All-Time All-Mariner Roster: Shortstop
Alex 1996. You get a 20-year-old Alex Rodriguez at shortstop, finally freed from the crazy vagaries of Lou’s favor and given a full season. He hits .358/.414/.631. If not the finest, certainly one of the finest offensive seasons by any shortstop ever. It’s stunning.
To put this in some perspective, in 1996 the other leaders in raw counting stats were:
Mark McGwire, immobile first baseman, .312/.467/.730
Frank Thomas, immobile first baseman, .349/.459/.626
Jim Thome, not entirely immobile first baseman, .311/.450/.612
Edgar Martinez, DH, .327/.464/.595
Alex ran offensively with those guys in his first full year in the league, at 20, as a shortstop.
It took three years– good seasons, to be sure — before he even approached that kind of production again in 2000. And those three down seasons are the 3-5th-best shortstop seasons in M’s history.
And at the same time you got that stellar offensive production from a premium defensive position, you got the super-primo Alex defense of 1996. I know it might be hard to imagine now that he’s at third, even having lightened up a little from last year, but Alex was ridiculous when he was that young, outstanding in all aspects of his defensive game as well.
I feel a little like we’re having the first baseman debate again, in a way: “What if you could have AD’s offensive presence…”
“…except he’d play good defense…”
“I like, I like…”
Yeah. Alex 1996. The best season by any Mariner ever makes him my starting shortstop.
1996 Alex Rodriguez was great. 2000 Alex Rodriguez was just a little bit better.
Trying to parse out their overall performances is basically useless. 1996 A-Rod was worth 56.5 runs over an average hitter, while 2000 A-Rod was worth 57.3 runs over an average hitter. Yep, the entire difference in offensive value comes out to 0.8 runs. The value of their batting lines couldn’t be any more equal, even down to the plate appearances (677 vs 672). There’s just one difference.
Kingdome vs Safeco Field.
As we noted, the Kingdome wasn’t the offensive paradise it’s remembered as, so I’m not downplaying the 1996 version of A-Rod, but we have to realize just how amazing it was that Alex had the kind of performance in Safeco Field that he did as a right-handed batter. We know the park is brutal on right handed power hitters, and even in his prime, even A-Rod couldn’t overcome the monstrous alley in left center field. Check out his home/road splits from 2000:
Home: .272/.406/.502, 330 PA, 18 2B, 2 3B, 13 HR
Road: .356/.433/.702, 342 PA, 16 2B, 0 3B, 28 HR
Now, I’m not arguing that 2000 A-Rod was going to post an 1.135 OPS if not for Safeco Field, but we have to acknowledge that he spent his last year in Seattle in the worst park on the planet for his particular skillset. As good as he was in Seattle, he was Ruthian as soon as he got out of the Northwest. Park adjusting the numbers doesn’t work in a park like Safeco, where lefties and righties are treated extremely differentliy by the park’s dimensions. In reality, it is far, far harder for a RH slugger like Rodriguez to thrive in Safeco, and the fact that he was able to put up an equal line to his 1996 performance while playing in the Safe makes his season more impressive.
Where’d you get the value over average hitter? In terms of value over replacement, VORP has it as 112 for 1996 and 102 for 2000. That seems odd, but it’s not — the AL shortstop in 1996 hit .269/.326/.408. In 2000, they hit .279/.345/.427.
And yet even as I type that, I feel a little uncomfortable writing that. League offense didn’t change that much, while the people playing shortstop got a lot better. Alex still played a premium defensive position, still hit just as well, and yes, he was in Safeco Field instead of the Kingdome.
It’s like — if we were talking about a great catcher who put up huge lines along with a host of others and kept going as they all collapsed, would we value those later years when their contributions stood out more? If you say yes, then it’s 1996 for you. And if it’s no, then we’re back to the argument.
I don’t know that I’m ready for the park issue. Not to return to my admiration for Mike Cameron’s career here, but how much do we boost someone for playing in a park that’s not suited to them? Are we picking the best season, or the best road season? If there were two Mike Camerons, say, one left-handed and one right-handed, who played in Safeco Field in 2001 and 2002, and the left-handed one put up huge numbers on the home and on the road, and the right-handed one was the Cameron we know and some of us appreciate. Which one had the better season? It’d be the lefty, right?
If the park suited Alex better, and he performed better overall, then I have to say that’s the better season.
But I think I’ve come around a little — I would have said it’s absolutely 1996 when I wrote the intro, and now I’m at a point where I think either choice is entirely defensible, where you can make the case that you set aside the increased competition at short, and it’d be a pick ’em.
I wish we could compare defensive numbers. Was a 20-year old Alex that much more awesome than the more experienced 24-year old version? (“Nope,” says the BP fielding numbers, “and by a ways, too!” “No BP fielding numbers,” you’ll reply.)
I miss seeing Alex play so much.
The value over average hitter is from Baseball Reference – it’s BtRuns in the “Special Batting” category, and it’s based on Pete Palmer’s linear weights formula. For various reasons, including the one you mention, I’ve given up on VORP.
As for the park issue, one of the main things we’ve tried to advance over the years is the idea of how context effects performance. I don’t think we can turn our backs on that fundamental principle now. Put another way, if you built a time machine and traded 2000 Alex Rodriguez for 1996 Alex Rodriguez, the results would change drastically because of the home park, and your vote would shift to a different player (you know what I mean) even though the only difference was the ballpark. You’re not voting for 1996 A-Rod – you’re voring for the Kingdome.
Or, to put it in another from, who had a better pitching season, 2003 Ryan Franklin or 2007 Felix Hernandez? Franklin threw more innings and prevented runs at a better rate, which is where value comes from. Of course, we understand that the ’03 M’s defense was outstanding while the ’07 M’s defnese was abominible, so we’d correctly assess that Hernandez had the better season, even though Franklin had the superior results, because of the context adjustment. Same thing here.
