Tag Archives: sabermetrics

Reining in the Moneyball Revolution’s Chief Excess, Twenty Years Later

Posted on by
Reply

What was the point of Moneyball? Nerds said on-base percentage is better than batting average and were technically correct on that narrow point? The real answer, of course, is that cheap owners‘ teams still can win by excelling at identifying and exploiting budget-friendly market inefficiencies. But when Brad Pitt says “on-base percentage is all we’re looking at now,” people tend to focus on that part and forget the rest.

Reaching base, however a player does it, is good, and counting only some ways players reach base necessarily misses relevant data points. By including walks and hits-by-pitch (“hit-by-pitches”?), on-base percentage (“OBP”) does paint a more complete picture of a baseball player’s offensive production than does batting average (“BA”), which only counts hits. The responsive inclination to look first to OBP rather than the traditional go-to, BA, thus is understandable.

Those comfortable with taking this new step, especially the early OBP adopters, often did so zealously and callously, even as they cloaked themselves in the mantle of measured reason. And when they did so, they very often took a second step: banishment of BA. Elevation of OBP was not enough; BA, the very embodiment of the old and impure way of thinking, must be cast out.

For the SABR revolutionaries, like not a few revolutionaries before them and to mix corporeal metaphors, that second step proved to be something of an overreach. As it turns out, the ancients were in fact onto something with BA, and there was something in that something that deserved to be conserved and carried forward through the revolutionary wave. BA, Eli Ben-Porat writes, not only deserves its place in baseball’s basic offensive statistic trinity– the Triple Slash Line of BA/OBP/SLG– but is the only component that actually belongs there.

As Ben-Porat explained over the weekend: “Dismissing batting average, in this author’s view, is just plain wrong. It is statistically significant in terms of predicting team runs, and on a per point basis, the most impactful component of” the building blocks of the triple slash line. After all, BA is a big part of both OBP and slugging percentage (“SLG”). And because of the way OBP weighs walks relative to hits, it can obscure the value of the offensive production it presents; in other words, not all OBPs are created equal. To Billy Beane’s point, it is important to account for a batter’s walks, but a hit– even a single– is better than a walk. Two players thus could post identical OBPs but have gotten there in much different fashion. Dumping BA would mask the real significance of a light-hitting, ball-taking batter’s empty OBP that matched the same mark of a more balanced player who hit more than he walked. Ben-Porat shows both that BA still matters and that presenting OBP without BA really makes the former less useful.

Whether Ben-Porat’s proposed adoption of an even more elemental triple slash line that omits the BA components of OBP and SLG and leaves the remainders (i.e., BA/BB%/ISO) catches on is another question. For now, rest with the satisfaction that you aren’t wrong to not get irritated when you see a player’s BA displayed during an upcoming MLB telecast.

__________________________________________________

Related
Why I’m not going to see Moneyball
Baseball Notes: Offensive Discrimination
Window Shopping: Ian Kinsler’s Walking, Not Running
Trout vs. Cabrera, and Aging with DRC+

Advertisement

RKB: Shifting the D to See Whether Analytics Drives the Motor City’s Baseball Team

Posted on by
Reply

The Detroit Tigers have the reputation of being a team late to baseball’s new analytical revolution, but they quietly have been making front-office hires (no, Brad Ausmus did not count) purportedly to try to catch up in that area, and there’s evidence that it’s happening. For example, two weeks ago, something occurredfor what I believe to be the first time in Tigers history, when manager Ron Gardenhire cited input from the analyitics department– excuse me, “analytic department”– as the reason for a decision he’d made:

If you’re excited — or angry — about seeing Jeimer Candelario in the lead-off spot Wednesday night, then feel to credit — or blame — the Detroit Tigers analytics department.

Tigers manager Ron Gardenhire said the recent spate of roster changes prompted a consultation with the club’s analytics and research department in an effort to find an ideal batting order.

“We did some research and the analytic department put all the data in there to try to see what gives up our best opportunities,” Gardenhire said. “(Candelario’s) name came up first as lead-off.”

