Sweeney Out, Berroa (strikes) Out

What would the reaction be if the report is that the Royals drafted the best singles hitter, I wonder? How many would call for the firing of GMDM for drafting a possible Hall of Famer without much power, say, like the next George Brett?

One of the false dichotomies you are setting up is OPS vs. “HOF singles hitters” and power hitters vs. “HOF singles hitters.” I put “HOF singles hitters” in quotes because none of the guys you mention are slappy singles hitters. Brett’s career SLG was .487 including seven seasons over .500. And that was in a low-power era. Brett had a lot of power. He wasn’t a power-only hitter like Branyan or Kingman. Even Tony Gwynn’s career SLG was .459 including four seasons over .500. You included Ted Williams in this group for some reason even though his career SLG was a whopping .634. The reality is that there aren’t any good hitters who don’t have either a good OBP or SLG or both. But there are certainly bad hitters with a pretty good batting average. Rey Sanchez’s career .272 comes to mind, including multiple seasons over .285. The fact that he didn’t draw walks or hit almost anything other than singles made him a very, very bad hitter, despite his good BA.
I

’m not suggesting that anyone look at average alone, but to remember that all things offensively follow batting average. There are no HR in an 0-for-22 slump with a HBP and seven walks. A drop in batting average will mean less OBP and less SLG%.

No, no, no. Nothing “follows batting average.” Just because BA includes some of the things in more meaningful measures does not mean that BA is particularly meaningful itself. Getting hits is important. That goes without saying. But in what way can those hits be measured that gives appropriate meaning to them? One can measure them with batting average and get a nearly meaningless tiny bit of information. Or they can be counted in OBP and SLG for a much more meaningful measure of hitting ability.

Ninja, if you actually care about stats and value them, then you should read up on some of the work that has been done in baseball statistics in the last 20 years. Hanging your hat on batting average is a really poor way to evaluate any hitter. It’s only virtue is tradition. Thankfully serious baseball researchers have gone beyond tradition to utilize more meaningful stats and develop even more meaningful metrics.

I’ve got a feeling that you value Wins and Losses in evaluating a pitcher too. Am I right?

Lots of things fluctuate from year to year, and so do HR/PA, XBH/PA, #PA, etc. There have been "juiceball years" like 1987. Dunn’s big contract is going to be primarily valued on his track record of 40+ HR per year. I don’t believe that he could hit 40 HR in KC, and I don’t believe that 30-40 HR by Adam Dunn would translate to the 20+ more wins needed to compete in the division. It would probably be easier to pitch around him here, too.

Batting average, however, is far more fluky in the way it fluctuates. That’s why you get players such as Freddy Sanchez hitting .344 one year. Is Sanchez really good enough to do that every year? Nope. Sometimes one year more hits fall in, sometimes more of those line drives get caught, and this leads to batting average being a terribly fluky statisitic. Some “average” players every year have well above average years because of a fluky batting average spike, while others lose their starting jobs because one extra hit a week gets taken away and they drop from .290 to .260. It’s just not very consistent for most players, and it fluctuates with more deviation than secondary skills do.

Formal statistical analysis in all other fields involves identifying and removing outliers first. That is never done by those who analyze baseball stats, because it takes the fun out of it. In a sample of 600+ plate appearances, the home runs would be removed as outliers. SLG% is a poor tool by that standard since it goes up exponentially as HR data are added. That does not reflect my opinion of SLG%; rather, that if one is concerned with analyzing data for consistent patterns and studying what happens most of the time, one should not be so concerned with the HR or SLG%. AVG is a decent gauge of consistency, and that is more important in 162 games than 30-40 HR.

Good grief. No one’s talking about your “formal statisical analysis,” then. We’re talking about baseball statistics, and it’s just ridiculous to say that because stats are applied in a different way in some other field, we shouldn’t take home runs into account the way we do. One of the most important skills a player can have-if not the most important-is power. SLG percentage gives us a good idea of how much power the player is producing, as does Isolated Power (SLG-BA). Average is absolutely not a decent gauge of consistency. It does not correlate well with a player’s overall worth.

