There are two ways to run the same underwriting model, and they look identical from across the room. In the first, the asking price goes in early, the assumptions get tuned until the returns clear the bar, and the model's job is to explain why the price works. In the second, the assumptions go in first, formed from the leases and the market and nothing else, the required return is stated up front, and the price falls out the other end as a result. Same spreadsheet, same formulas, opposite directions, and the direction is the difference between underwriting and rationalizing.
The first mode has a mechanism, and knowing it is the defense. Anchoring is not a character flaw; it is how attention works. Once the ask is in the model, every soft assumption faces a quiet gravitational pull: rent growth drifts a quarter point, the downtime shortens a month, the exit cap compresses a notch, each adjustment individually defensible, none of them made for a reason other than the gap they closed. Nobody decides to cook a model. Models get cooked one plausible nudge at a time, by people who would tell you, sincerely, that they were just refining the inputs.
The second mode has a procedure. Build the model blind to the ask where you can: leases first, rollover assumptions per tenant, expenses from the actuals, the capital stack the deal genuinely supports, your required return stated before the first number goes in. Then solve for the price those inputs justify, and only then look at the ask. Now the gap between your number and theirs is information instead of an assignment. A small gap is a negotiation. A large gap is a fact about the market or about the seller, and either way it is not your model's problem to fix.
This direction of work changes what happens at the table, which is where it pays. An offer that is the output of documented assumptions can be defended assumption by assumption: here is why the vacancy costs what it costs, here is the rollover schedule, here is what the debt supports, and therefore here is the price. Sellers and brokers argue with numbers all day. They have a much harder time arguing with reasoning, and more than a few deals close at the buyer's price because the buyer was the only party in the room who could show their work.
Every page in this library is, one way or another, this page: vacancy as events, rollover priced per tenant, structure modeled honestly, exits stressed rather than hoped. PropCalc™ is built around the second mode, assumptions in, price out, fast enough that discipline does not cost you the deal timing. The asking price is the seller's opinion. The model exists to form yours.