The numbers work. The contractor finishes in eight weeks. The rehab stays on budget. The appraisal comes in where expected. Financing doesn't get more expensive. The property rents for what you projected. The tenant pays. Nothing interesting is discovered behind the drywall.

Every one of those assumptions may be perfectly reasonable. But here's a better question: "How many perfectly reasonable assumptions have to be right at the same time?" 

There's an old rule that applies particularly well to real estate: Murphy's Law. Whatever can go wrong, will go wrong.

I prefer a slightly less fatalistic version: whatever can go astray eventually will go astray. Not necessarily because you're incompetent. Not because your contractor is a crook. Not because the universe has selected your LLC for special punishment.

Because you're trying to predict the future. And human beings are terrible at that.

If we were really good at predicting the future, everybody would be a stock-market billionaire. For that matter, look at the futures people confidently predicted fifty years ago. By now we should be commuting in flying cars, vacationing on the Moon and letting robots do the housework.

Instead, we're still asking contractors why something that was supposed to take Tuesday through Thursday isn't finished three Tuesdays later. The future has an irritating habit of refusing to cooperate with the spreadsheet.

And there's another problem. We almost always overestimate the good. The renovation will probably go smoothly. We'll probably get that rent. The tenant will probably stay. Rates probably won't move enough to matter. The roof probably has another five years.

That's normal. In fact, it's probably necessary. If human beings spent their lives accurately contemplating everything that could possibly go wrong, we'd be paralyzed. Nobody would start a company, buy an investment property, get married or possibly even leave the house.

Optimism isn't a defect. It's part of what allows us to act. Entrepreneurs probably need more of it than most people.

The mistake is letting your underwriting share your optimism. You can be optimistic. Your spreadsheet shouldn't be. Your spreadsheet should be the miserable old bastard sitting across the table asking: "What if rehab is 20% over?" "What if it takes four months instead of two?" "What if rent is $200 lower?" "What if the appraisal comes in short?" "What if we have to carry it another six months?"

Not because all of those things are going to happen. Because something will. And problems have an unpleasant habit of breeding. The contractor gets delayed. That extends your financing. That costs additional interest. Now you're rushing the finish. Rushing creates mistakes. The mistakes cost more money. What looked like five separate risks were actually dominoes.

That's one reason experienced people can sound unnecessarily pessimistic. They've simply been around long enough to see one perfectly manageable problem have children. 

So I don't think the objective of underwriting is to predict exactly what will go wrong. You can't. The objective is to make sure you don't have to predict correctly. A good deal shouldn't require the contractor, lender, appraiser, tenant, market, weather and Almighty God to cooperate simultaneously.

A good deal gives you room to be wrong. You need optimism to recognize an opportunity. You need pessimism to underwrite it. And you need enough margin between the two to survive reality.

So here's the question for you: "How many things can go wrong before it stops being a good deal?" If the answer is one, was it ever really a good deal?