Junior Felix, right fielder, 1992 California Angels (team OPS+: 79)

After he was selected by the Marlins in the expansion draft, I once saw a pitcher throw Junior Felix 10 straight balls in one at bat. He struck out.

My parents went to games two Saturdays in a row one year, and between the two games they saw, he struck out 10 times.

I once saw a pitcher throw Junior Felix 10 straight balls in one at bat. He struck out.

This just cracked me up. It reminded me of some of the bad Giants I watched in the 70's and 80's but I am pretty sure that none- not even Dave Kingman, matched that feat.

In 2003, Miguel Tejada hit .239/.296/.398 (81 OPS+) in 361 PA as the #3 hitter. He hit .321/.379/.544 in 338 PA batting cleanup.

The A's #3 hitters collectively sucked again in 2005, to the tune of an 85 OPS+.

no tike redman?

This list is acceptable because Steve Garvey made the honorable mentions...

Any list that derides the Garv gets blanket kudos in my book.

"Derides the Garv"

NEW BAND NAME!

I know these are for full seasons, but the Torre era Mets had horrendous #3 hitters. I'll check Retrosheet, but I'm sure Bruce Boisclair was one of them.

I was surprised to see Herr 1987 on that list - surprised mostly because I had the lineup fixed in my head as Coleman-Smith-Pendleton-Clark. But I do see that Herr batted 3rd twice as often as Pendleton, and Pendleton mostly batted 6th.

There is a warning worth making here - that the 1987 Cardinals are a very dangerous team to be making any sort of OPS+ based conclusions on. The team OPS+ was 94, well below average and 3rd from worst in the league, but the team runs scored (in a near-neutral ballpark) were well above average and 3rd from best in the league. That's a huge discrepancy between OPS+ and actual runs. In more detail: that team was dead last in the league in HR (in small part a park effect but mostly that's who they were) and well below average in SLG, but first in the league in OBP. They're a testament to the true value of OBP. And they were a great, great basestealing team (248-72) and likely an excellent baserunning team otherwise. Four regulars or near-regulars on that team had "upside-down" stat lines with raw OBP higher than SLG: Coleman, Smith, Herr, and Oquendo. I'm not saying that's unprecedented - it's probably happened elsewhere - but it can't be common post-1920.

There's synergy there. Pile up too many high-SLG low-OBP guys in a row, too many Joe Carters, Mel Halls and Cory Snyders (or Pedro Felizes) in a row, and the whole is less than the sum of the parts, as no one is on base when the HR are hit. But pile up a bunch of high-OBP low-SLG guys in a row (like the '87 Cardinals) and, despite the number of LOB, the whole will be more than the sum of the parts. And a single power hitter in the middle of all that will have his value magnified. (And Jack Clark did slug .597 that year.)

Yes, the Cardinals would have been a little better off with the order as I remembered it - Pendelton third - then they actually had with Herr third. But even though Herr wasn't a particularly good hitter, he still had a better than league average OBP - and what else do you really want in front of Clark? And with two great basestealers in front of him, Herr's singles had value. Yes, there was slightly more power available with either Pendleton or McGee - but it also makes sense to put that slight power behind Clark, becuase you need more power to score Clark from first than you need to score Coleman or Smith after a SB. And yes, Oquendo would have made slightly more sense in the #3 spot than Herr even though Oquendo had even less power (but a .408 OBP.) For that matter, Oquendo would have made a better regular 2B than Herr. However, the real hole on the team was the lack of any semblance of a RF, and Oquendo wound up filling in there quite a bit.

Given how many runs that team actually scored for the raw offensive events it produced, I'd be very reluctant to cast any aspersions on its batting order.

Piling on a little: take Steve's "top 10" and ask how many runs they batted in while batting third. I'll show it as RBI/AB and AB/RBI:

50/451 9.0
47/388 8.3
51/398 7.8
27/360 13.3
75/505 6.7
58/520 9.0
32/332 10.4
50/463 9.3
41/392 9.6
37/364 9.8

That's Herr 1987 in boldface. Yes, that he drove in that many runs really tells you more about Coleman and Smith than about Herr, and yes, he's being compared to a stinko list. But just maybe, he doesn't really belong on that list.

