JAWS and the 2020 bWAR Update, Part 2

Josh Donaldson is one of the game’s elite two-way players, but like the late Ernie Lombardi, he received rude treatment when it came to Baseball-Reference’s latest update to its version of WAR. Last week I began a breakdown of B-Ref’s influx of new data, which resulted in alterations to five different areas of its version of WAR, some aspects of which affect players as far back as 1904 and others as recent as last season. The introduction of detailed play-by-play baserunning and caught stealing data from the 1930s and ’40s, for example, cost Lombardi — a heavy-hitting Hall of Fame catcher who played from 1934-47 — a whopping 7.3 WAR. Donaldson took the largest hit among contemporary players, losing 3.8 WAR via changes in the way Defensive Run Saved is calculated. For the 34-year-old third baseman, the loss adds a bit of insult to the injury of this delayed season, which won’t make it any easier for him to build what is admittedly a long-shot case for the Hall of Fame.

B-Ref’s version of WAR is different from that of FanGraphs, but as bWAR is the currency for JAWS, it’s of particular interest to me. While the Hall of Fame itself is as closed right now as any museum due to the COVID-19 pandemic, Hall arguments are never out of season, nor is taking stock of greatness, particularly when it provides a diversion from considering stockpiles of toilet paper and shortages of N95 masks. B-Ref’s adjustments are hardly unprecedented for the site, which adds new data annually. The earliest boundaries for game logs and play-by-play data have moved backwards by decades over the years, for example, and last year’s big-ticket addition was a major update to catchers’ defensive statistics for the 1890-1952 period.

Reordered for their chronological effect, this year’s update has incorporated the following:

• New Retrosheet Game Logs (1904-07)
• Caught Stealing Totals from Game Logs (1926-40)
• Baserunning and Double Plays from play-by-play data (1931-47)
• Defensive Runs Saved changes (2013-19)
• Park factor changes (2018)

As I noted last week, the career WAR totals of 11 Hall of Fame position players swung by at least 2.5 WAR, some positive and others negative. Where Lombardi was the biggest loser in that update, shortstop Arky Vaughan was the biggest gainer from among the enshrined; his 5.1-WAR gain was the second-largest swing overall, 0.1 less than that of three-time All-Star Lonny Frey (a teammate of Lombardi’s with the Reds from 1938-41). Because nobody needed 3,000 words from me in the first installment of a series as we await the green light on the 2020 season, I didn’t publish the table of the position-by-position changes or delve into the effects on other groups of players, such as Donaldson and his contemporaries. This time around, we’ll do just that.

First, here’s the table, in all its glory:

You’re on your own if you want to figure the deltas for career and seven-year peak WARs; I didn’t want to clutter the display, so the Change column only reflects the movement in JAWS. The standard for catchers fell by the greatest margin, as Lombardi, Gabby Hartnett, and Rick Ferrell lost about 13 WAR between them due to their below-average baserunning and double-play avoidance. The standards for pitchers of both flavors were the only others that dropped; two positions went unchanged, at least when rounding to the first decimal, while the standards for five other positions moved by anywhere from the equivalent of one to five runs.

What’s interesting when you note the magnitude of the changes (modest though they are) is that they mostly conform to the expectations that come with their spots along the defensive spectrum. Given the speed and athleticism required to play center field, we shouldn’t be too surprised that additional baserunning data nudged the standard at the position in a positive direction, where it had the opposite effect for the generally plodding catchers, athletic though they may otherwise be. The standards for middle infielders rose a bit, though it’s somewhat surprising that right fielders moved more; Mel Ott (+2.9 WAR), Chuck Klein (+2.4), Enos Slaughter (+1.7), and Paul Waner (+1.1) were among the latter group that had notable gains. None of the positions moved by much, but if you memorized where your favorite candidates stood with respect to JAWS totals, rankings, or standards, you may need to refresh the positional leaderboards.

As for the changes that affect contemporary players, updated park factors led to adjustments to pitchers’ WARs. At FanGraphs, we use five-year regressed park factors for our pitching WAR, where B-Ref uses three-year ones, and so with the addition of 2019 data, their 2018 park factors have been recomputed. The biggest beneficiary of this was José Ureña, whose 2018 WAR increased by 0.7 because the Marlins’ park factor rose from 90 to 95; on the other side of the coin, the Yankees’ Luis Severino — who’s already laid up due to Tommy John surgery — lost 0.7 WAR from the same season. Among Hall-relevant pitchers, CC Sabathia lost 0.5 WAR from his 2018 total, while Zack Greinke (-0.7) and Justin Verlander (0.7) moved in opposite directions enough for the latter to recapture the active lead in JAWS among starting pitchers (71.6 career WAR/50.0 peak WAR/60.8 JAWS for Verlander, 71.0/48.2/59.6 for Greinke).

I’ll have more to say about the pitchers, both ancient and modern, in my next installment. For the moment, of greater interest is the change to DRS, which Sports Info Solutions announced in January, and which has already been incorporated into our player cards, adjacent to the UZR that we use in our version of WAR. We also have a much more detailed look at the changes from SIS’s Brian Reiff here, as part of the FanGraphs library (which will soon receive a JAWS entry).

The short version is that the stat used to rely on one “Range and Positioning” figure for each player (formerly known as the Plus/Minus System), and added a team-level total for shift runs, but it now breaks each player’s total into PARTs, a clever acronym for runs saved via Positioning (pre-pitch, which it has been charting since 2013 as part of its shift data), Air balls, Range, and Throwing. The Positioning component is now credited to the team, with SIS’s reasoning being that some credit or debit for a player’s defensive performance owes to the club’s guidance in positioning rather than player skill; over-crediting a shifted player for plays made in a zone that no other third baseman can reach skews the totals, as was the case for the Blue Jays’ Brett Lawrie circa 2012. Matt Chapman was the biggest single-season beneficiary of this change, gaining 1.6 WAR in 2019 via a DRS that jumped from +19 to +34, with 13 runs from that change debited to the A’s team defense. At the other end of the spectrum, Adrián Beltré lost 1.5 WAR in 2015, and 2.0 WAR overall, which wasn’t enough to bump him out of fourth place in the JAWS third base rankings, but if George Brett, who trails him (71.0 to 71.2), decides to come out of retirement when the season resumes…

Here are the contemporary position players whose career WARs moved by at least 2.0, either in a positive or negative direction:

B-Ref didn’t say anything about the Law of the Conservation of WAR Within a Family, but it worked out that way for the Seagers. As for Donaldson, here’s a closer look at how his annual DRS and WAR totals have changed:

As you can see, the slugging third baseman’s best seasons have taken the biggest hits, starting with his 2015 AL MVP-winning campaign; where he previously ranked second in the league in WAR there (behind Mike Trout’s 9.4), he’s down to third (with Trout up to 9.6 and Manny Machado climbing from 7.1 to 7.5). Thus his career/peak/JAWS line slipped from 44.8/43.8/44.3 to 41.0/40.5/40.8. Not only did he fall from 24th to 29th in JAWS at the position, but he went from 0.8 above the peak standard — the cornerstone of his burgeoning Hall of Fame case — to 2.6 below it. As for that case, Donaldson has made just three All-Star teams and has yet to win a Gold Glove, but nobody’s taking that MVP award away, and he’s got five top-10 finishes in WAR and four apiece in on-base and slugging percentages, and he’s been a key part of seven teams that reached the postseason. He’s not somebody who would draw significant support from the voters if he retired today, but given a few more years of strong play and he likely would.

The good news for Donaldson is that his seventh-best season, still valued at 1.5 WAR, is low-hanging fruit from a performance standpoint, a total that he should be able to outdo that in any reasonably healthy season, even an incomplete one. The bad news is that players who meet the peak standard but fall significantly short of the career standard tend not to do well in Hall voting; almost invariably, they’re the ones who fall victim to the Rule of 2,000: no player whose career took place in the post-1960 expansion era who has fewer than 2,000 hits has been elected either by the BBWAA or a small committee. Donaldson, who played just 89 major league games before 2013, his age-27 season, has just 1,048 hits, and would need just over five seasons of matching his career-high 184 hits to get to 2,000, a highly improbable occurrence given his age and penchant for drawing walks (100 last year).

Meanwhile, here’s a look at Correa:

Houston’s shortstop gained ground in every season of his young career, though he’s still going to have to stay on the field with greater frequency than the past two seasons, during which he totaled just 185 games, and put up more seasons of his 2016-17 caliber to boost his peak score. At just 25 years old, time is on his side, and if this season is significantly shortened, well, he’s got a much longer read ahead of him to make up the lost ground.

Older pitchers of Verlander and Greinke’s ilk have less time to make up ground. I’ll be back with a look at them — and older pitchers such as Christy Mathewson — in the final installment of this series. And while it won’t be the only topic I cover during the delay to the start of the 2020 regular season, I’ve gathered a handful of good Hall of Fame-related ideas from readers that can help pass the time as we await the return of our favorite pastime.