Just the one analytic so far, but it’s a start. Now that we know the Tigers have sabermetric analysts and those analysts convey strategic input to the coaching staff, it’s fair to inquire into the quality of that input. As it turned out with respect to the above example, Candelario only hit leadoff for two games, and while he performed well (four hits, including a double and home run, and two strikeouts in eight plate appearances), it did not seem to be a part of Gardenhire’s long-term plan. Very likely coincidentally, the team lost both of those games, and Gardenhire moved Candelario back to fifth, where he’s hit for most of the season, for the next game, a win. As Lindbergh and Miller’s The Only Rule Is It Has To Work reminds, it’s one thing to develop sabermetrically informed strategies and another to implement them with coaches and players. (And, as beat writer Evan Woodbery pointed out in the article quoting Gardenhire, Detroit didn’t have many good options for the leadoff position anyway.)

More recently, Tigers observers and fans have cited with excitement a data point on defensive shifts an FSD producer pointed out over the weekend as more good evidence in this area, even suggesting that the team was becoming a leader (first place!) in the realm of new analytics-based strategy:

The irony of the timing of this was that it came as lead Baseball Prospectus writer Russell Carleton was in the process of dismantling the notion of the shift as a useful defensive strategy.  Continue reading →

Miguel Cabrera in the bWAR era

Posted on by
1

miguel cabrera 2003

I have been monitoring the effects of Baseball Prospectus’ recent modifications to its wins-above-replacement metric, WARP, on Miguel Cabrera’s career valuation numbers, and, on the whole, the results for Cabrera have been positive.

On Monday, former Baseball Prospectus editor in chief Ben Lindbergh discussed the ways in which WAR metrics always are in some state of flux as they incorporate newly available information and adapt to significant changes in game strategy and play:

In a sense, it’s unsettling that WAR is always in motion. Batting average may not be an accurate indicator of overall (or even offensive) value, but barring an overturned ruling by an official scorer or an unearthed error in archaic records, it always stays the same. Ted Williams will always have hit .406 in 1941, but his FanGraphs WAR for that season was 11.9 in 2011, and today it’s 11.0. That’s one reason why WAR values may never achieve the emotional resonance of evocative stats such as .406, 56, or 755, or even milestones like 3,000 hits or 500 homers.

WAR reminds us that objective truth tends to be slippery. And the metric is likely to get more unstable before it someday settles down. None of the big three versions of WAR(P) currently incorporates Statcast data. Thus far, MLBAM has drawn on that data to quantify aspects of player production without generating one unified number, but Tango describes it as “inevitable” that “eventually they will get rolled into one Statcast WAR metric.” He acknowledges that WAR’s amorphousness may make some fans more hesitant to trust it. Even so, he says, “Our focus should be on representing the truth as best we can estimate it. And it’s the truth that will attract the people.”

Baseball-Reference founder Sean Forman has responded to criticism of WAR’s mutability—not to mention its multiple implementations—by comparing it to Gross Domestic Product (GDP), another complex statistic that also changes retroactively and comes in more than one form. WAR works the way all science does: Discoveries are scrutinized, assumptions are examined, errors are rooted out, and breakthrough by breakthrough, we learn.

The focus of Lindbergh’s article was on the ways in which teams are straying from the traditional sequencing of starting and relief pitchers– frequently referred to as “the opener” strategy– are affecting WAR calculations, particularly Baseball-References bWAR.

An obstacle I encountered in analyzing changes in Cabrera’s WARP is that BP doesn’t keep a public record of statistical changes. By contrast, as Lindbergh helpfully noted, B-R does keep a public bWAR index, which effectively permits the tracking of changes to individual players’ seasonal bWAR totals on a daily basis since March 29, 2013.

In light of my prior documentation of the recent set of changes to Cabrera’s career seasonal WARP totals, I decided to take a quick and very rough look at how Cabrera’s seasonal bWAR totals had changed over the last six years. What I found was that, at least through 2012 (covering the first ten years of his career, which was all that was included in the March 29, 2013 data set), the difference was negligible. Some years’ bWAR numbers had increased a bit, some had decreased a bit, and some didn’t change; in total, the aggregate difference was -0.13 bWAR over those ten seasons. Doing a similar thing for the next six seasons by using the bWAR value from the first available date on the calendar year immediately following the completed season yielded a similar mix of results, with an aggregate difference of +0.38 bWAR. Combined, the total change is an increase of 0.25 bWAR, basically a negligible amount. Coincidentally, “negligible” also describes the value over replacement blog post (VORBP) of what you’ve just read.

