On Monday, The New York Times reported that “Republican leaders” responsible for targeting key races have pulled the plug on Mike DeWine’s senate reelection bid in Ohio. If true, I can only imagine the complex decision-making process that preceded such a move.

Committee targeting decisions are heady stuff, combining the bloody gore of emergency room triage with the drama of NFL draft day and a level of technical acumen that might land a probe on the moon. But at the end of the day, seasoned experience and wise judgment are the foundation of good targeting decisions.

For three prior election cycles, I was fortunate to be part of a team that did GOP targeting for Michigan State House races. The team was led by John Llewellyn, a former legislator turned operative. John would start each cycle with a list of targeted races, dividing them into categories according to degree of difficulty and opportunity.

The most important category of races, of course, was for endangered incumbents. The national and state committees that do the targeting and raise the money to fund races are principally devoted to incumbent protection. That’s job one. But identifying endangered incumbents worthy of committee assistance is not as simple and straightforward as it sounds. You can’t be too lenient or incumbents will get lazy and expect the committee to do all their work, especially fundraising. So you have to make your list of protected incumbents shorter than some may like. It’s the only way to save money and personnel resources for contesting swing and targeted takeover seats.

Swing seats are obvious targets, but you have to refine that category further. Llewellyn always wanted a mix of swing seats that lean Republican, Democrat and true toss-ups. He didn’t want us solely to be defending GOP-leaning swing seats, but rather to use the targeting process to go after Democrat-leaning seats, forcing the Democrats to play some defense. In this spirit, John always targeted one to three genuine Democratic seats. He was mainly interested in open-seats that traditionally elect Democrats, but sometimes he’d pick out a Democrat stalwart to go after.

The logic for Michigan’s House Republicans sometimes going on offense with their targeting was to force the Democrats to lavish money on protecting their incumbents, leaving less for them to attack our vulnerables. This was classic cold war spy-versus-spy stuff. They messed with us. We messed with them. I suppose there were few surprises because everyone had access to all the same polling and cash-on-hand financial data. Assuming that everyone was equally armed, there was some sort of mutually-assured détente, except in a few select races.

But despite the technocratic and empirical nature of the committee proceedings, it’s still sound judgment that serves as the basis for wise targeting decisions. You can’t just look at a list of poll numbers and say that everyone who is 5 points or more behind gets cut off. You can’t necessarily look at a listing of net cash-on-hand balances and see where opportunity lies.

I used to admire the care that Llewellyn put into his decisions. He’d drive out into swing districts and look over the yard sign war, visit with candidates personally, and listen to his campaign committee staff members’ anecdotes about what they were seeing and hearing. He would even ask he pollster to look beyond the topline numbers to spot trends or patterns that might suggest opportunity or impending disaster.

John Llewellyn successfully protected the House GOP majority during his time at the helm. Now he’s moved on to a real job in the real world. But like me, he’s probably following the decisions Republican targeters make in D.C., hoping they have the wisdom to make sound choices. A lot depends on it.

Hill is director of Hill Research Consultants, a Texas-based firm that has polled for GOP candidates and causes since 1988.