These models exist to answer two recurring questions in our process: "how much does this single name move the portfolio?" and "given my discount and my conviction, what is the right size for this position?"
Both tools take simple inputs and return numbers we use as a sanity check before sizing or rebalancing. They are not forecasts — they do not predict returns, do not adjust for correlation, and do not capture the qualitative work that drives our actual decisions.
The qualitative work — building conviction, understanding the moat, reading filings — happens elsewhere. These sliders are the last 1% of arithmetic before pulling the trigger.
"Sliders cannot substitute for the work of building conviction in the underlying business."
— Atlas AM · how we use these tools