All-Star Games? Really? People still care about All-Star Games? Haven’t we collectively, as a society, properly slotted All-Star games, like batting average or pitcher wins, as the primitive measuring tool they are?
— Chubbs Peterson (@TigerWoodenhand) April 29, 2015
Since we vowed to try and objectively address each and every concern about Jeff Bagwell’s Hall of Fame candidacy, let’s take a look at All-Star Games…
Right away, we should mention that Bagwell, who played 15 years, racking up only four All-Star appearances is notable. And potentially dubious. There are only three players currently in the Hall of Fame with fewer All-Star appearances (Bert Blyleven, Ferguson Jenkins and Robin Yount; Don Sutton would tie Bagwell with four). More distressing, other than Yount, none were first or even second ballot selections (Jenkins went in on his third ballot, Sutton fifth, Blyleven 1,784th). And frankly, given the advancement in statistical analysis, it’s likely the candidacies of Sutton (who went in on the almost singular virtue of totaling 324 wins, an archaic milestone that would no longer carry as much weight) and possibly Jenkins, not to mention Yount, might be different if they were revisited today.
So, yeah – that’s uninspiring company, to say the least. And even if you reduce the significance of All-Star appearances to the silliest of criteria, it still speaks to a fairly unanimous opinion that Bagwell was not consistently among the 2 or 3 best first basemen of his era, which makes it more difficult to promote the idea that he was actually one of the 2 or 3 best hitters of his era. Like his somewhat overblown but nonetheless legitimate playoff struggles, this might be one of the best arguments against Jeff Bagwell’s Cooperstown credentials.
But please keep five things in mind as we venture into a deeper All-Star discussion:
- Every team in the league is required to be represented in the All-Star Game, regardless of merit, which can compromise choices, especially if you have a position (like first base in the National League during Bagwell’s era) that is stacked with worthy candidates.
- All-Stars starters – 8 each year – are selected by the fans; and while fans actually do a better-than-you-probably-realize job of picking starters (this year’s Royal-flush shenanigans aside)– it’s still a popularity contest that can be impacted by factors that might only cursorily relate to what’s actually happening on the field (like this year’s Royal-flush shenanigans).
- The roster has to be constructed by the guidelines of a standard baseball roster in order to play what is a standard baseball game. In other words, you can’t simply pick the 30 best players in each league but, rather, the 30 best players as dictated by positional needs. Sure, a team with 12 first basemen would probably rake – but defense is going to be a bit of an issue.
- Active managers are tasked with picking reserves, which further adds to the selectivity of All-Star rosters as they, too, will (consciously or otherwise) invariably introduce additional biases to the proceedings, including favoring either their own players, or players they see more often, based on divisional alignment
- All-Star rosters are constructed on the backbone of small sample sizes and have no year-to-year consistency.
Let’s examine this last point more closely because it’s important.
The Hall Of Fame was created to honor players who excel over an extended period of the time. The All-Star Game is, in many ways, the antithesis. It doesn’t consider past performance; it doesn’t recognize an entire season; it doesn’t even technically require a standard of quality, not when fans vote for starters and every team must send at least one representative.
Meaning, every season is its own unique creature and prone to unpredictable, unsustainable outliers which we’ll highlight shortly. In other words, you can’t truly draw any lengthy pattern from a series of completely random, unrelated outcomes.
Still, we get it – logging just three more career All-Star appearances than Alfredo Griffin is troubling so let’s try and find out what happened.
We’ll start here: we mentioned the small sample size (it’s actually less than a half season when rosters are finalized) and there’s a reason for that – Bagwell, for his career, was routinely better after the All-Star break*. Here are his career splits (tOPS+ measures how the player performed relative to their overall performance, with 100 being neutral):
1st Half: .291/.400/.531/.932 (tOPS+ 96)
2nd Half: .305/.417/.552/.969 (tOPS+ 104)
* Which, oh by the way, peripherally addresses another common complaint about Bagwell, that he was a regular season monster and playoff no-show. Given the Astros finished first or second in their division 11 times during Bagwell’s 15-year career, doesn’t having better second halves, when pennant races heat up, at least temper some of that?
