It was impractical to hit shots from the fringe, fairway, or rough because no human (not even Perfy, my putting/chipping robot) could hit the flagstick often enough or accurately enough to run the test in a reasonable amount of time. However, by precisely rolling balls on a green from a short distance, I could measure how the flagstick affected the results. To guarantee measurable, reliable results, I used a putting machine called the "TruRoller," which I invented to roll balls precisely controlled directions at carefully controlled speeds. For each test, I set the TruRoller about two feet from the cup and measured 1) how far the ball rolled past the hole when the hole was covered, 2) how many putts stayed in the hole when the hole was not covered and the flagstick was out, and 3) how many putts stayed in the hole when the flagstick was left in.
Each test was run at three different speeds: On a perfectly flat green, the speeds were fast enough to send the ball three feet past the hole, six feet past, and nine feet past. Each test also included putts that approached the target at different parts of the hole: dead center; left- and right-center of the pin; left and right edge of the pin. Finally, the tests were run, first on level greens, then on ones that sloped sharply uphill and downhill. (The speeds remained consistent, but because the slope changed, the balls, if they missed, would finish considerably farther away on downhill putts and closer on uphillers. But it is the speed, not the final distance from the hole, that matters.)
All told, TruRoller launched thousands of "shots" at the hole, an equal number with the flagstick in and out, on a number of different greens, at five different parts of the hole. Once that was done, PGA Tour veteran Tom Jenkins, the former lead instructor at my short-game schools, did his best to duplicate those tests. Although Tom couldn't control his putts as precisely as the TruRoller, I felt it was important to compare machine and human results. Tom hit more than a thousand putts, the results of which supported the TruRoller's results.
Of course, there were variables in conditions, including imperfect green surfaces, the edges of the cup becoming ragged and worn, the hole being higher in back than in front and acting as a "backstop," and so on. But over thousands and thousands of putts, these variables were more than compensated for. What did I learn? Leave the flagstick in whenever the Rules allow, unless it is leaning so far toward you that the ball can't fit. Here are a few special cases.
- Perhaps most surprising, when the flagstick leans either slightly toward the golfer or away, the odds of it helping to keep the ball in the hole increase: With the flagstick leaning away from the golfer, the hole becomes effectively larger; when the flagstick leans toward the golfer, the ball rebounds downward, again helping shots find the hole.
- Only in the most obvious case, when the flagstick is leaning so far toward the golfer that there isn't enough room for the ball, is leaving the flagstick in a bad idea. Check the flagstick before you chip to be sure it is sitting properly in the cup. (The Rules of Golf prohibit you from positioning a flagstick to your advantage. But you may leave a tilting flagstick as is or else center it in the hole.)
- Even if you don't hit the flagstick dead center, it still will aid you. It proved especially advantageous when chipping downhill and at faster speeds. I even believe the flagstick should be left in when you're putting from an inch or two off the green in the fringe. The flagstick will help you make more putts unless it is leaning severely toward you or it's so windy that it is moving and might knock your ball away.