And perhaps the real story about this lineup wasn't how bad the #3 hitter was but rather how good the #2 hitter was. Smith was .303/.392/.383. Upside-down OPB/SLG relationship and still a 105 OPS+, with a 43-9 SB/CS record thrown in. That's a lot better than a typical #2 hitter.

OK, it is here. Don't know how I missed it.

Oh, no doubt OPS+ is a crude metric, and teams at the extremes in any manner, especially a team like the Cardinals with tremendous baserunning (that OPS+ incorporates not at all) will be misread by OPS+.

But, dude. From the article:

The stat we're using to rank these stinkers is OPS+. Certainly, OPS+ isn't the be-all and end-all of offensive measurement, but, hey: This isn't a serious "study;" it's just for fun. One might quibble with the exact ordering of these guys, but even the blunt instrument of OPS+ has properly identified them all as hitters leaving something to be desired.

And even with that caveat, I see no reason not to stand by this:

Had Herr continued to hit as well as he did in 1985, batting him third wouldn't have been a terrible idea. But Herr didn't continue to hit that well; he didn't continue to hit anything close to that well. As we see on the honorable mentions list above, Herr's OPS+ plunged to 87 in 1986, yet Herzog stuck with him in the third slot. And in 1987 Herr's OPS+ was a mere 80, yet there he was still batting third, leaving runners on base instead of driving them home.

The Cardinals scored the second-most runs in the NL in 1987 despite Herr's performance in the third slot, not because of it. And while many of the bold decisions the very assertive Herzog made in his long and successful managerial career were good ones, the decision to deploy Herr as his No. 3 hitter was definitely not.

I was surprised to see Herr 1987 on that list - surprised mostly because I had the lineup fixed in my head as Coleman-Smith-Pendleton-Clark. But I do see that Herr batted 3rd twice as often as Pendleton, and Pendleton mostly batted 6th.

If you played RBI Baseball, you wouldn't have been ignorant of this important fact.

The Cardinals scored the second-most runs in the NL in 1987 despite Herr's performance in the third slot, not because of it. And while many of the bold decisions the very assertive Herzog made in his long and successful managerial career were good ones, the decision to deploy Herr as his No. 3 hitter was definitely not.

I strongly disagree with this. Herr was as good a choice to bat third on this team as anyone. Look at the alternatives: Pena was awful. Clark was the clean up hitter. Ozzie Smith was the # 2 hitter. Terry Pendleton was coming off back to back years with an on base percentage around .280. Vince Coleman was the leadoff hitter. Willie McGee had low on base percentages. and their right fielder was Curt Ford.

At least Herr was a good bet to have a good on base percentage ahead of Clark (which he did). Whitey made the right call - and it worked out, given the number of runs scored by this team.

Clark was the clean up hitter.

This was, what, carved on a stone tablet? Clark could have been moved up to #3, and everyone else moved up accordingly, with Herr dropped back to 7th or something. That alone would have produced more runs than batting Herr ahead of Clark.

Or, McGee could have batted third. Even in his down years he was distinctly more productive than Herr. Hell, Curt Ford was distinctly more productive than Herr.

The purpose of a #3 hitter isn't just to get on base, it's to get on base and drive in runners. This is especially true on a team like the '87 Cardinals, whose #1 and #2 slots were setting the table so efficiently.

The fact remains that the Cardinals scored as many runs as they did despite Herr batting third, not because of it. There were any number of plausible alternatives available to Herzog that would have been more productive.

Batting order doesn't matter a whole lot, of course. But neither is it meaningless. It has some impact, and the particular choice that Herzog made was clearly suboptimal.

Clark could have been moved up to #3, and everyone else moved up accordingly, with Herr dropped back to 7th or something. That alone would have produced more runs than batting Herr ahead of Clark.