I’m not suggesting that anyone look at average alone, but to remember that all things offensively follow batting average. There are no HR in an 0-for-22 slump with a HBP and seven walks. A drop in batting average will mean less OBP and less SLG%.

NYRoyal basically covered what I wanted to say here in a thorough way. See his comment. I mean, really, you’re not making any real point here. A drop in batting average will mean less OBP and less SLG? Of course it will. But that doesn’t make it anywhere near as useful a statistic as the other too because it’s much more important to how often a player is getting on base or how much XBH he’s generating. If I were to evaluate a hitter on any one of these three stats, it’s a given I wouldn’t know the entire picture, but batting average would be BY FAR the worst choice of the three.

Just as a walk cannot become a HR, a K cannot become an error or a hit or a sacrifice fly as a contact out can. Had the 1986 World Series ended with a strikeout, history would have been quite different. At minimum, a contact out tests the defense and has value for that alone. That’s why strikeouts are good for pitchers—they advance no runners, produce no runs, and are defense-independent outs. Baseball is a double-entry system. A strikeout is among the best outcomes a pitcher can produce and among the worst results a batter can produce.

I said that strikeouts aren’t MUCH worse, not they weren’t worse. The fact is, Ks may be worse than other outs but they’re still dramatically overrated because they are yet another statistic that independently means zero (unless we’re talking about King 400 times or something ridiculous like that). A player can have a high k-rate, obviously, and still be the best player on his team or even one of the best in the league.

And you throwing out the last out of the 1986 world series, which is a grand total of one out out of millions recorded in baseball history…doesn’t mean anything to me at all. If Stanley had struck out Wilson, Buckner never would have committed the error, right? No shit. Stanley is far more to blame for the events of that inning than Buckner. Take it from a rational Red Sox fan.

Your hypothetical example was a player with 100 BB and 100 K’s, and I’ve already explained why 60 hits and 40 are better than 100 walks, so let’s get back to who I was writing about: Adam Dunn. He of the 190+ strikeouts in a season more than once. 2006 was one of those years. He did not have a 900 OPS in 2006, probably since 30 of those should have been hits to raise his batting average and OPS.

60 hits and 40 walks are better than 100 walks? Stat Ninja, you’re stating the obvious to an almost embarrassing degree. That’s not any point you’re scored against me, that’s a ridiculously obvious truism. However, it’s equally silly that you assume the trade-off would be “60 walks for 60 hits” when we’re talking about our 40:40 player and 100:100 player. That’s just not practical.

And as for the Adam Dunn point…that’s for pointing out a case study of what I was saying about batting average. Dunn’s BA fluctuating all the way down to .234 in ‘06, fluky low considering he’s managed .260 in the past. That was the only real difference between a 23 VORP and one in the forties, showing once again that secondary skills are usually more consistent than batting average.

With RISP, batting average matters most. (I’m not one of those who thinks that some players are "clutch" and thrive with RISP and others aren’t, aside from the fact that some may be better at hitting for average, but that’s another topic.) A walk doesn’t advance or score a runner on second base, but it does give the opposing pitcher a chance to get the very best possible outcome from his perspective (the double play). Same with a HBP, plus a chance of injuring the batter. BB and HBP are almost always better than an out, but not always better than a sacrifice fly, and never better than (but usually equal to) an error.

The most important thing a hitter can do in any given at-bat is not make an out. If a pitcher walks someone, that’s not a positive outcome and it can’t be assumed to be a positive outcome. Batting average, of course, doesn’t matter most with RISP. Being a good hitter overall does.

A player with a high batting average can contribute to more than 80% of his games played, and is more consistent than a low-average slugger. The higher the batting average (demonstrated over 1-3+ seasons), the more consistent the player.

