May 19, 2019

Small Ball = Smaller Odds of Winning

Last May Clemson entered the top of 9th inning with a 7-2 lead against third ranked North Carolina in their second game of the ACC Tournament and, despite losing the opening game, the Tigers were in good shape to host a regional with a 5 run lead and 3 outs standing between them and their 40th win against 18 losses.

To that point in the game the Tigers had managed to scratch out 7 runs on 9 hits, 8 of which were singles. Small ball was winning.

North Carolina also entered the 9th with 9 hits, but the Tar Heels only had 2 runs to show for their efforts. The Tiger pitching staff had held a power-laden (20th in home runs and 31st in slugging) North Carolina team to one extra base hit to this point.

A walk, 3 singles and a sacrifice fly brought home a couple of runs, but Clemson still led by 3 with two outs and two on. The odds still favored the Tigers, but the great equalizer was waiting in the wings. As Brian Holberton’s bat met Scott Firth’s pitch and the ball sailed over the fence to tie the game all of the missed opportunities that plagued the Tar Heels that evening were erased.

Clemson went on to lose that game in 14 innings, was shut out the next day and relegated to traveling to Columbia for the second straight year for a regional.

For all the talk about the lack of offense in college baseball since the bat changes after the 2010 season power and slugging still rule offensively and teams that have those qualities have a much larger margin of error than small ball teams.

Much as I did with football last summer I took look at college baseball and again found some oft-repeated themes don’t meet the statistical test as “important” to winning and losing.  I looked at 23 metrics for all 296 Division I teams for the 2013 season. Some findings were mundane, some surprising.

With the current state of the bats in college baseball who would have guessed that stolen bases and sacrifices are far and away the metrics that have the least correlation to scoring runs?

Small ball seems to be what the majority of college baseball teams have turned to in the “dead bat era”. It’s not unusual to see the 3-hole or cleanup hitter sacrifice. Get a guy on, steal or bunt him over and hope someone knocks him in. Play for one run at a time. The power game is gone. There’s no use in playing for the big inning. Hang around. Keeep it close.

Except that’s not what the data shows gives you the best chance of winning.

On some level the statistics below are obvious and those metrics show that, not surprisingly, on the offensive side runs correlate highest to winning %.
Baseball Batting Metrics Pearson Wins

Once we have confirmed that runs lead to wins, we needed to determine how teams score runs by finding the correlation to runs for each of the other metrics. There’s also no surprise that hits is the metric that correlates most highly to runs.
Baseball Batting Metrics Pearson Runs

The surprise in the data is not at the top, but the bottom. Stolen bases and sacrifices are not only at the bottom, but they are not close to any of the other metrics in terms of correlation to runs.

With the aforementioned changes in the bat several years back it would seem logical to assume a higher correlation between runs, stolen bases and sacrifices, but that’s simply not so.

Sacrifices and stolen bases are what teams do to generate runs when they don’t hit as well or with as much power as other teams and not something better hitting teams do to score runs. Teams that score runs (and have better odds of winning) generally hit for power (slugging %, OPS).

To some extent coaches play with the hands they are dealt, field and personnel-wise. But many small ball coaches recruit small ball type players and it becomes a self-fulfilling prophecy on some level.

The findings above are not earth shattering, but it does confirm that a lot of sacrifices and stolen bases and not a lot of power (slugging%, OPS) reduces your chances of scoring runs and leaves these teams a much smaller margin of error.  Small ball  type teams have to capitalize on a higher percentage of opportunities on offense and rely more on pitching and defense.

So while small ball is often cheered and celebrated as indicating a “well-coached” team, it also means the odds of scoring runs, and therefore winning games, are reduced.

Looking at the pitching and defensive metrics it’s ERA that corresponds highest with wins which, again, is not surprising.
Baseball Pitching Metrics Pearson Wins

But what metric drives a team’s ERA the most? You might be mildly surprised (as I was) to find that it’s WHIP, which as the table below shows correlates extremely high to ERA. Not splitting the atom that the two are related, but I wonder how many of us were aware of just how closely these metrics are tied?
Baseball Pitching Metrics Pitchng ERA

On the defensive side a surprise is how low on the list fielding is. I’ve been of the belief that fielding is one of the most important metrics in driving wins and apparently I was wrong. There’s a correlation, but it’s a moderate level.

Hits/9 innings is also important, but WHIP dominates because walks and hits are low on the list individually, but the combination of the two per inning is a powerful indication of winning and losing.

So what’s all this about? It doesn’t take a rocket scientist to understand that scoring runs is tougher for teams that have little power and rely on sacrifice bunts and stolen bases to score. A walk, stolen base and single is much less dangerous than a walk, single and 3 run homer. You generally need 3 of the former to equal 1 of the latter.

But the drum beat is nearly non-stop that small ball is the way to win in college baseball with the current bats. These numbers indicate that not only is that not true statistically, but relying on these tactics could in fact retard scoring (by definition you are “sacrificing” a chance for a hit and almost always giving the defense an out when attempting to advance a runner by bunting in most situations).

The truth is small ball wins some games and some teams and coaches are better at it than others, which is a benefit when playing another small ball team. But the odds are against teams using this approach against a team with more power, even during the dead bat era.

These numbers suggest that disadvantage is bigger than most of us realized.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: