28 homers, 98 RBIs
https://twitter.com/chico1640101/status/1128669539334594562
Paul's tweet triggered something in my mind. We don't often build posts around home runs and RBIs because there are, I think, so many more interesting and revealing statistics. But the way Paul put that -- how much are 28 homers and 98 RBIs worth today? -- struck a chord with me. It got me thinking about what the sabermetric revolution is all about.
There's something powerful about "28 homers and 98 RBIs." You see just those two numbers and, if you're like me, the image of a player comes to mind. But what player? That' the magic of the numbers. It's like the Old woman/Young woman illusion:
What do you see -- the old woman? The young woman? Both?
When you think of 28 homers, 98 RBIs, do you think of Mike Trout's 2016 season when he hit 29 homers and drove in 100? Or do you think of Joe Pepitone's 1964 season when he hit 28 homers and drove in 100?
Do you think of Willie Mays in 1958 hit 29 home runs and driving in 96?
Or do you think of Mark Reynolds' 1997 season with 28 homers, 97 RBIs.
These seasons are as different as can be -- Trout's 10.3 WAR vs. Pepitone's -0.7 WAR -- and though I'd never thought of it this way, it seems clear that the difference is at the heart of modern baseball analysis. As Bill James will be quick to say, we will never stop having massive blind spots when it comes to understanding baseball. The blind spots change, but they are always there and always will be there because baseball's complexity defies our efforts to simplify it.
That said: It is true that baseball offense was for more than a half-century reduced to three numbers: Batting average, home runs, RBIs. These were not just the main stats, they were the only stats. A quick look at Topps baseball card backs tells you a story:
1952: Stats on the back included: Games, At-Bats, Runs, Hits, Home Runs, RBIs, batting average. There were also fielding records including putouts, assists, errors and fielding average.
1954: Topps added Doubles and Triples.
1957: Topps removed defensive numbers.
1967: Topps removed Games and Runs.
1969: Topps put back in Games and Runs
1971: Topps only put one year of stats on the back but did add Total Bases and Stolen Bases.
1972: Topps went back to full career stats but took away Total Bases and Stolen Bases.
1973: Topps removed Games and Runs again.
1976: Topps put back in Games and Runs.
1981: And finally, in a banner year, Topps added Stolen Bases, Slugging Percentage, Walks and Strikeouts. It was a new age.
You can see the point -- for a quarter century, all other stats except batting average, home runs and RBIs were fungible. They were fine, but not necessary. As kids, we didn't question it -- or at least no one I knew questioned it. Topps set the conversation. If a number wasn't on the back of your baseball card, it wasn't worth thinking about.
So, while, yes, I must have KNOWN that there was a difference between Cal Ripken's 1983 season (.318, 27 homers, 102 RBIs) and Willie Upshaw's 1983 season (.306, 27 homers, 104 RBIs), I'm not sure that I could have explained it well enough to get a decent grade on the report.
And realistically, I'm not sure many other people could have told you the difference either. You look at it now, Ripken's season (league-leading 8.3/.8/5 WAR) dwarfs Upshaw's still solid season (4.3/4.5 WAR). We know why now:
Ripken played shortstop and played it extremely well. Upshaw was a good defender too but he played first base, which is considerably less valuable.
Ripken played in a tough hitter's park. Upshaw played in a dream of a hitters park.
Ripken led the league in runs (121), in large part because he also led the league in doubles (47). Upshaw hit 20 fewer doubles and scored 22 fewer runs.
Even now, these things can seem somewhat inconsequential to many baseball fans. But most people understand that defensive positioning, context, etc. add up to a lot.
That wasn't common knowledge in 1983. Yes, it's true, Ripken won the MVP award that year and Upshaw finished 11th. But, I would venture to say, this had almost nothing to do with the way we view the game now. It had only to do with one simple fact: Ripken's Orioles won the division (and, after the votes were cast, the World Series) while Upshaw's Blue Jays finished fourth. If it had been the other way around, there's no chance Ripken would have won the MVP, and a pretty good chance that Upshaw would have.