First of all, batting your only power hitter clean up is not that strange an idea. Second, Tom Herr was not so much worse a hitter than the other players that you could say with any certainty that your suggested change would produce more runs.

Or, McGee could have batted third. Even in his down years he was distinctly more productive than Herr

I disagree. In his down years, Willie McGee has a significantly lower on base percentage than Herr. Given the makeup of the Cards, I'd rather have Herr bat high in the lineup than McGee - I'll take Herr's advantage in on base percentage (hitting in front of Clark) than McGee's small advantage in slugging.


Hell, Curt Ford was distinctly more productive than Herr.

He wasn't much more productive, and there was no way of knowing, going into the 1987 season, that he would be more productive at all. Not to mention you are now blaming Whitey Herzog for not batting a guy third who had a career on base percentage of .305.

Batting order doesn't matter a whole lot, of course. But neither is it meaningless. It has some impact, and the particular choice that Herzog made was clearly suboptimal.

It is in no way clear. In fact, given the extreme productivity of this offense, given the low overall OPS, that the lineup construction wasn't so bad.

First of all, batting your only power hitter clean up is not that strange an idea.

It is if you bat the second-weakest hitter among your eight starters third. Is there a law that says you can't bat your only power hitter third?

Tom Herr was not so much worse a hitter than the other players that you could say with any certainty that your suggested change would produce more runs.

Actually, yes, you can. While OPS+ is a blunt instrument, it isn't so blunt as to render a 10-to-15 point gap as anything less than a certainty of differential production.

In his down years, Willie McGee has a significantly lower on base percentage than Herr.

And, a significantly higher slugging averaage than Herr. Offensive production isn't only a matter of getting on base, it's the combination of getting on base and driving runners home.

McGee's small advantage in slugging.

It wasn't small. It was 103 points in 1987, and it had been 39 points in '86, and 87 points in '85.

It is in no way clear.

As a matter of fact, it is. Herr's production was distinctly lower than every other regular in the lineup except Pena. No lineup is optimally productive with its second-worst hitter batting third.

In fact, given the extreme productivity of this offense, given the low overall OPS, that the lineup construction wasn't so bad.

The extreme productivity of this offense given the low overall OPS was achieved despite Herr's contribution from the third slot, not because of it, just as it was achieved despite Clark's missing 30 games and Pena having a lousy year, not because of it.

Well, the ideal #3 hitter would probably have been Andy Van Slyke. Van Slyke in RF with Mike LaValliere catching. Unfortunately, that ship had already sailed.

The best opportunity on that team to drive in runs with singles only was in the #3 spot. And the best opportunity for driving in runs with a .410-.440 SLG was after Clark in the lineup. How do you "protect" a highly selective power hitter? You change the verb; it's not protection you want, it's revenge. You want someone to cash in the opportunities created by his walks. (And McGee, having merely a typical McGee year, had 105 RBI.) I can defend putting the hitters with a little power after Clark rather than before him.

And you already put in the article the primary reason not to bat Clark 3rd - you don't want him drawing walks with two out in the first. With that lineup, cleanup looks like the best place for him.

Herr's production was distinctly lower than every other regular in the lineup except Pena.

Uh huh. I don't know if you think there are any statistics worth looking at other than OPS+, but lets try runs created per game. Here are the runs created per game in 1986 and 1987 of the four real options to bat third:
Herr: 4.2, 4.3
McGee: 3.6, 4.3
Ford: 4.3, 4.2
Pendleton: 2.9, 5.0

So according to this stat (which is just as good as OPS+ if not better in my opinion, Pendleton was the only player who was distinctly better than Herr in 1987. And Pendleton was flat awful in the two prior years. Neither McGee nor Ford were more productive than Herr in 1987. And none of them were really better in 1986, either.

We can look at other stats if you want. The only stats that make the decision to bat Herr third look like a bad one are those that underrate OBP and overrate SP, such as OPS. Given Herzog's options, I would have done the exact same thing.

By the way Steve, I know the list was just meant to be fun, not a real analysis. And certainly Herr was not the ideal third hitter when compared to the options that other teams had. And he also wasn't much better than the other Cardinal options. But you make it seem as if it was insane for Herzog to bat him third, which to me is frankly crazy.

lets try runs created per game.

