Some figures are impossible before anything is built. Some indicators are blind at exactly the point where a system fails. Floor1200 finds out which, before you sign or rely on it, and shows the calculation.
Send the figure or the indicator, the documents behind it, and the one question you need answered. You get a written finding with derivation, assumptions and sources. Fixed price per finding, agreed before work starts.
gernot.feldbacher@floor1200.ch
Why now: claims are certified and sold, and indicators are signed off, before the systems that must deliver them are tested.
Several reserves are watched through one ratio, score or traffic light. At some operating points that number holds steady or rises while one reserve falls to its wall — and geometry, not your weights, picks which.
The margin your indicator stops reporting, where, and under which conditions.
Send the indicator documentation instead of relying on backtests alone
You are asked to fund, buy or underwrite a performance figure, and the vendor's model is the only source.
Possible, tight or excluded — against the bound the process cannot beat.
Send the figure before the term sheet, not after the first shortfall
A certified object is optimised against long after the rule dropped the physical quantity that governed it.
Where the coupling breaks — through multipliers, dimensions, models or scope — located before the rule circulates.
Send the draft methodology before consultation closes
A readiness or load score blends reserves that fail separately — rig, crew, conditions, recovery. The score can look fine while one of them runs out.
The reserve your score stops showing, and when it matters in a campaign.
Send the scoring sheet before the season, not after the retirement
Also on request: margin audits of cyclic processes — pressure and temperature swing, drying, gas cleaning — and second opinions before signature.
For verifiers and rating firms: technical assessments as subcontracted work, under your name or ours.
Rescale every margin and every indicator, and the results do not change. That is why they hold for a separation plant, a balance sheet, a grid, a rule and a racing yacht alike.
The results are local and existential. Unlike a model validation, they do not test against your data; they tell you what no data can show.
A limited number of assignments per month, answered within two working days. We do not assess anything we hold a stake in. An anonymised sample report is available on request.
Floor1200 also licenses its published methods.
The figure or the indicator, the documents, and one question. Reply within two working days.
gernot.feldbacher@floor1200.ch