I think we absolutely have to make corrections for degree of difficulty. 1996 A-Rod simply had an easier path to achieving the same results as 2000 A-Rod. Therefore, 2000 A-Rod wins.
I entirely agree on context affecting performance, and I see what your point is. But I come down on the opposite side of what you might expect here. This may well reveal that we have differing assumptions about what we’re selecting for. I would vote for the swapped Alex.
To take an extreme example, say that instead of the Kingdome/Safeco Field debate we had Superkingdome and SuperSafeco Field — one that favors righties as much as any park in history and the other punishes them as severely as any park ever.
Alex will put up two dramatically different lines:
in Superkingdome, he’ll hit .350/.550/.950 with 50 home runs or something ridiculous
in SuperSafeco Field, he’ll hit .300/.380/.450 with 15 home runs (the adjustment here… never mind, it’s not important).
On the road, they hit exactly the same.
Sure, the difference in their season lines is the park, but in one, Alex is a huge superstar who helps his team as much as any player ever, and in the other, he’s — well, he’s still Alex, but he didn’t do things that scored runs nearly as often.
If you’re building a team to play in Safeco Field today you’ll have a strong preference for left-handed hitters with power. Given two equally talented options, you wouldn’t pick a right-handed player and then, when he doesn’t do well, claim that performance-wise he was hampered by his park and your team should be awarded an extra ten park-adjusted runs.
In terms of who had the best season, I think it’s entirely fair to consider how they performed in the park they were in, and –
Which brings me back around to Alex. In arguing his home/road splits:
Year AVG OBP SLG OPS
1996: .352 .405 .619 1.024
2000: .356 .433 .702 1.135
That’s not such a huge gap in any event. But I don’t see why we should count the performance of those 81 games on the road, performed in parks with a certain set of characteristics that more or less form a neutral environment, while discarding the 81 played at home.
I feel like I’m struggling to make the argument I want here. In a way, it’s like considering clutch hitting in MVP discussions. Clutch hitting isn’t a hitter skill, but if someone hits way better than you’d expect one season and as a result the team scores 10 more runs than they would have normally, then that’s worth considering as part of the sum of what they did. Similarly, though the park is not something a player can do anything about, their accomplishments in a season reflect how they did in that environment.
Alex in Superkingdome scores tons more runs than Alex in SuperSafeco even though they may overall be similar run-scoring environments.
I sense an impending chastisement for results-based analysis. I press on!
I see this differing from the pitching example in a significant way. A hitter, or pitcher, can be well-suited for their environment. I think, for instance, that it’s possible to consider Ryan Franklin’s seasons in a fly-ball friendly park to be better seasons than his horror-show stints in bandboxes. In one, he’s a much more effective pitcher than the other, because his skills even removed entirely from those playing around him differ greatly: in the larger stadium result in outs, while in the other, they result in home runs.
Or to your Franklin v Felix example: I see the point here, but I disagree that this line of reasoning leads to Franklin being selected. A pitcher does have things that they control the outcome of, and we can measure those. How many strikeouts did they record? How many walks? How many home runs?
But even then, those raw stats are tilted by the environment. More foul ground means fewer balls in play turn into hits, because more are caught for outs. The dimensions of the stadium, the air, the wind, the weather all affect how many line drives go for home runs. And on and on. At some point, performance is inseparable from the environment.
Overall, to circle back to something I mentioned back a bit, I wonder if we’re not arguing two different things: I’m in favor of which season was best, and I’m willing to say that Alex performing better in one environment over another made my choice, and I feel like you’re arguing that Alex’s road splits demonstrate that as a complete hitter, he was better in 2000 than he was in 1996.
It’s like… Best Mariner Player-Season in Performance versus Best Mariner Player-Season in Skill.
I have a feeling we’re going to be revisiting this argument again.
The problem with the SuperStadium example, though, is that you’re not adjusting for relative value of runs. In SuperKingdome, you could stick any generic RHB in that park and get a quality line, so the .350/.550/.950 mark is still only marginally better by the same amount that A-Rod is better than his peers. Same deal in SuperSafeco – if A-Rod could only hit .300/.380/.450 there, then an inferior RHB would do worse by the same amount as he would in SuperKingdome.
Runs, in and of themselves, aren’t valuable. They’re only valuable to the extent that they help you score more runs than the team you’re playing, and if you’re in a park that turns every RHB into Babe Ruth or Mario Mendoza, then comparing statlines without the context adjustment is pointless.
And, to be honest, I’d be more understanding of your argument if we were talking about a significant performance difference, where you could argue that the parks don’t matter enough because 1996 A-Rod was just that much better. But we’re not talking about a huge performance difference here. 1996 version had 5 more plate appearances, made 14 more outs, and amassed 43 more total bases. Even without doing any adjustment for the run environment of the times, the total production difference adds up to about 7 runs.
Once you deflate 1996’s raw line to account for the fact that a run was worth less than it was in 2000 (since offense was higher in the mid-90s), and you account for the fact that a run at Safeco is worth more than a run at the Kingdome (even ignoring the L/R thing), you’re forced to conclude that their performances were basically equal. The linear weights formulas that adjust for park and league give 2000 A-Rod a one run advantage, so if we’re strictly based off of who had the better performance, well, you still have to give it to 2000 A-Rod by the slimmest of margins.
I guess I just don’t even agree that 1996 wins the “In Season Performance” award. You just have to ignore too many things to draw the conclusion that he helped his team win more games in 1996 than he did in 2000, and when you factor in the degree of difficulty, it becomes hard for me to see why 2000 isn’t the clear choice.