More Plates More Dates Blood Work: What Derek's Panel Measures and How to Run It Yourself

Woolf Software
bloodwork biomarkers hormones proteomics lab-assays
A translucent creature in a glowing wetland with dozens of colored fluid channels in its flank, mirrored by glass-like reeds nearby.

The panel people have in mind when they search for “More Plates More Dates blood work” is a broad serum chemistry and hormone workup. It is usually drawn through Marek Health, the telehealth company Derek co-founded, at a Quest or Labcorp patient service center. It is ordered more often than a general practitioner would order it. It is also read against “optimal” ranges rather than against the reference intervals the lab prints on the report.

The panel composition is good, and the assay selection is better than what most primary care ordering produces. The weak part is interpretation. The optimal ranges circulating in that ecosystem are consensus opinion rather than outcome-validated cut points. The population being tested, meaning people using anabolic steroids, SERMs, aromatase inhibitors, GLP-1s and peptides, is one in which almost no reference interval was derived. If you are going to run this on yourself, the value lies in the longitudinal series and in pre-analytic discipline rather than in any single flagged value.

What the panel contains

It helps to see what is inside one of these panels before arguing about how to read it. Strip the branding and a typical “comprehensive” male panel is roughly 50 to 70 analytes across six blocks:

  • CBC with differential is the complete blood count. It reports hemoglobin and hematocrit, RBC and MCV, and platelets and WBC subsets.
  • CMP is the comprehensive metabolic panel. It reports sodium, potassium and chloride, along with CO2 and glucose. Kidney markers are BUN, creatinine and eGFR. Protein markers are albumin and total protein. Liver markers are bilirubin, ALP, AST and ALT.
  • Lipids covers total cholesterol, LDL-C (calculated or direct) and HDL-C. It also covers triglycerides, apolipoprotein B and lipoprotein(a).
  • Androgens and pituitary covers total testosterone, free testosterone and SHBG. It also covers LH, FSH and estradiol. Prolactin and DHEA-S are included, and sometimes DHT.
  • Thyroid covers TSH, free T4 and free T3. Sometimes reverse T3 and TPO antibodies are added.
  • Metabolic and inflammatory covers HbA1c, fasting insulin and hs-CRP. It also covers ferritin, iron/TIBC and homocysteine. The remaining markers are uric acid, IGF-1, 25-OH vitamin D and cystatin C.

None of this is exotic. The same set can be ordered through any direct-to-consumer lab portal. In most US states it will be run by the same two reference labs. What a service like Marek sells is order-writing, panel curation, and a provider who will look at the result. Whether that is worth the membership fee depends mostly on whether you would otherwise sit down and interpret the series yourself.

The assay choices that matter more than the panel size

A panel’s usefulness depends less on how many analytes it contains than on how a few of them are measured. Three method decisions drive most of the signal. This is where the MPMD material is genuinely correct against mainstream practice.

For estradiol, order the sensitive assay rather than the standard chemiluminescent immunoassay. The sensitive version uses liquid chromatography with tandem mass spectrometry, abbreviated LC-MS/MS, which separates the molecule chemically before measuring it. The immunoassay was designed around female physiologic ranges and cross-reacts badly at the much lower concentrations seen in men. Its functional sensitivity sits somewhere around 15–20 pg/mL. LC-MS/MS reaches roughly 2–5 pg/mL with far less interference. For a male serial-tracking use case, immunoassay estradiol sits close to noise.

For free testosterone, use equilibrium dialysis or ultrafiltration if you can get it. Both physically separate the unbound fraction before measuring it. The cheap “direct” analog free testosterone immunoassay is unreliable. Calculated free testosterone is acceptable and reproducible. It applies the Vermeulen equation to total testosterone, SHBG and albumin. That holds provided you always use the same equation and always draw SHBG at the same time. Mixing calculated values from one panel with direct-assay values from another produces a trend line that means nothing.

For total testosterone, ask for LC-MS/MS harmonized to the CDC standard. The commonly cited harmonized reference interval for healthy non-obese young men is about 264–916 ng/dL. Immunoassay platforms disagree with one another by enough to swamp a real change.