Huh? Per game? Why not let's try runs created per plate appearance, which removes the noise of partial game appearances:

Herr: .107, .109
McGee: .098, .118
Ford: .113, .113
Pendleton: .080, .129

Herr was in fact the least productive in 1987 of all these options, by this stat as well as by OPS+, and he hadn't been as productive as Ford in 1986.

But you make it seem as if it was insane for Herzog to bat him third, which to me is frankly crazy.

I haven't said it was insane. The ways I've described it are "definitely not a good decision" and "suboptimal," and I've said that the Cardinals scored as many runs as they did despite Herr's performance in the #3 slot, not because of it. I stand by all of that, because it's just the truth.

The best opportunity on that team to drive in runs with singles only was in the #3 spot.

Yes, but why should Herzog have been required to limit himself to driving in runs with singles only? The fact that that was the best place to drive in runs with singles only doesn't mean that singles only was the best way to drive in runs from that place.

And the best opportunity for driving in runs with a .410-.440 SLG was after Clark in the lineup. How do you "protect" a highly selective power hitter? You change the verb; it's not protection you want, it's revenge. You want someone to cash in the opportunities created by his walks. (And McGee, having merely a typical McGee year, had 105 RBI.) I can defend putting the hitters with a little power after Clark rather than before him.

And you already put in the article the primary reason not to bat Clark 3rd - you don't want him drawing walks with two out in the first. With that lineup, cleanup looks like the best place for him.

Yes, but all that argues for is Clark #3 and McGee #4.

Sure, you don't want Clark walking with 2 outs and the bases empty, but with hitters with OBPs as high as Coleman's and Smith's ahead of him, his 2-outs, bases-empty PAs would be far less than with almost any other team.

Basically, hitting Herr third to minimize that problem is a great illustration of solving one problem and leaving a bigger one in its place. The more times Clark, McGee, and Pendleton (and, yes, even Ford) came to bat, and the fewer Herr did, the better, and bunching the production of Clark/McGee/Pendleton (and, yes, even Ford) as closely as possible to the on-base-ness of Coleman and Smith, the better.

Huh? Per game? Why not let's try runs created per plate appearance, which removes the noise of partial game appearances:

Steve, are you really unfamiliar with the concept? Have you really never heard the term before? I thought it was pretty well known in these parts but apparently not. Its a statistic created by Bill James in the 1980's, which measures runs created per 26 or 27 outs depending on what formula you use. Its on baseballreference if you want to look it up. Offensive winning percentage shows the same thing. And once again, Herr came out about the same as McGee and Ford.

The ways I've described it are "definitely not a good decision" and "suboptimal,"

Fine. It seems obvious to me that Herr was just as good offensively as McGee and Ford - you might make a case that he was 1% worse, but I don't agree, and it doesn't make andy difference if I did. Calling the decision to bat Herr third as "definitely not a good decision" is just not supported by the facts.

Just the retrospective shudder on a trade that worked out really, really badly for the Cardinals: Tony Pena from the Pirates to the Cardinals in exchange for Mike Dunne, Andy Van Slyke, and Mike Lavalliere. The actual 1987 stats:

---
Pena: .214/.281/.307, OPS+ 55
Lavalliere: .300/.377/.365, OPS+ 98 (typical C playing time for both - plenty of work for a 2nd catcher.)
---
Van Slyke: .293/.359/.507, OPS+ 127 as a full-time player
Curt Ford: .285/.325/.408, OPS+ 92. Also Jim Lindeman, OPS+ 66, and John Morris, OPS+ 88, and David Green, 30 AB.
---
Mike Dunne: 163.3 IP, ERA+ 135

What a trade! And yet the Cardinals won the pennant anyway.

It's also worth noting that Herr was pretty solid throughout the 1987 season before falling apart in September. At the end of August he was hitting .282/.360/.365. I don't know how to translate that into an OPS+, but it looks like it's over 90.