Pittsburgh Pirates' Fans' Gut Feelings Supported by "Sabermetrics"

Most Pittsburgh Pirate fans feel that Manager Clint Hurdle made the wrong move in asking Andrew McCutchen to bunt with one man on and nobody out in the bottom of the eighth inning Friday, down 3-2 against the Houston Astros.

"Sabermetrics," or baseball science, says that the fans were right and the manager was wrong.

Put another way, would a team rather have a man on first with zero out, or a runner at second with one out? Sabermetrics says it's the former.

The chances of a home team winning with a no out, a man on first, and a one-run deficit in the bottom of the eighth inning are 41 percent. Sacrifice him to second, incur an out and the chances of winning fall to 37 percent.

That's a swing of four percentage points. If a manager did this in every game of a 162-game season, the expected cost would be between six and seven games.

These probabilities are based on baseball history, and assume random batters, pitchers, fielders etc.

Now, it might be right for a PITCHER (in a National League park) to try to sacrifice. A pitcher might strike out, or ground into a double play, or hit into a fielder's choice that has the lead runner cut down at second and the batter safe at first. If the pitcher is going to get out ANYWAY, then let him try to make it a "productive" out.

A position player might do any of these unfortunate things, of course. But the average pitcher gets on base at about half the normal rate. Given this particular (non-random) prospect, it's a good move to try a sacrifice for the pitcher, but not for the position player.

Even so, you wouldn't want to do this if your pitcher happened to be a former Pirate and a .300-plus hitter named Ken Brett, or someone like Milwaukee's Yovani Gallardo. In this case, you would have to consider his chances of getting on base as good as that of any position player.

So you wouldn't sacrifice with a random POSITION player at the plate. And on the Pirates, Andy McCutchen is as "non-random" (best) as they come.

Except that Clint Hurdle did.

His rationale was that he wanted a (speedy) runner at second, with a chance to come home. If the score were tied, and the Pirates needed ONE run to win, that might make sense.

But the Pirates were DOWN by a run, and needed TWO to win. That's what made sacrificing McCutchen so wrong.

Because left to himself, McCutchen  could win the game right there by clearing the bases with a homer, as Ryan Doumit actually did in the third game. Or he could hit a long single to right that puts the runner on THIRD base instead of second.

With no outs, there are plenty of sacrifice combinations that can score a tying run. The chances of scoring a runner from second with one out aren't that great by comparison.

And even so, do you really want to tie the game just to go into extra innings and tire the bullpen? It's probably better to "take your chances" with the batter and win-some, lose-some. Because the result of the tie will be just that.

Clint Hurdle has been a great manager during his short tenure with the Pirates, but even great managers are fallible. Friday night was an instance of that. Fortunately, there was no repeat on Sunday.