Saturday, 10 August 2013

Least Valuable Sluggers (LVSs)

Home runs are a nice thing. More home runs = better player, right?

Clearly I think the answer is no, or I would not be writing this. There are other factors involved. Like every other thing that player does - defence, GIDPs, baserunning, walks, strikeouts, etc etc etc. As usual, let's turn to WAR.

Here are the 10 least valuable 50 home run seasons of all time:
 Those are all good seasons. But considering 50+ HR, not great seasons. Wha happened?

*cough*defense*cough* - defence is going to come up a lot during this post.

Prince Fielder should be a DH, especially as he ages over the duration of that monster contract. He is entertaining to watch as he plays first base (the easiest position), but is not especially effective. In 2007 baseball reference rated him as 15 runs below average at first (better than -22 in 2006 when he posted a -0.7 WAR). That hurts.

Prince still walks a lot, hit hits for average, and he slugs. So it's not a terrible season by any means. But still, 50 home runs is supposed to be legendary, right?

What about the others on the list?
Out of 10 seasons, there are only 6 players, who fit 2 profiles:
1. Slugging first baseman who should be a DH because their defence is so bad
(Mark McGwire, Prince Fielder, Ryan Howard, David Ortiz - already a DH)
2. Slugging corner outfielder who should be a DH because their defence is so bad
(3x Sammy Sosa, Greg Vaughn)

Also, notice that every one of these seasons occurred between 1996 and 2007. 7 of 10 of these seasons occurred within the 5 year period 1996-2000.

WAR is a context neutral stat, that is, it compares players relative to their peers. When the league as a whole hits a lot of home runs and scores a lot of runs, the relative value of a home run decreases. Sammy Sosa hit 63 home runs in 1999* and was worth less than 5 wins to a 67-95 Chicago Cubs team. That's crazy - except it's not. Sosa played bad defence, did bad baserunning, and did all his damage during one of the two craziest (the other being the 1930s) offensive eras ever.

*I have two favourite Sammy Sosa stats:
1. Sammy Sosa hit more than 60 home runs three times, in 1998, 1999 and 2001. He led the league in home runs twice. Which years did he lead the league in home runs? This is easier because of the above chart. But anyways, the correct answer is: not during any of these three seasons! He led with 50 in 2000 and 49 in 2002. 
2. Sammy Sosa is listed at 165 pounds on his baseball reference page. Um... no.

Moving on, here are the 10 least valuable 40-49 home run seasons of all time:

None of these seasons are "replacement value" yet - but we are getting close. How are these players so un-valuable despite their slugging prowess?

Well yes, its the other parts of baseball, but it's fun to look at how those parts can overshadow what is ostensibly a good season, isn't it?

First up is Adam Dunn. What a fascinating baseball player. In probably the most boring way possible. He hits home runs, he walks, he strikes out. He plays poor defence, if he is allowed to play defence at all. He is third all time (behind Mr. Derek Jeter and Gary Sheffield) in defensive runs below replacement. And he is not very valuable considering he hits lots of home runs. The only reason he did not lead the first list is probably that he has never hit 50 home runs. He hits 40 home runs. Every year*. Dunn is probably not a key part of a good baseball team. Over his first 106 games this year (leading up to today), he has hit 26 home runs** and has a WAR of 0.0.

*Adam Dunn hit exactly 40 home runs for four consecutive years 2005-2008. 

**26 home runs in 106 games projects to exactly 40 home runs over 162 games. Okay, 39.74.

But Adam Dunn seasons don't really appear to be great seasons. At least not to me. "In 2006, Dunn hit .234 with 40 home runs and 92 RBIs, and struck out 194 times" - even Joe Morgan could tell that this is not a particularly great season, even with the home runs.

Most of the other seasons on this list match that pattern - pretty low batting average, high strikeouts. Some of them have very high RBI totals, which could be sort of confusing and confounding. Most/all of these seasons are accompanied by terrible defensive ratings from 1B or corner outfield.

One number in that table sticks out like crazy. One year, really. And that is Dante Bichette's 1.2 WAR Triple Crown attempt in 1995. Dante hit .340 (but since Tony Gwynn was still in the league he finished 3rd) and led the league with both 40 home runs and 128 RBIs in only 139 games*. He even led the league with 197 hits. And again, he finished with 1.2 WAR. If Mike Piazza and Tony Gwynn were not awesomer at batting average, Dante Bichette could have won the Triple Crown and probably would have won the MVP**. Am I wrong?