_________________________________________________________

Previously
Miguel Cabrera continues to shine in the DRC era
Miguel Cabrera further bolstered by sabermetric update
Trout vs. Cabrera, and Aging with DRC+ (via Baseball Prospectus)

Average Hit Band: Photograph of the DRC Era’s New Normal

Posted on by
1

This MLB offseason, while arguably a bit chilly by hot stove standards, did offer baseball fans a hot new hitting metric in Baseball Prospectus’ Deserved Runs Created Plus (DRC+). In the words of its creators, DRC+ is “designed to parse out more accurately . . . batters’ expected individual contributions — separate from all other player and environmental factors — to their teams’ offensive production.” (My summary of that introductory article, which was nominated for a SABR research award, can be found through here.)

Unlike traditional, rate-based hitting metrics such as batting average (BA) and on-base percentage (OBP), DRC+ is an index statistic, meaning that it’s arranged to indicate the degree to which a player is above or below average, where 100 represents average. As part of its DRC+ rollout, BP published an homage to rate statistics (link and summary available through here) that touts their simple approach to delivering contextual information.

This undoubtedly is a user advantage for metrics like DRC+, but, by placing the focus so squarely on the average reference point, the initial transition from the rate-stat world of BA/OBP/SLG to the index-stat world of DRC+ can be a little bit rough. To help smooth things, I thought it would be beneficial to illustrate the translation with a quick look at all of the hitters who had “average,” according to DRC+, seasons at the plate in 2018.

Last season, eleven batters finished with at least 275 plate appearances and DRC+ marks of 100. As their traditional slash lines illustrate, they got to that point in a variety of ways.

The ranges for these eleven on each of the traditional hitting rate statistics are:

  • BA: .224 – .280
  • OBP: .294 – .351
  • SLG: .359 – .484

Obviously, because of the multitude of factors DRC+ considers, including both player-performance factors and environmental factors, these rate bands only serve as rough guidelines for fans making the mental shift from the rate world of BA/OBP to DRC+ that want a little help finding their bearings. (Also keep in mind that these “average” slash-line bands will vary from year to year. For example, in 1998, there were four players with at least 275 PA who posted DRC+ marks of 100, Matt Williams, Devon White, Luis Alicea, and Robin Ventura: BA between .263 and .279; OBP between .327 and .372; and SLG between .425 and .456. For reference, Mark McGwire, .299/.470/.752, led MLB with a DRC+ of 211 that year.)

_________________________________________________________

Previously
Miguel Cabrera continues to shine in the DRC era
Miguel Cabrera further bolstered by sabermetric update
Trout vs. Cabrera, and Aging with DRC+ (via Baseball Prospectus)

Related
The Best Baseball Research of the Past Year (2018)

The Best Baseball Research of the Past Year

Posted on by
1

20190207_133211

Once again, the Society for American Baseball Research has chosen fifteen (non-ALDLAND) finalists for awards in the areas of contemporary and historical baseball analysis and commentary.

My latest post at Banished to the Pen highlights each finalist. The winners will be announced on Sunday.

The full post is available here.

Miguel Cabrera continues to shine in the DRC era

Posted on by
5

Last month, I wrote about the substantial change in the way Baseball Prospectus is measuring hitter value and the significance of that change to Miguel Cabrera’s statistical legacy. Yesterday morning, BP announced “updates” to its hitter-value metric, DRC+. The description of the updates is pretty technical, and I commend you to the linked article if you want to get into the nuts and bolts, to the extent BP exposes them to the public. The short story seems to be that the original version of DRC+ undervalued two types of players: 1) those who play many of their games in “extreme ballparks” (Coors Field is the only one I’ve seen mentioned in the early DRC+ critiques and the update article, but I assume others are included) and 2) “extreme”-output hitters who do one thing really well (the examples I’ve seen discussed usually include singles hitters like Tony Gwynn and Ichiro Suzuki).

For Cabrera, the update credited him with even more productive value, adding almost two wins to his career total. The following chart, which I’ve adapted from the one I created for the BttP article, compares Cabrera’s career and season-by-season win totals under three different WARP regimes: a) TAv-based WARP; b) the original DRC+-based WARP; and c) the updated DRC+-based WARP.

cabrera warp drc update

(Notes: TAv-based WARP isn’t available for 2018, which affects the WARP totals in the bottom row. Orange highlighting signals seasons in which TAv and original DRC+ disagree about whether Cabrera’s offense was above or below average. Updated DRC+ was consistent with original DRC+ in that respect.)