Whatever you order, write the method down for every result. A tidy long table with one row per measurement works well: date, marker, value, unit, method, lab, loinc, fasting_hours, time_of_draw, days_since_hard_session. If you change labs or platforms partway through, treat that as a break in the time series rather than as another data point.

Markers that mislead in trained people

Several common markers behave differently in people who train hard. Reading them against an ordinary reference interval invites the wrong conclusion. Four are worth understanding in detail.

The first is creatinine and the estimated glomerular filtration rate, or eGFR, derived from it. Serum creatinine scales with skeletal muscle mass and with creatine supplementation. A lifter can therefore sit at 1.3–1.5 mg/dL with normal kidney function and a CKD-EPI eGFR that looks like stage 2 or 3 chronic kidney disease. Cystatin C is produced by all nucleated cells and is essentially independent of muscle mass. That is why a cystatin C based eGFR is the marker to track in this population. Discordance between the two estimates is itself informative, and a persistent one belongs in front of a nephrologist rather than in a spreadsheet.

The second is the enzyme group of AST, ALT and CK. Resistance training, especially eccentric-heavy work, raises creatine kinase into the thousands and drags AST and ALT up with it. A “liver panel” drawn 48 hours after a heavy session during an untrained-to-training transition is close to uninterpretable. Draw after 5–7 days without hard training, and read GGT alongside ALT. GGT is not released by skeletal muscle, so an isolated ALT rise with normal GGT and high CK points at muscle rather than liver. Serum enzymes are a low-resolution readout of liver state in general. Plasma proteomic panels distinguish fibrosis stages in alcohol-related liver disease better than the standard enzymes do, which gives a sense of how much liver information the CMP is not carrying.1

The third is hematocrit. Androgen use raises hematocrit and erythropoietic drive, and the measured value is sensitive to hydration and posture. Standardize the conditions. Sit for 10 minutes before the draw, avoid fasting-induced dehydration, and keep the time of day constant. A 2-point hematocrit swing between draws is usually pre-analytic rather than physiologic.

The fourth is ferritin. It is an acute-phase reactant and rises with inflammation independently of iron stores. Interpret it only alongside an hs-CRP drawn in the same tube.

Pre-analytic control and the reference change value

Serial self-testing fails most often on pre-analytics rather than on the assay itself. The conditions of the draw therefore deserve as much attention as the panel design. Fix the following and hold them constant:

  • Draw between 7:00 and 9:00 am, since testosterone, cortisol and prolactin are strongly diurnal.
  • Fast for 10–12 hours.
  • Avoid training for 48 hours at minimum, and 5 days if you care about the liver enzymes.
  • Avoid alcohol for 72 hours.
  • Use the same lab, the same collection site and the same posture each time.

Even the fasting pattern itself moves cardiometabolic and inflammatory markers measurably, which is precisely why the protocol has to be identical between draws.2

Once the conditions are stable, stop reading single-point flags and compute a reference change value, or RCV. The RCV is the size of difference between two measurements that cannot easily be explained by analytical and biological noise:

RCV = 2^0.5 * Z * sqrt(CVa^2 + CVi^2)
# Z = 1.96 for two-sided 95%

Here CVa is the lab’s analytical coefficient of variation. You can ask for it or take it from the platform’s published imprecision. CVi is the within-subject biological variation listed in the EFLM database. Approximate values are around 9–10% for total testosterone and around 5% for creatinine. For TSH the figure is around 19–20%, and for ALT it is in the 20–25% range. On that basis, a TSH that moves from 1.8 to 2.6 is inside the noise floor, while a total testosterone that moves 35% is not. This single calculation kills most of the panic that direct-to-consumer panels generate.