*Strike-shortened Schedule

**He finished 2nd to 5.9 WAR Barry Larkin

This was the result of a brutal 18 runs below replacement at defence, poor baserunning, and a new ingredient: Coors Field, the greatest offensive ball park since the Baker Bowl*, or maybe ever. Since I know what's coming next, I am going to leave my rant until later. That's called foreshadowing**.

*280 ft to the right field wall

**it's not

Here are the 10 least valuable 30 home run seasons of all time:

Now we are really getting somewhere. There are nine seasons where a "replacement" player would have been more valuable than the 30 home run slugger on the team. I have a lot of digging and commenting to do about all of this, so try to stay with me. There has to be a lot of interesting stuff in here.

10. Richie Sexson (+0.1)
The 1999 Indians were one of the monster offensive teams we have ever seen, scoring 1009 runs. Sexson played 1B, LF, RF and DH poorly. He didn't walk - his OPS+ was only 102 (2% above average). He hit home runs and his teammates were pretty spectacular at getting on base - six had OBPs of .397 or better. They are the only team in history that had 6 players with an OBP over .395 (100 games or more each):

By WAR, Sexson was the 15th most valuable position player on this team. But the rest of the team is more interesting than Sexson.

The '99 Indians juggernaut lost to the Red Sox in the first round of the playoffs in a crazy and memorable playoff series. Cleveland went up 2-0, then lost games 3 and 4: 9-3 and (yes) 23-7. This was the steroid era. Pedro Martinez left game 1 with a back injury. In game 5, he entered the game to start the fourth, with the score already tied 8-8. This was the steroid era. Pedro pitched 6 scoreless no-hit innings and the red sox won 12-8.Wow.

9. Tony Batista (-0.2)
Tony was a crazy guess hitter. I have never seen a batter swing at so many terrible pitches he had clearly decided to swing at before the ball left the pitcher's hand. By 2004, he was a one dimensional third baseman. His fielding actually wasn't that bad. His negative WAR, as far as I can tell, can mostly be attributed to Batista getting on base at a .270 clip. That's bad. His OPS+ was 81.

.241 with 32 home runs and 110 RBIs doesn't sound that bad - but if that's all you get, then yeah, it's bad.

8. Cory Snyder (-0.4)
In 1987, I did not know about baseball or that Cleveland had any Indians. (In)famously, Sports Illustrated picked Cory Snyder, Joe Carter and the Cleveland Indians to win it all. They finished last in the major leagues at 61-101. Oops. Negative WAR from Snyder could not have helped.

Snyder regressed during his second season, and his 33 home runs were not enough to offset a .273 OBP with 166 strikeouts. Snyder never fulfilled any potential and retired with a career .291 OBP.

7. Adam Dunn (-0.4)
We've already discussed Dunn - apparently when he doesn't hit 40 home runs, he doesn't have any sort of value at all. But, home runs are probably the best way to get paid for providing about 0 value.

6. Joe Carter (-0.5)
As a Jays fan, I'm sad to see aging hero Joe Carter on this list. We certainly didn't realize it at the time - Joe Carter drives in 100 runs every year - but his lack of on base skills certainly limited his overall career value.

Carter retired after the 1998 season. After his legendary home run to end the 1993 World Series, Carter actually provided -1.8 WAR for the rest of his career despite averaging 24 home runs and 90 RBIs for his age 34-38 seasons. Carter went negative war 6 times in his 16 year career. This makes me kinda sad. I didn't think that Joe was a hall of famer, but I thought he was a pretty great outfielder.

But he did hit that home run.

5. Leon Wagner (-0.6)
I'm not gonna lie - I've never heard of Leon Wagner before. In this 1964 season, he did post an OPS+ of 107 with those 30 home runs, so that's pretty good. It seems that this negative WAR season is largely attributable to terrible defensive ratings - 19 runs below replacement that year as a low value corner outfielder, and 79 runs below replacement over his 11 year career. I don't really know who is putting together defensive run metrics for players from the 1960s, but I personally will take them with a grain of salt.

4. Tony Armas (-0.9)
Is this Tony Armas Sr.? Yeah, apparently. Well, that is interesting in itself. In 1984, Armas hit a league leading 43 home runs to go with his 121 OPS. But it's 1983 on our list. That year, he hit 36 home runs with a dreadful .254 OBP en route to an 85 OPS+. Ouch. To top it off, he was a below replacement outfielder that year.

Which leads me to a different question, that has sort of been percolating for this entire post. And which will likely lead to another blog post. When we consider a "replacement level player", are we considering an overall player? Or are we considering, of the whole pool of players, replacement level defence and replacement level offence separately?

