In the top of the 13th inning, the Red Sox were able to push home a run on an infield hit that basically was no different than a bunt. However, seeing eye infield hits that act like a bunt are not counted as a sacrifice.
So of course sabermetric statistics on sacrifice bunts are going to produce SLIGHTLY lower runs per innings scored than when a sacrifice is not done because anytime an out is not recorded, the batter is not credited with a sacrifice.
When the Dodgers were able to score specifically because the runner at first base, Max Muncy, was able to advance from first to second base on a tag up when the Red Sox third baseman Eduardo Nunez landed in the stands making a catch on a foul ball, was a sacrifice credited on the play? I just checked the box score and this was not ruled a sacrifice either. Amazing, the next play would have been a game ending play at second base if Muncy had not advanced on the caught foul ball that put Nunez in the stands.
If Box scores are going to ignore plays that mimic exactly what a sacrifice does, and does not give a sacrifice credit when a baserunner advances, then the stats are going to skew towards runner on first base, no outs, as scoring more runs per inning than a runner on second base, one out.
And lets not forget the real comparison is not whether or not a runner on first no outs scores SLIGHTLY more often than a runner on second base with one out, the REAL TEST is does a runner on second base, one out, score more often than a conventional inning. If a runner on second base by way of a sacrifice results in more runs scored than no one on, no one out, then a sacrifice still INCREASES the likelihood of scoring a run, and that is the real test that should be used to valuate a sacrifice.