Granted, he wasn’t *bad* in the first half – but the guy was prone to some amazingly ugly slumps and those seemed to, for whatever reason, often take place between April and June. Again, this is why sample size is important; a player, in theory, shouldn’t be punished for having a tough month. The mark of a truly great player encompasses the entirety of their career. All-Star Games measure 3 months.
In terms of each season being unique, let’s look at the ten seasons in which Bagwell did not make the All-Star team. (Note, it’s technically 11 seasons but he was on the disabled list in 2005 so we didn’t include it.)
NL All-Star First Basemen: Will Clark*, John Kruk, Eddie Murray (* denotes starter)
Bagwell’s First Half Numbers: .299/.387/.460/.847; 8 HR; 36 RsBI
Details: Right off the bat, we see how subjective this process can be. Clark (.858 OPS; 15 HR; 59 RsBI) was a worthy starter but Bagwell outplayed Kruk (.795) and, especially, Murray (.728). So, too, did Fred McGriff (.899; 16/53), by the way. In fact, McGriff’s snub was as puzzling as Murray’s inclusion – neither made a lick of sense. Unfortunately, the Phillies were better whole than their individual parts, and Kruk was their lone representative.
Verdict: Bagwell’s numbers were certainly worthy of consideration. Unfortunately, because of the roster requirements, you can’t bump Kruk for Bagwell without throwing the whole thing into tumult, even though Bagwell was probably more deserving.
NL All-Star First Basemen: Fred McGriff*, Will Clark, John Kruk
Bagwell’s First Half Numbers: .254/.356 /.413/.768; 9 HR; 51 RsBI
Details: All three players earned their place; McGriff was clearly the best of the bunch while Kruk and Clark both posted .900+ OPSs through the season’s first half.
Verdict: Bagwell’s sophomore season was a bit underwhelming after his Rookie of the Year campaign and he wasn’t deserving of being on the All-Star team.
NL All-Star First Basemen: John Kruk*, Mark Grace+, Andres Galarraga, Gregg Jefferies (+ denotes DH)
Bagwell’s First Half Numbers: .324/.393 /.524/.917; 14HR; 58 RsBI
Details: Oh boy. First things first: Kruk (.350/9/51; 1.008 OPS) and Galarraga (.391/13/65; 1.070 OPS) were runaway no-brainers. Grace and Jefferies, though, were a little less so. Given it was 22 years ago (prior to the advent of more robust analytics), we’re guessing Jefferies’ and Grace’s .343 and .332 batting averages, respectively, is what tipped those two over Bagwell, who had a higher OPS, as well as more home runs (and more RsBI than Jefferies – look, we know those aren’t definitive stats but no one’s doing a deep dive when constructing All-Star rosters; then or now).
Verdict: Among the five, Bagwell ranked third overall by OPS, which qualifies as a snub if you’re going to put four first basemen on the roster. But neither Grace nor Jefferies were unworthy by any means. Tough year (which underscores why it’s so silly to view All-Star appearances in total – there are legitimately distinctive factors from year-to-year that make it an incredibly unpredictable metric.)
NL All-Star First Basemen: Fred McGriff*, Mark Grace
Bagwell’s First Half Numbers: .260/.372/.438/.810; 10 HR; 45 RsBI
Details: As in 1992, McGriff and, more so Grace were both worthy selections. You could argue Eric Karros made more sense than McGriff.
Verdict: Bagwell was recovering from a broken wrist the previous August and had a dreadful start to the season (.183/.668 OPS in April/May with 4 HR – we told you the guy could slump). He was a worthy non-invitee.