Looking first at the table’s seventh column, the DRC+ update added to Cabrera’s totals, not infrequently by double digits, in every season save two minor decreases in 2007 and 2014. Looking next to the table’s final column, though, there isn’t really a consistent correlation between either the direction or magnitude of the update’s DRC+ adjustments and WARP; indeed, in 2008 and 2012, the update resulted in increases in Cabrera’s DRC+ but decreases in WARP. As the totals in the bottom row indicate, however, overall, the DRC+ update boosted Cabrera’s career WARP total by 1.8 wins. Not bad.

Here I will add the same caveat I included in my previous article on this subject, which is that I don’t have a deep enough understanding of DRC+, a proprietary metric, to explain with any further detail why this happened. (I also will note that, because BP does not archive its statistical reports from prior metric regimes, the foregoing is reliant on data previously captured by Archive.org’s Wayback Machine and me.)

What outsiders like us can say is that the Deserved-Runs-Created era has been good to Cabrera, from validating his MVP wins over Mike Trout to restoring all of his season-by-season WARP numbers into the black to, following yesterday’s update, increasing his career WARP total. None of this is likely to stir any concern on the parts of Al Avila or Chris Ilitch that Cabrera suddenly is on track to challenge for MVP votes in 2023 such that his $30 million option for his age-forty-one season in 2024 will vest, but the growing– even if by very small amounts– recognition of Cabrera’s past achievements is nice to see.

_________________________________________________________

Previously
Miguel Cabrera further bolstered by sabermetric update
Trout vs. Cabrera, and Aging with DRC+ (via Baseball Prospectus)

Trout vs. Cabrera, and Aging with DRC+ (via Baseball Prospectus)

Posted on by
6

MLB: All Star GameIt was about as clear as these things get, and the writers got it wrong. In fact, they got it wrong twice. That was the consensus, in our sabermetric corner of the internet, when Miguel Cabrera stole consecutive MVP awards from Mike Trout in 2012 and 2013.

Cabrera was a lumbering first baseman, shoved across the diamond only because the Tigers decided to force-fit Prince Fielder onto their plodding roster. He was a great hitter, but he added no value beyond that hitting. Trout, at the tender ages of 20 and 21, lit up the field in ways Cabrera couldn’t. He robbed home runs in center field, stole bases both often and efficiently, was one of the most consistent hitters in baseball, and according to the best information we had at the time, he was also Cabrera’s equal (or very nearly so, or perhaps even his superior) at the plate.

Baseball-Reference and FanGraphs each had Trout about 3.0 WAR better than Cabrera in 2012, and about 1.5 WAR better than him in 2013. We had the gap slightly smaller in 2012, but slightly larger in 2013. When such a clear gap between the best player and the field exists, it’s rare that the award goes to the “wrong” one. In this case, though, more or less everyone with a stat-savvy bone in their body espoused the belief that it had happened.

We were, all of us, deceived. … Read More

(via Baseball Prospectus)

One-Paragraph Preview: 2018 NLDS – Dodgers-Braves

Posted on by
Reply

ozzie

The Los Angeles Dodgers and Atlanta Braves begin their best-of-five National League Divisional Series tonight at Chavez Ravine. There are dozens of articles on reputable baseball websites previewing this series, all of which will leave you with the impression that the Dodgers are the superior team and probably but not definitely will prevail in this round. I write separately for the sole purpose of highlighting a conflict in the literature. Baseball Prospectus’ projection system, PECOTA, gives Los Angeles a solid (57%/43%) advantage. Interestingly, a postseason-prediction formula sabermetric forefather Bill James originally developed in 1972 that Rob Mains recently updated likes the Braves in this round. My own view is that it’s difficult to find an exploitable weakness in any phase of the Dodgers’ game, but I can’t count the Braves out in a competitive developmental environment that, more and more, seems to run on Coughlin Time. First pitch is set for 8:37 pm on MLB Network.

Baseball Notes: Offensive Discrimination

Posted on by
3

baseball notes

Although they may continue to cite them because of their familiarity as reference points, baseball analysts largely have moved on from the historically conventional hallmarks of pitcher and batter performance– ERA and batting average (“BA”), respectively– in favor of more comprehensive metrics that provide a more accurate picture of player performance by addressing some of those traditional statistics’ blind spots.

