Saturday, January 5, 2008

Scoring Baserunners in the NL: 2007

Obviously, the main goal of a team on offense in baseball is to score runs. Teams with a lot of home run hitters have a leg up on the competition in that regard since they probably don't have to bother with moving runners over quite as much. Of course, no matter how many home runs a team hits, they have to also get some of the players they put on base around to score in order to have a chance at winning games. Using the Play Index, it's possible to quickly get a list of how many times each team put runners on base during the season (totaling the Times on Base from each game). It's also easy to figure out how many runs the team scored from those baserunners by taking total runs scored and subtracting the team's home run total (since, obviously, a home run doesn't leave a runner on the bases).

In the following table, "Baserunners" refers to the team's total Times on Base minus Home Runs and BRR stands for "Baserunner Runs," my term for runs scored by players who actually were on the bases (derived from R-HR). BRR% is then the total percentage of baserunners that came around to score (BRR/Baserunners).

New York2032627.3086.342
Los Angeles2019606.3001.337
St. Louis1996584.2926.337
San Francisco1901552.2904.322
San Diego1889570.3017.322
NL Sums315569503.3011.334

It seems pretty apparent (and should be obvious) that teams with higher on base percentages end up with more baserunners. Milwaukee seems to be an exception, but the Brewers also led the league in home and had over fifty more than every other team with a sub-.330 OBP. Thus it's perhaps understandable that they had relatively few baserunners for their OBP.

In any case, it's interesting to note that the percentages of baserunners scored seems to bounce around a lot, especially near the bottom of the list. Remember, however, that teams that hit a lot of home runs will get some benefits here. Though I took out runs scored by the batters who hit home runs, I didn't take out the runs scored by baserunners on home runs. Here's a chart of BRR% in order to show visually which teams succeeded and which teams struggled.

(click image to enlarge in a new window)

With a couple exceptions (Pittsburgh and Arizona), the teams that were in contention for the playoffs are on the left side of the NL Sums (average) bar and teams that didn't are on the right. I'm not at all sure this reflects any sort of skill on the part of teams beyond getting on base at a better clip, but perhaps the next image should help us tell. It's the same chart as the last picture, but it also has each team's OBP next to their BRR%.

(click image to enlarge in a new window)

I'm sure the scale used makes it look more dramatic than it really is, but it's obvious team OBP bounces around even as BRR% declines. I don't think this reflects a team skill any more than, say, leaving fewer men on base than the rest of the NL teams (more on this at a later date, but every team in the NL left between 58 and 61 percent of their baserunners on base in 2007).

As final charts before messing around with hypothetical stuff, it might be worthwhile to see what teams were the best at scoring the "smallball" way. As I alluded to above, teams like the Brewers and Reds who hit a lot of home runs have perhaps skewed BRR% numbers since I only took out runs scored by batters who hit home runs, not the runs scored by baserunners at the time. Let's look at the chart of the percentage of baserunners scored in ways other than via the home run.

(click image to enlarge in a new window)

As suspected, Milwaukee and Cincinnati take a hit, falling all the way down to third and second lowest. The rest of the league jumbles up as well. Pittsburgh rises to third, while Philadelphia falls to the middle of the pack, among other changes. (Note: all the images used in the bars for each team are the background images from their official site which explains why the Cardinals are blue).

Finally, here is a chart of both the overall baserunners scored percentage and that of baserunners scored other than on home runs. I think it is good proof of how the baserunners scored percentage fluctuates wildly when you take the home runs out of it:

(click image to enlarge in a new window)

This entry has gotten pretty long, so I'm going to leave you hanging and say I'll post again tomorrow with some numbers derived from playing around hypothetically with data from today's post. As a preview, if the Houston Astros had scored even a league average percentage of their baserunners, they would have scored 31 more runs on the season. Those 31 runs would have changed their Pythagorean Winning Percentage from .447 (72-90) to .466 (75-87).


tlb said...

I don't mean this in a rude way, but why do you want to evaluate offense without the HR? Honest question. It would follow from that logic that you could derive something from evaluating offenses by filtering out BBs, 2Bs, Ks, or GIDPs (or any single element).

Perhaps you could figure something out from these factors, but wouldn't all of them lead you to the same place?

It honestly doesn't surprise me that, when HRs are taken out of the equation, you can't really predict or identify offensive impact - or, at least, you can't find the teams who, in reality, had the bigger impact near the top of your sample group.

Since batters' BA w/RISP seems to fluctuate from year to year (except in rare cases), wouldn't it follow that a team's success in driving in runners (whether or not you filter out HRs.. but not the runs driven in by them?) would be sporadic as well?

Again, I'm sorry if I come off like a jerk. I just really don't understand what filtering out HRs would or wouldn't tell you about offense.

TheJay said...

There's always this theory bandied about that smallball teams who don't rely on home runs are better at manufacturing runs in other ways so that's why I took HR out of the equation. I personally don't agree with that line of thinking but I thought it'd be interesting to see if anything changed with homers included or not.