Where a 60-analyte panel runs out

Even a well-run panel measures only a narrow slice of what circulates in blood. Clinical chemistry covers a small set of high-abundance, mostly liver-secreted proteins and small molecules. Plasma protein abundance spans roughly ten orders of magnitude, and albumin plus the top handful of proteins account for the overwhelming majority of the mass. The disease-informative content is concentrated in the low-abundance tail that no clinical panel touches.34

Mass-spectrometry plasma proteomics now has standardized acquisition and quality-control frameworks that make cross-batch and cross-lab quantification tractable. That standardization is the requirement for using the method longitudinally in one person.5 Pairing it with whole-genome sequencing also changes how you read the chemistry you already have. Lp(a) is largely set by LPA genotype and needs measuring only once. Pharmacogenomic variants explain a good deal of the individual variation in how drug and hormone metabolism shows up in serum.

There is one limitation that no panel design fixes. If you are using anabolic steroids, the reference intervals were not built for you. Much of the clinical literature also describes people who avoid disclosing use to their physician, which is a documented barrier to receiving competent care rather than an argument for self-management.6 The better path is to bring the raw results to a clinician who will read them without moralizing. Interpretation of any abnormal value, and any decision about what to do next, is a medical decision rather than a data-analysis one.

Questions people also ask

Is More Plates More Dates credible on blood work? On assay selection and panel breadth, yes, and it is better than typical primary care ordering. On “optimal ranges,” treat the material as opinion, since those cut points are not derived from outcome studies in the population being tested.

Is Marek Health legitimate, and how fast are results? It is a real telehealth and lab-ordering business that uses Quest and Labcorp collection, so the analytics are standard reference-lab quality. Draw-to-result is typically a few business days for routine chemistry and longer for LC-MS/MS send-outs, plus provider review time. Pricing is a la carte panels plus an optional monthly membership, so check their current published rates rather than any number you read secondhand.

What blood work do people on steroids get? Broadly the panel described above, run more often, with emphasis on the CBC for hematocrit, lipids with ApoB, liver enzymes plus GGT, kidney function via cystatin C, and the androgen and pituitary block including LH, FSH, SHBG and sensitive estradiol. Cadence is usually quarterly. What to do about any given result belongs with a physician.

Do lifters have high creatinine? Frequently, and usually because of muscle mass and creatine intake rather than kidney disease. Confirm with cystatin C before drawing any conclusion.

Woolf Software builds longitudinal molecular profiles of individuals: whole-genome sequencing, RNA sequencing, proteomics, blood biomarkers, and continuous glucose data, integrated into one model of you. Build your profile.

Footnotes

  1. Lili Niu, Maja Thiele, Philipp E. Geyer, et al. Noninvasive proteomic biomarkers for alcohol-related liver disease. Nature Medicine, 2022. https://doi.org/10.1038/s41591-022-01850-y ↩

  2. Iftikhar Alam, Rahmat Gul, Joni Chong, et al. Recurrent circadian fasting (RCF) improves blood pressure, biomarkers of cardiometabolic risk and regulates inflammation in men. Journal of Translational Medicine, 2019. https://doi.org/10.1186/s12967-019-2007-z ↩

  3. Marco Tognetti, Kamil Sklodowski, Sebastian Müller, et al. Biomarker Candidates for Tumors Identified from Deep-Profiled Plasma Stem Predominantly from the Low Abundant Area. Journal of Proteome Research, 2022. https://doi.org/10.1021/acs.jproteome.2c00122 ↩

  4. David H Geho, Lance A Liotta, Emanuel F Petricoin, et al. The amplified peptidome: the new treasure chest of candidate biomarkers. Current Opinion in Chemical Biology, 2006. https://doi.org/10.1016/j.cbpa.2006.01.008 ↩

  5. Xue Cai, Philipp E. Geyer, Yasset Perez-Riverol, et al. A standardized framework for circulating blood proteomics. Nature Genetics, 2025. https://doi.org/10.1038/s41588-025-02319-7 ↩

  6. Jason Kreutz, Michael Potemkin, Harshil Shah, et al. A qualitative exploration of family physicians and people who use anabolic steroids: Barriers to accessing evidence-based care. Performance Enhancement & Health, 2025. https://doi.org/10.1016/j.peh.2025.100319 ↩