured twice, the sheet should show both values and which one you used.
  • Keep the step size constant. The interpolation above assumes 20 W everywhere; a single 30 W step changes the slope maths for that segment.

The calculator does not see your sheet

Online LT1 and LT2 calculators accept the numbers you type. Whether they use Dmax, a fixed 4 mmol/L value or another method, a wrong input produces a wrong output with the same number of decimals as a correct one. Methods also differ from each other on the same clean data, which is a separate source of spread from the input errors shown here; Step Test Analysis: 4 Threshold Methods, 33 Watt Difference shows how large that spread can be.

Which fields LactateThreshold asks for, and which stay on your sheet

The LactateThreshold analysis asks for the test date, resting and maximum heart rate, the stage duration in minutes and the blood sample site (earlobe or fingertip). For each stage you enter power or pace, heart rate and the lactate value; a second mid-stage lactate value and RPE are optional. The tool flags values outside plausible ranges as invalid, for example lactate below 0.1 or above 30 mmol/L, and it warns when stages are shorter than four minutes. It cannot tell whether a plausible number is the wrong one: A 3.8 typed instead of 3.6 passes every check.

The analysis has no separate fields for room temperature, last meal, meter model or strip lot. When you save a test, the editor has a free-text notes field, and that is where those comparison fields belong. Write them there or on your paper sheet before you look at the curve, not afterwards. VLamax is not estimated from the step test at all; the tool calculates it only from a separate sprint test.

The strip budget behind a complete record

A complete record costs more strips than the stage count suggests. For the seven-stage example:

ItemStrips
Resting baseline1
One sample per stage7
Repeat on the two flattest stages2
Reserve for one failed strip1
Total11

Planning for 11 strips instead of 7 is what makes the "why this matters" notes affordable in practice. A tester who runs out at stage 6 either skips a stage or reuses a questionable reading, and both show up as noise in the curve.

Why the record matters more over the winter

Many endurance athletes move testing indoors over the coming winter months, and the holiday period interrupts routines. That changes exactly the comparison fields from the first table.

Indoor conditions

A test on a smart trainer in a heated room is a different setup from an autumn test in a cool garage, even with the same protocol. Record room temperature and fan use at the start and end. If two tests differ by several degrees, note it next to the result before comparing curves.

Breaks, travel and changed routines

A test scheduled shortly after a holiday break, travel or a period of disrupted sleep is still a valid data point, but it belongs in a different comparison bucket. A simple rule for the record: If anything in the week before the test differed from your usual pattern, write one line about it. You decide later whether it was relevant; you cannot reconstruct it later.

Device changes over the gift season

New meters and new strip lots often enter the kit around this time of year. Record the model and lot number on the first test with new equipment. If a curve shifts on that test, the device change is the first thing to rule out, and a qualified professional or the device manufacturer is the right place to ask about measurement agreement. How much a single reading error on one specific meter changes the result is worked through in the Lactate Pro 2 analysis of a 0.2 mmol/L reading error.

FAQ

Why does every field on a lactate test record matter?

Because each field protects one part of the analysis. Workload, duration, sampling minute and lactate value decide whether one curve can be fitted at all. Baseline, device, temperature and pre-test state decide whether two tests can be compared. A missing field rarely causes an error message; it causes a result you cannot check.

Can an LT1 and LT2 calculator correct a bad reading?

No. A calculator fits whatever pairs it receives. Some tools flag implausible jumps, but none can know that a reading came from a contaminated sample or a stage that ended early. That check happens on the record sheet before data entry.

Which field should I never skip?

The sampling minute, in most home setups. Workload and lactate value are obvious to record; the exact timing is the one testers drop, and without it you cannot tell whether two stages are comparable.

How do I use the record to compare tests over time?

Compare only tests whose comparison fields match, or note the difference next to the result. A worked approach is in how to compare lactate test results over time without false precision, and a full log template is in the 16-field lactate test progress log.

Before your next test, add a column to your sheet titled "why I record this" and fill it once for every field. Any row you cannot explain is either unnecessary or a gap you have not noticed yet.