John Korsgaard / The Players
[Author's note: The Brewers are only the latest example of whiff kings. The idea isn't to barb them. The idea is to explain that what strikeouts take away from offense is different depending on the strategies of the day. That's whya comparison of the Brewers to the other teams is more relevant than whether its a new team record or not. The new team record is almost a random product of the times. One suspects that whoever is the worst stirkeout team next year will set another new record.]
by John Korsgaard
This record breaking thing about the Brewers and team strikeouts in 2001 is not being covered all that thoroughly in the pre-season magazines.
Strikeouts by hitters is something I do not believe is very well understood.
Two guys argue. I'll call them "whiff" and "wait"
Whiff: Hey what 's the big deal. An out is an out. Popping out ain't no worse than taking a good cut. The long balls are worth the whiffs. This ain't 1904 and you ain't Roy Thomas.
Wait: You been playing too much dice baseball. With two out and nobody on, it doesn't make any difference in dice baseball. But on the field, we are talking about a guy who has two strikes on him and makes no attempt to meet the ball. He should choke up and get his club a baserunner.
Whiff: Against Armando Benitez, you're going to choke up and dribble the ball to Alomar? You think that's going to help?
Wait: You're talking worst case scenario. Over the long haul if modern players would protect the plate with two strikes or force pitchers to make a good pitch on 2 and 2 .
Whiff: Hey the Brew Crew had a 426 slugging percentage. That's league average ..
This argument goes on and on. Largely, it is not resolvable because we will never have data for something as refined as attempted choke-ups verses attempted jerk-its in close games.
But the argument can have a lot of light shed on it. Enough light for GMs and managers to evaluate their lineups properly.
Three things are needed:
1. Contextual comparison
2. Willingness to approximate statistically instead of worrying about the ability to nail down every scenario to the ninth decimal point.
3. Understanding units
Instead of ten pages on theory, I'm going to go ahead and use the Brewers season to help us see all of this.
For context, we aren't going to hall out Roy Thomas or Eddie Yost or even Jawn Mize. Those of us (me included) who fondly point to what those guys meant to an offense will defer to you "up to date" "the modern game" folks on that point. So, here ya' go:
In 2001, the other 29 teams averaged 1069 whiffs with 194 homers. The BREWERS tallied 1399 and 209. [If you like further contextualization you might remove the Rockies, always abberant, and the Giants because of Bonds. The totals for 27 teams, however, are revealingly similar at 1070 and 192.]
I say revealingly because the 27 or 29 team results show how far off the Brewers are.
In essence, the Brewers homered 15-17 more than the average and struck out 330 more times.
Now, the Brewers pitchers tied for fourth in pitchers batting average and walks. They hit zero homers. 14 other NL teams hit 16 homers and the Rockies hit 10. Brewers pitchers whiffed 119 times. The average staff, including the Rockies was 115.
So, the Brewer whiff record was not about the ninth slot in the batting order.
The average whiffs and circuit clouts for the other eight batting order spots comes to:
119-23 other clubs
So, the question is: Was 41 more whiffs per batting order spot worth 3 more homers each?
What we have done here is placed the Brewers in context of 2001 and we have drawn some hypothetical approximations. It is not necessary to computerize every thing down to precise data. I love the computer formulas that we are now able to do. For "fun with history" they are wonderful. For many analyses they upgrade what we've had before.
But sometimes we can get the "concepts" we need even without all the math.
I'm here to tell you that the Brewers are simply a case in point of the need for clubs to get hold of some "concepts" or remain losers.
Let me tweek what we have just learned once more to see how obvious the need is.
Compare that hypothetical average to four main regulars:
The departed Mr Burnitz 150-34. Regulars 119-23. Do you trade a guy with 119-23 for a Burnitz with 150-34. Get 11 more homers but it cost you 31 whiffs. I do. Other things being equal, I'd give up Right fielder Joe Blow @ 119-23 for Jeremy Burnitz. Right field is an offense position.
Do I trade 1B John Smith at 119-23 for Richie Sexson at 178-45. That's a lot more strikeouts (59) but for double the homers 22 more. I think that's a good move.
What about shortstop Jose Hernandez? 185-25. Is 66 more whiffs worth the extra two homers? Do I want that out of the batting order if he can not correct it? You are able to answer that question.
Here's some additional data:
Jenkins was injured and in only 104 games was 120-20. That's a virtual wash with the league average. Five other guys played between a dozen and 21 games in LF. I don't have the breakdown but it's irrelevant because Jenkins has the job when healthy.
Blanco and Casanova caught in 158 games. They struck out 101 times and hit 17 homers. That was a small part of the problem but hardly the extreme. You take good defensive catching a little bit below league norm in production any time you can.
The point is that you are looking for OBA at lead off and you are hiding your defensive whizzes 7th or 8th in the batting order.
You are really looking at only 5 or 6 spots in the batting order.
The Brewers' 330 extra Ks with 17 more homers divided among the six middle spots of the batting order is in the vicinity of 55 whiffs and 3 homers per spot. Tweek that as much as you want, you have the same basic trade-off. Make it 5 positions, make it 7. Deduct Blanco and Casanova from the average catching staff. It just doesn't matter.
The 55-3 for six spots [or the 47-4 for 7 spots] is not the Burnitz or Sexson we thought might be a good exchange. That's much closer to Hernandez. It's not acceptable.
That brings me to the final thought. The units. I used to do consulting to small businesses. One of their great failures in analysis, next to not selecting a large enough sample, was failure to know what their units were. I emphasize units plural. Every arena of analysis has units rather than a single unit.
In batting average a main unit is at bats. Same for slugging and then we do plate appearances for on-base. I've always thought that pitchers' opponent BA, OBA, SLA were good. We combine that with another unit: seasons. That combo is our main ingredient.
For pitcher's we use innings pitched. Perfect. We don't use that for batters for the obvious reason that it is not an individual batter's unit.
On the other hand, one of Bill James' great advances, I always thought, was when he put his Runs Created into a context of "per 27 outs". The reason that was enlightening is that INNINGS is definitely a unit of team offense.
You've probably all seen the stats which show that the first inning is the highest scoring and the second is the lowest. Has to do with the construction of a batting order.
Once you recognize that an inning is a unit of team offense, these numbers take on more meaning. What does that differential of 330-17 mean in terms of innings contributed to and innings taken out of.
Not all innings are ruined by a whiff, Not all innings with homers contribute to a win. Not every one of 330 whiffs would convert to safeties by making contact.
We have to approximate. You can not say that you'd get 25 more runs on 17 homers which would win you 6-9 more games in trade for being no-hit 12 times resulting in that many more losses. Although I think 6-12 or 7-11 comes very close rather accidentally.
There's more ways to guess-timate at it than you can shake a stick at. I'm not going to put my favorite out here for you to shoot down. I can tell you this. If I'm a GM I'm not shopping for a guy who strikes out 165 times and hits 8.5 homers.
Have fun pondering all that.