You’re killing me. Absolutely killing me. This is an amazing statement. So you’d rather have Julio Lugo at .280/.350/.350 than someone with a .250/.380/.500 statline because of some made-up lack of consistency? Someone who can put up the latter statline will be a helluva lot MORE consistent than someone with the former line player b’s production is not entirely dependent on batting average, a notoriously fluky metric! Consistency as a hitter is not illustrated by batting average, it is much more clearly shown by all-encompassing stats like VORP or EQA.

Ortiz had a pretty good average (about 20+ points better than Dunn’s career avg) with the Twins except for the year that he broke his wrist. I’m a native Minnesotan, and every dedicated Twins fan I know knew that he could hit for average if he could only stay healthy. I keep telling my dad and uncle that he never could have done in MN what he did in Boston, mostly because of the lineup and ballpark effects, but that’s another story…

I meant Ortiz had never consistently hit .300 before coming to Boston, and obviously the Red Sox didn’t sign him for his “pretty good batting average.”

Sure. They spend 30-100 AB producing hits that Dunn spends striking out. The worst thing about excessive strikeouts is the opportunity cost which a skilled hitter could have brought to the team instead.

David Ortiz and Manny Ramirez are…surprise! Better hitters than Adam Dunn! No way! There’s no way I could’ve known that already! Stop telling me stuff I already know.

See Richie Sexson. How are the Mariners doing, by the way? Were they able to build a winning team around him while he was their big-money elite FA? I hope the Royals think about that very carefully.

Believe the numbers if you don’t want to believe me as I type this: hitting for average well is good.

Just how stupid do you think I am?

Maybe I’m missing what people are really arguing about here. So I apologize, Stat NInja, if I’m missing you’re point. But isn’t it quite well -established, aside from the logical points argued about above (with which I agree with the OBP crowd) that historically OPS, OBP, and SLG (whatever their limitation) all correlate better with runs scored than batting average? I mean, I think the logical/stastical arguments also carry the day, but the historical record proves it. Here’s a simple example taken from Neyer and Epstein’s Baseball Dynasties, pages 145-147.

Talking about the 1936-39 Yankees (they later both conclude that the ‘39 Yanks were tthe best team of all-time, but that’s not directly relevant here), they show the following chart of those Yankee teams and their ranks in AL batting stats (forgive the crappy chart):

1936 Avg : 3 Runs 1 OBP 1 SLG 1 OPS 1
1937 Avg: 3 Runs 1 OBP 2 SLG 1 OPS 1
1938 Avg: 6 Runs 1 OBP 2 SLG 1 OPS 1
1939 Avg: 2 Runs 1 OBP 1 SLG 1 OPS 1

Keep in mind that there were only 8 teams in the AL, so even being third in avg only put the team just ahead of middle of the pack, while being 6th shows them to be a “poor hitting” team that year when judged by avg. Yet every year they lead the Al in scoring. Now, it’s just an exampe. It isn’t definitive, and there are and will be aberrations, but note that, despite fluctuations in their BA rank, they always scored the most runs, and also had the highest OPS and SLG. Granted, this underrates the overall significance of OBP if taken on its own, but it’s just a historical example, a whatever else it proves, it shows the relative unimportance of BA compared to other stats. I think you could do a similar thing with recent scoring ranks, too.

and it probably means nothing when it comes to his overall value.

while at the same time proving it with your “challenge”.

You are absolutely correct in one respect:

Adam Dunn is more likely to strike out in a crucial situation than, say, Luis Castillo.

But anyone who’d rather have Castillo at the plate than Dunn in said crucial situation is a blithering idiot, because when not striking out Dunn is FAR more likely to produce than Castillo. There are tons of players who hit .280 while rarely striking out; that’s because they put the ball in play. When Dunn doesn’t strike out, he still doesn’t put the ball in play half the time. He either walks or homers.

Bottom line, you’re still stuck in the dark ages where people didn’t understand that the currency of baseball is NOT HITS. It’s OUTS. Your job as a hitter is to avoid making outs. Those hitters who avoid making outs the most frequently are the most valuable. A guy who hits .100 is still valuable if he’s got an OBP of .450, because it means that 55% of the time, he is not using up the one commodity which an offense has in limited quantity.