The Value of Intermediate Milestones

Carl Milsted, Jr on May 7 20:25:27

View QTML

Originally published January 16, 2020. A few tweaks made when porting.

And thus flame wars, cat pics, and casino gambling all compete successfully for our attentions over more useful projects.

In creative politics just what is the value of convert won, a petition drive finished, or an ad campaign launched? None of these milestones by themselves change the laws and regulations of the land at all.

Let’s try breaking the Simple Formula for Success into pieces, to apply it to a multi step project which does have intrinsic value when completed. To keep the algebra manageable, we’ll make a a two step project. (This is more general than it looks if we think of the first step being our next step, and the second being all the remaining steps lumped together.)

The estimated costs simply add together:

$C=C_1+C_2$

The probability of success is the product of the probabilities of success for the two steps.

$P_s=P_{s1}{P}_{s2}$

Now suppose we have completed the first step. $P_{s1}$ is now 1. We now have a new opportunity with a higher probability and less expected cost, since$C_1$ is now a sunk cost.

$V_{ne1}=P_{s2}{V}_s-C_2$

Our improvement is:

$\Delta V_{ne}=V_{ne1}-V_{ne}$

$=\left(P_{s2}{V}_s-C_2\right)-\left(P_{s1}{P}_{s2}{V}_s-C_1-C_2\right)$

$=V_s\left(P_{s2}-P_{s2}{P}_{s1}\right)+C_1$

$=V_s{P}_{s2}(1-P_{s1})+C_1$

This is definitely a higher value opportunity since we have gotten a cost out of the way and the probability of success is up. How much better off are we?

By completing Step 1 we still have no intrinsic benefit yet, but we do have a better opportunity ahead of us. Our opportunity has gained value for two reasons:

1. The probability of success has gone up, since P1 is now 1.0.
2. The estimated cost has gone down by the amount we estimated for Step 1.

The new opportunity is definitely better than the old one. But we had to expend time/energy/money to get there. Are we better off? If so, by how much?

Suppose Step 1 is a deterministic step, such as filing for office. And suppose everything went according to plan so that the actual cost for Step 1 matches our original estimate.

By finishing up Step 1 we have a new opportunity whose expected value is better by exactly the same amount that we expended to get there. If we value the opportunity by its net expected value when done, we are not better off. Our gain is exactly equal to what we spent to get there.

Ouch!

Back to the drawing board! Value of an opportunity is not the same as the value when finished. Don’t count your chickens before they hatch!

The Value of an Opportunity

Suppose you were given a $10 coupon. And that coupon was for a store that you shop at frequently: your usual grocery store, Amazon, Walmart… Such a coupon would be virtually as valuable as $10 in cash.

But suppose that coupon was “Buy $10 of merchandise and get another $10 worth for free.” The coupon would be almost worth $10, but not quite.

Let’s go further: “Buy $100 of merchandise and get another $10 worth for free.” Now the coupon is getting noticeably less convenient than the equivalent cash. But it is still a valuable coupon.

Another factor of ten: “Buy $1000 worth of merchandise and get $10 more for free.” For me, such a coupon from Food Lion would be utterly worthless. I might be able to use it on Amazon with great effort, but even there I would not value the coupon nearly as much as $10 in cash.

Back to Politics

The net expected value of a project function much like the nominal value of such a coupon above. The buying requirement functions like the remaining expected cost.

Bang/buck needs to be part of the function for valuing an opportunity, but exactly what function is still beyond me.

Let’s put some numbers into a somewhat realistic political scenario to make it easier to visualize. A small city council race has an expected remaining cost of $10,000 (of time, money, unpleasantness,…). The expected net value of the project is $20,000. That is, if you (plural) expect to be $20,000 better off after the race if you slog on to the end. The overall bang/buck ratio is 2.0.

Your next step has an estimated cost of $1000 and a 90% chance of success.

You succeed, but it turned out the effort was a harder than expected. You expended $1500 of timemoney. Are you better or worse off? Should you continue?

Well, our original first term in the Simple Formula was $30,000 of expected value, since we expected a net improvement of $20,000. Our expected cost drops by our original estimate for Step 1, not by what we actually spent. So our expected net value for continuing becomes:

$\frac{\$30,000}{0.9}-\$9000$

$=\$33,333-\$9,000$

$=\$24,333$

The net expected value has gone up by over $4,000. Not too bad for spending $1,500 of effort, even though we had a cost overrun.

Our windfall comes from the reduction in uncertainty. Take note!

The bang/buck ratio has improved as well. It’s now 2.7. If your group is not burned out, do continue!

Even though I still don’t have a decent formula for valuing intermediate stages, there are some lessons here.

Do the Uncertain Steps First

Suppose you have some choice over which steps to do first. The lesson above teaches that you should do the most uncertain steps first whenever possible. Bonus points for those steps that are both cheap and uncertain.

Why?

For starters if you fail early you can quit before sinking too much costs. (Software startup gurus talk of Failing Fast.)

Secondly, the expected value function goes up nicely as the uncertain stages are crossed. This provides motivation to continue after the initial energy has been used up.

NOTE: here I mean uncertain steps of a project, not do uncertain projects first!!

Use Your Hardcore Activists for the Early Stages

When the going is uncertain, your borderline activists will sit on the sidelines. Deal with it. Don’t try to hype them into action with wishful thinking.

Use your hardcore activists, you true believers, to prove an uncertain strategy. If the results are good, then send out the screaming fundraising/recruitment letters to activate your dormant followers.

Pro tip: be honest with your hardcore base when pitching an uncertain project. Prepare them for possible failure. Complement them for being willing to take a risk. Don’t work too hard to motivate at this stage. If the mo isn’t there when the prospects are deeply uncertain, don’t do it.

(Case study: look up the history of the Free State Project. The original plan was to verify that enough people would move to the Free State before asking anyone to do so. Jason Sorens intended the project to shut down since they did not meet the original deadline. But the project refused to die and eventually they met the required 20,000 people commitment and 4,400 people have moved.)

Some Steps are General

Some early steps are prerequisites for many possible projects. For example, when you successfully educate people on the virtue of your programme, you increase the pool of volunteers, donors, petition signers, and voters – for a wide variety of projects.

This includes the project of starting a political party. There is much to be said for doing some kind of education effort first. There are cheaper ways to spread ideas and build mailing lists than direct politics.

Think Small

Are there projects that are intrinsically worthwhile that you can complete using existing resources? Do them! Even if the victories aren’t glorious, it is good have some actual victories.

Winning is a good habit.

Keep the Early Costs Low

But since so much power has been moved to the state and federal level, you do need to think big to have a real impact. That means long slogs, with many steps before policies get changed.

Since there is no political return until the endgame, keep the early net costs as low as possible. Don’t burn out.

Political activism can be a way to make friends, useful contacts, and build social skills. It can even be fun at times.

Postpone the annoying overhead tasks as long as possible.

Have fun.