Why do I believe this? Simply: Look at the next year. In 1984, Ripken had an even better season than he did in 1983. It didn't look that way at all back then because Ripken hit .304 instead of .318 and drove in 86 RBIs instead of 102. But Ripken' WAR in 1984 (10/.9.8) wasn't just the best in baseball that year, it was second back of the decade. It was phenomenal stuff.
AND ... he finished 27th in the MVP voting behind, among others, Willie Upshaw (whose 1984 season was nowhere close to as good as his 1983). Ripken also finished behind Kent Hrbek (27 homers, 107 RBIs), Kirk Gibson (27 homers, 91 RBis), Alvin Davis (27 homers, 116 RBIs), Jim Rice (28 homers, 122 RBIs), Steve Balboni (28 homers, 77 RBIs) and George Bell (26 homers, 87 RBIs).
That's just what happened in those days when your team dropped from first place to fifth place in the division.
It's funny that Paul's tweet would make me think of all this ... but I have a somewhat new thought about why people connected (and some still connect) the MVP award to the team's performance: I can't help but think that this is (an admittedly awkward) effort to determine Wins Above Replacement. Go back to 1984. The winner of the MVP that year was reliever Willie Hernandez. It's lunacy from a pure production standpoint to think that Willie Hernandez -- a reliever who had an excellent season -- was the MOST VALUABLE PLAYER in baseball. Come on.
But look at it another way: The Tigers ran away. They were 12 wins better in 1984 than they were in 1983. Hernandez was really the only significant change the team made between the two years. You can easily see how someone would decide that Hernandez was worth 12 wins, which is even more than WAR gives Cal Ripken. Hernandez ran away with the MVP award even though he wasn't the best player in the game, the best pitcher in the game or the best player on the Tigers.
You can see how -- long before computers gave us more detailed baseball stats than we could have dreamed of -- this would not only make sense but would be common wisdom.
Anyway, this is where Paul's tweet took my mind.
Back to Bryce Harper. He struck out again on Thursday, giving him 20 Ks in his last 12 games. He's hitting .221, slugging .429, and if he has three more 40-or-so-game stretches like this one, he will finish right around Paul's 28-homer, 98 RBI season.
So, the original question: What is that worth? Well, we now know that Harper's value as a baseball player will come down to various things that have nothing at all to do with his homers and RBIs such as:
-- How well he plays defense. Last year, Harper was a fiasco defensively. But people who have watched him play this year say he's playing much better out there now, and the statistics mostly back this up. If Harper is a plus defender all year, that will make him significantly more valuable.*
*Though as Fangraphs' Brad Johnson points out -- with Harper, defensive effort also equals injury risk.
-- His ability to avoid making outs. Harper's on-base percentages tend to be very high because he walks a lot. Even this year, he leads the league in walks. But there's a big difference between his .365 OBP so far this year and the .410 OBP he had between 2015 and 2018. Harper's strikeout percentage has been on a frightening path the last two years.
2015: 20%
2016: 18.6%
2017: 20.1%
2018: 24.3%
2019: 30.8%
As various people have pointed out, Harper is streaky. He can go on a one- or two-month stretch where he takes his number closer to the moon. But if he continues to strike out at this pace, it will be tougher for him to go on one of those crazy hot streaks.
-- His performance in high-leverage situations. This isn't necessarily fair -- there's scant evidence to suggest that hitters can consistently perform better in the most important moments -- but, so far Harper is crushing the ball in high leverage situations this year (.344/.417/.750) as compared to medium (.218/.306/.327) and low leverage (.159/.381/.365). There is no reason to believe this is sustainable (for his career, Harper has not hit any better in high leverage than the rest) but he will do himself a lot of good with the Philadelphia fans if he can keep delivering in the clutch.
-- Taking advantage of his ballpark. Philadelphia is one of the better hitting parks in baseball the last three or four years. So far, Harper is hitting .176 at home. That has to get better.
And then, beyond that, yes, actually hitting more than 28 home runs and driving in more than 98 RBIs would serve him well. Harper has become much more of an all-or-nothing hitter, which I don't think is a great trend. But if that's what he's becoming, he has to embrace it. That means he needs a bit more ALL to go with that increasing amount of NOTHING.