I'll try to explain what I mean. I think that of all the AAA+ players, including major leaguers, a good defender is easier to find than a good hitter. That makes the defensive standard higher for the players who are good hitters - they have to match the standard of the easily-found defensive specialist. Hmm. More on that later.

3. Dave Kingman (-1.0)
I read a great piece about Dave Kingman once that made me look up his career stats. It's a strange career. I'm not sure how much this happens, but he had one season that was much better than all of his other seasons, his 1979 season when he led the league in home runs, slugging and OPS.

His last season was 1986, and he mostly played (played?) DH. The man obviously had power, to hit 35 home runs in his age 37 season. He signed on with the Giants the next year but couldn't get out of the minors and retired. His below-replacementness was clearly noted, despite his home run power.

The 1986 Oakland A's went 76-86 with 21 year old Rookie of the Year Jose Canseco.
The 1987 Oakland A's went 81-81 with 23 year old Rookie of the Year Mark McGwire.
The 1988 Oakland A's went 104-58 and lost the world series to Kirk Gibson and the LA Dodgers.
The 1989 Oakland A's went 99-63 and swept the Giants in the earthquake World Series.

That team changed the game and probably launched the steroid era. But that's a topic for a different post.

2. Mike Jacobs (-2.0)
This only happened 5 years ago, so I'm kind of disappointed for not knowing about this already. Mike Jacobs was a power hitting first baseman who might be done with MLB by now. The Diamondbacks gave him 13 games last year, and he gave them -0.2 WAR. His career WAR totals go in the wrong direction:

He starts good, gets worse, get's even worse for his monster (monster?) 27 year old 2008 season, then gets less worse as he fades into retirement. Which I find very interesting. Is it just me?

Most of the negative value comes from terrible defensive ratings: -24 runs as a first basement is pretty tough to overcome. A .299 OBP is not really worth keeping on the field for bad defence, even with the power.

1. Dante Bichette (-2.3)
I already got to write a little bit about Dante Bichette, since his 1995 season appeared on our 40+ home run LVS list. But his 34-year-old 1999 takes the LVS award home for good! Congrats! (congrats?)

In a 14 year career, Bichette played what I would call full seasons 12 times. In those years, he averaged .301/.339/.504 with 23 home runs and 93 RBIs. That's good, right? He accumulated 5.7 WAR in those 12 seasons. That's pretty insubstantial. He was 95 runs below replacement as a corner outfielder.

The numbers are a mirage of course because he played most of those years in Coors Field pre-humidor. In his age 26-28 seasons with the Angels and the Brewers, he averaged .259/.293/.409 with 12 home runs and 51 RBIs during 118 games played. That's a first clue.

The Rockies traded for Bichette as they prepared for their first season. He was readily available - that's the definition of a replacement player, right? Over the next seven seasons in Colorado, he totaled .316/.352/.540 and averaged 29 home runs and 118 RBIs. At Coors Field he hit .360/.397/.640 - it was a crazy place!

Let's check out some of the Rockies's home splits before 2002, when the humidor was installed:

And now the same list of players's away splits:

A crazy place.

Okay, but more about 1999 and LVS Dante. He hit 20 of those 34 home runs at home, but that still leaves 14 for the road. Not the worst. It's a boring reason, but the reason that 1999 is on the list so high is because of... a bad defensive rating - a whopping 34 runs below replacement as a corner outfielder. Coors Field was a hitters park in part because they put the fences back to compensate for the thin dry air that made baseballs travel so far. That left a lot of space for LVS Dante to cover.

The 1999 Rockies scored 906 runs - but they allowed 1028. They were a disaster, going 72-90 to finish last in the NL West. Their hitters totaled 1.5 WAR - including a 5.1 WAR Larry Walker*. They allowed 1028 runs with a pitching staff that totaled 21.7 WAR - what a terrible environment to be in.

Staff ace Pedro Astacio happened to be my fantasy baseball secret weapon that year - I started him on the road only and he was awesome. Overall he had a 5.04 ERA in 232 innings for a 5.9 WAR. On the road he went 12-6 with a 3.60 ERA. Of pitchers with an ERA over 5, Astacio has the highest WAR by far, with 1938 Bobo Newsom in second at 4.8. The ERA standard has to be lowered to 4.45 in order to find a higher WAR, 2000 Brad Radke at 6.2 (also during the steroid era).

Which is all to say that players on the 1993-2001 Colorado Rockies probably collected some of the most misleading statistics there are - with Dante Bichette's weird 1999 leading the way.

Well, that was fun.


  1. On behalf of first basemen everywhere: ouch!

  2. But how many of these guys can stretch it like you?