NL All-Star First Basemen: Mark McGwire*, Andres Galarraga
Bagwell’s First Half Numbers: .278/.411/.556/.967; 19 HR; 55 RsBI
Details: In many ways, 1998 illuminates the desperate inconsistencies that plague Bagwell’s Hall of Fame candidacy. He was a muscle-bound power-hitter, they’ll say. But in 1998, when home run totals exploded – he “only” hit 34. At the All-Star break – the All-Star break –McGwire had 37. Sammy Sosa 33. Galarraga 28. At the same time, we have people deriding Bagwell’s lack of All-Star Game appearances – well, yeah – he wasn’t a muscle-bound power hitter! How is he going to surge past guys like McGwire, who, in the first half of the ‘98 season, totaled more home runs than Bagwell did in nine of his 15 seasons?
Verdict: Unbelievable that a guy with a .967 OPS wasn’t as All-Star – but that’s the world Bagwell inhabited in the 90s.
NL All-Star First Basemen: Andres Galarraga*, Mark McGwire, Todd Helton
Bagwell’s First Half Numbers: .294/.404/.569/.973; 23 HR; 64 RsBI
Details: Helton joined the party with authority, hitting .383 with 20 HRs. And McGwire was still treating baseballs with utter disdain, totaling 30 HRs and 69 RsBI with a ridiculous OPS of 1.230. The wild card – and, again, this is why a blanket approach to the All-Star Game can get tricky – is Galarraga. He was elected a starter (despite inferior numbers) after successfully bouncing back from cancer, which cost him the entire 1999 season.
Verdict: Strictly from a numbers perspective, Bagwell should have been an All-Star if the team was going t oselect three first basemen. He had arguably his best first half as a Major Leaguer. But McGwire and Helton – neither of which had better careers than Bagwell – went bonkers and Galarraga was an incredible feel-good story that transcended numbers. (And Galarraga’s numbers, btw, were very good. Not as good as Bagwell’s – but very good.)
2001 and 2002
NL All-Star First Basemen: Todd Helton*, Sean Casey, Ryan Klesko (2001); Helton*, Richie Sexson (2002);
Bagwell’s First Half Numbers: .269/.382/.523/.905; 21 HR; 68 RsBI (2001); .270/.378/.492/.870; 15 HR; 49 RsBI (2002)
Details: We’ve grouped these two seasons together because they illustrate one of the primary limitations of the All-Star Game. Casey’s (.329; .892 OPS) and Sexson’s (19 HRs; .882 OPS) first halves, while very good, were rewarded primarily because they were both the default best players on their respective teams and the All-Star Game mandates that every team must send at least one representative. Neither was a slam-dunk All-Star, by any means.
Verdict: This is where the exercise, fun as it might be, starts to break down a bit. Bagwell was better than Casey; in fact, by nearly every available measure, Bagwell was having a terrific season – but 14 years ago, that .269 batting average was more ghastly than it would be today. And he’s neck-and-neck with Sexson in 2003 but is ultimately punished for playing on a better team with a better roster that had more deserving candidates than Milwaukee’s. Hard to argue deserving – but certainly not undeserving.
Obviously, this exercise is subjective; we don’t mean to imply these findings are absolute. Instead, we wanted to shine a light on why All-Star Games are a terrible measure of a player’s greatness. They’re essentially small sample sizes of disconnected seasons in which wildly unpredictable outliers can skewer the results. You could reasonably argue that Bagwell was robbed of at least three more All-Star appearances (’93, ’98, ’00) and was a legitimate candidate in at least two more seasons (’91 and ’01). So it’s not like the game just passed him by. But how does he compete with a guy bashing 37 first-half HRs? Or recovering from cancer? Or being the best player – who also happens to play first base – on a mediocre team that nonetheless needs a representative?
Bagwell’s lack of All-Star Games is a slight mark against him as a legitimate Hall of Fame candidate and is worthy of mention. But anyone using it, and it alone, against Bagwell is purposefully being obtuse and narrow-minded. There are simply too many statistical tools available for us to lean on such a lazy measurement.