So, over the last couple of days I’ve heard a bunch of people here in Silicon Valley talk about how we might abolish the Electoral College. It’s not an unreasonable impulse coming from people for whom Disruption is a religion – and who live in a state that voted two-to-one for Clinton.

But here’s the catch: in this sunny and magical land, we take for granted the ability to disrupt without asking permission of the disruptees. Unfortunately, the Electoral College isn’t some lumbering incumbent; it’s a structure codified in the Constitution. And the Constitution, while routinely ignored and sometimes trampled on, is very specific about how it might be amended, requiring three-quarters of states to give a nod.

And why is that a problem? Well, let’s look at the numbers. In the table I’ve enclosed at the bottom of this post I’ve compared each state’s population (as a percentage of the nation’s) with its percentage of electoral votes and defined this ratio as “Thump Value.” So a state like Colorado that has 1.67% of the nation’s Electors and 1.67% of its population has a Thump Value of 1.00. California, on the other hand, has 10.2% of Electors and 12.1% of population, resulting in a Thump Value of 0.84. At the other end of the spectrum, Wyoming has 0.56% of Electors and 0.18% of population, resulting in a Thump Value of 3.03.

Now, why does this matter? Well, as it turns out, there are 32, ahem, states (31 excluding DC, which isn’t a state to the dismay of some) with a Thump Value in excess of one. These guys are overrepresented in presidential elections relative to their population; their swagger comes out of the hide of the most populous states.

Since a “no” vote by a mere 13 states can shut any amendment down, it’s important to note that abolishing the Electoral College would mean that 20 states would be voting against their self-interest (some more than others, to be sure). In fact, 14 states have an electoral influence that’s at least 50% greater than their population might suggest.

So what does this have to do with venture capital? Well, I sat there pondering the odds of Wyoming’s delegation voting against their state’s interests at the Constitutional Convention. Then I got to thinking about Vermont’s odds, and North Dakota’s and Alaska’s and Maine’s, etc. And it reminded me that the joint probability of all these states voting against the status quo that granted them extra thump in presidential elections was the product of all their probabilities multiplied together, a pretty minuscule number.

This fundamental arithmetic of probabilities can be tough to internalize, even for savvy startup founders.  Every now and again, I find myself pointing people to Josh Kopleman’s seminal 2006 blog post, Domino Rally Business Models.  By way of quick summary, in this decade-old post Josh was bemoaning business models that required a bunch of disparate events to all turn out favorably for success to be probable. For example if a business required four things to happen, each with a probability of 75%, the odds of success for the business were about 30% (0.75 X 0.75 X 0.75 X 0.75 = 0.316); needless to say, rarely are the odds of any one thing happening that high.  Unfortunately, these Domino Rally business models, where success is predicated on a series of binary outcomes or big leaps, are endemic to venture. While companies that thread the needle can be rewarded handsomely, most entrepreneurs radically overestimate their chances of success because they think of risks individually, not in a compounded fashion.  Moreover, there are two kinds of risks: those you can mitigate and those you can’t.  As consumers of risk, we investors have to work really hard to mitigate those that are mitigable and get properly compensated for those that aren’t.  Founders may squawk, but we’re all well served to heed Buffet’s equation: Opportunity = Value – Perception.  Keeping the hype in check to maximize Opportunity around the fulcrum of value is one way to make chances of success higher than the chances of abolishing the Electoral College ever will be.

Thump Value by State, 2016