How Best to Fail

[Wally Mo Pena]

Runs are created by offensive production and this for the most part refers to what players do when they don’t get out. But to stop an investigation of offense there is to ignore what in my mind are a few of the great things about baseball:

1) just how many ways there are for a batter to fail,
2) the different effects this can have on the game, and
3) the fact that all teams have to fail a certain amount.

In a 10-0 no-hitter blowout, the two teams can look different in every statistical category except one, the number of outs. It makes sense then (because of #2 above) that the team on top may have done a better job using it’s outs to help the offense to those ten runs.

As such, there has been a lot of talk lately on these boards and throughout the baseball world about how teams make their outs, and how this relates to their offensive production. I support this fully, but I haven’t liked the approach that most people take in these examinations, which is to look first at the number of outs a team makes of a certain type, and then compare this with the runs scored.

You get results like this (2005):

NYA 2nd in ground ball (GB) outs, third in runs.
TEX 1st in fly ball (FB) outs, 2nd in runs.
CIN 1st in strikeouts (K), 5th in runs.

It’s clear that you can get your outs in any way and still score a TON of runs, because of what I said all the way back in my first sentence. Runs have a lot more to do with what a team does when they don’t get out than they do with how the team makes it’s outs. If you had a team that K’d for every out they made but Slugged 2.000, I’m going to guess they’d still break every record around. Same for GB or FB, or the category I’m calling given up outs (GU) which consists of times caught stealing and sacrifice hits.

The way I think this issue should be approached is by looking first at the number of runs that teams (plural) score, and then examining whether trends exist in the type of outs made by the high scoring teams and the lower scoring teams. This will show any small differences that may exist in the understandably small effect that outs have on offense. So what I’ve done here is to stratify the major league teams into three groups: the top ten scorers, the middle ten scorers, and the bottom ten scorers. I looked at how these teams made their outs for the this year, and the previous two, and came up with an average across these years of how many outs of each type these teams made per game played.

(note: when I say a looked at the top scoring teams across the three years, I am not looking at the teams that have scored the most over the last three years combined, but rather the average of the top ten scoring teams for each year. The same is true for the middle and bottom groups).

The results are displayed below in average outs per game of each type.

..........................GB FB K GU
Average top 12.30 10.55 6.25 .47
Average mid 12.25 10.13 6.36 .59
Average bot 12.59 9.54 6.52 .63

The results for each individual year are very close to the overall average, and the trends are maintained. PM me for all of the numbers.

So what can be made of this? The top scoring teams make about 1 more FB out per game than the lower scoring teams, strike out less than the middle and low scoring teams, and give themselves up less than both other groups.

The GB outs are a little less clear because the middle group has fewer than the top scoring group, but the bottom group has the most. The main conclusion that can be drawn from these findings is that teams that score the most runs consistently make more outs in the air than teams that score less. The lower scoring teams seem to make their outs in different ways, with no strong trend, which is to say the lower teams aren’t making one less FB out and one more K per game or anything that obvious. The one less FB is fairly evenly distributed in the other types of outs.

This makes sense, in that double plays are less likely to occur on FBs than GBs, and runners can be advanced on FBs where they usually cannot be on Ks. Also the higher FB total might reflect teams that make better contact and hit balls harder.

Since this is a very simple examination, only simple conclusions can be reached, but these can still be helpful. Overall, it seems a FB is the best way to get out, and this may in the end be of less importance to offenses than it is to pitchers. A batter can easily overcome a few extra ks or GBs if thier non-out numbers are good, but unless a pitcher can balance FBs with high K numbers, these numbers would lead one to believe that they will not fare well overall. Eric Milton agrees.

[ghettochild]

[Saber]

The moral of the story is to stop *****ing about Adam Dunn striking out so much.

[missionhockey21]

Very interesting article Wally. I really appreaciate the thoughts you bring to the forums with your articles. I am one who uses individual stats for strikeouts, groundouts and flyouts to evaluate a player but never as indepthly as you have here to analyze their effect on a team's performance. You deserve reputation for this article, keep up the good work!

..........................GB FB K GU
Average top 12.30 10.55 6.25 .47
Average mid 12.25 10.13 6.36 .59
Average bot 12.59 9.54 6.52 .63

About what I expected here but I didn't expect it to be so close (I don't know why I didnt, but still). The top teams, many (but not all) often rely on power in their lineup so it's understandable to see how they have more flyball outs. And the bottom tier teams often are made up of players with weaker plate discipline so I can understand the higher amount of outs via the strikeout.

This makes sense, in that double plays are less likely to occur on FBs than GBs, and runners can be advanced on FBs where they usually cannot be on Ks. Also the higher FB total might reflect teams that make better contact and hit balls harder.

Flyballs are the most productive out in my book. The double play chance is virtually gone, there is always a possibility for a misplay in the outfield and if there are runners on base something can possibly happen in terms of advancing them. And like you point out, the flyball generall comes from players who hit the ball hard. Few power hitters will have a greatly skewed ratio of groundouts to flyouts (at least not that I am aware.)

Anyway, great article again Wally. Just curious though, while compiling the statistics, did you notice any anomalies in the groups? Teams that just really didn't fit with the teams they were with in terms of the stats you were looking at? I am not suspecting there were or weren't any, just interested.

[awefullspellare]

Very Very entertaining read. Well worth the time! Thanks for sharing the info. GB outs can be good, when you are smart with them, but I agree that FB is the best.

[The Philly Fanatic]

Another very interesting piece to read of your's.

"A batter can easily overcome a few extra ks or GBs if thier non-out numbers are good, but unless a pitcher can balance FBs with high K numbers, these numbers would lead one to believe that they will not fare well overall. Eric Milton agrees."
You make a good point here but I think a lot of it has to do with the situation. Eric Milton is a pitch who leaves pitches hanging and a result he gets crushed, especially since the Great American Ballpark is a cozy stadium for homerun hitters. But despite that, I think the GB-FB stats can be quite helpful to show just how dominant a pitcher is where with many batters it isn't quite as important. A dominant pitcher usually has a favorable ratio of grounders to flyballs where there are good hitters that have all sorts of ratios when relating to those stats. I was concerned about Ben Sheets awhile ago for while still having a good season, his strikeouts were down and his flyballs were up. Again, I believe a lot of it has to do with situation (would Eric Milton have more success in larger ballpark? probably) but these stats are very useful in my opinion to show just how dominant a pitcher is in his outings and getting grounders and strikeouts usually is the optimal thing to do.

Good article though Wally.