Focusing here on hitters, some of BA’s most notable blind spots include walks; the fact that each park has different dimensions; and the significant variance in the values of different types of hits (e.g., a single versus a home run). As they have with WAR, the three main baseball-analytics websites each offer their own improved versions of BA: Baseball Prospectus’ True Average (“TAv”); Baseball-Reference’s adjusted on-base-plus-slugging (“OPS+”); and FanGraphs’ Weighed Runs Created Plus (“wRC+”). Visually, TAv looks like a batting average but is scaled every year such that an average hitter has a TAv of .260, while OPS+ and wRC+ are scaled to an average of 100.

If you’ve read baseball articles here or at those websites, then you’ve seen those metrics cited, sometimes seemingly interchangeably, in the course of an examination of hitting performance. As BP’s Rob Mains notes in the first part of a recent two-part series at that site, there’s good reason to treat these three metrics similarly: they all correlate very strongly with each other. (In other words, most batters who are, for example, average according to TAv (i.e., .260) also are average according to OPS+ and wRC+ (i.e., 100).)

There are differences between the three, however, and those differences arise because each regards the elements of batting performance slightly differently. As Rob explained:

How the three derive the numbers themselves, including their respective park factors, is pretty small ball. Bigger ball, though, it what goes into them.

  • OPS+ incorporates the same basic statistics as OPS: At-bats, hits, total bases, walks, hit by pitches, and sacrifice flies.
  • wRC+ weights singles, doubles, triples, home runs, walks, and HBPs, with the weighting changing from year to year. For example, a home run had a weight of 2.337 in 1968 but only 1.975 in 1996, reflecting the scarcity of runs in the former year. Additionally, wRC+ considers only unintentional walks.
  • TAv also weights outcomes, including strikeouts (slightly worse than other outs) and sacrifices (slightly better than other outs). TAv also includes batters reaching base on error and incorporates situational hitting[, which refers to hitting that occurs only when runners are on base: Sacrifice hits, sacrifice flies, and hitting into double and triple plays].

So while all three measures look at the same thing—hitting—they’re not doing it quite the same. For OPS+, a walk is as good as a hit, from an OBP perspective, and a home run is four times as good as a single, per SLG. FanGraphs’ wRC+ weights them, but it doesn’t weight outs, as TAv does. Only TAv considers situational hitting.

When applied to players who are especially good or bad in those areas where the three metrics diverge, the result is a lack of correlation between the three with respect to that player. (Cf. the divergent views of the three WAR metrics with respect to Robbie Ray.) Mains’ second article examines some of those players of whom TAv, OPS+, and wRC+ take different views (e.g., Barry Bonds, Kris Bryant, Ian Kinsler, and David Ortiz) before explaining a few general conclusions:

[TAv, OPS+, and wRC+ are] very similar. You can use any of them and feel confident that you’re usually capturing the key characteristics of a batter.

If you want to drill down, though, here are the differences I found:

  • The lack of weighting in OPS+ means that it gives slightly less weight to singles and slightly more weight to home runs and walks than TAv and wRC+.

  • TAv’s inclusion of situational hitting means that batters who are extremely good or bad at avoiding double plays are going to get rewarded or penalized. (Situational hitting also includes bunting, but nobody does that anymore anyway.)

  • The black box factor in these calculations is park factors. Each of the three sites calculates them their own way. They can account for some changes, though not in a predictable or transparent way like high walk totals or low GIDP rates can.

I expect I’ll continue to use these three metrics somewhat interchangeably in articles at this site, although my preexisting (mostly uneducated) preferences for TAv and wRC+ likely will continue. Articles like Rob’s serve as both an important reminder that, at the edges, these updated metrics aren’t exactly the same and an entry point into thinking more precisely about what we ourselves value in the process of evaluating hitter performance.

___________________________________________________________________

Previously
Baseball Notes: Current Issues Roundup
Baseball Notes: Baseball’s growth spurt, visualized
Baseball Notes: The WAR on Robbie Ray
Baseball Notes: Save Tonight
Baseball Notes: Current Issues Roundup
Baseball Notes: The In-Game Half Lives of Professional Pitchers
Baseball Notes: Rule Interpretation Unintentionally Shifts Power to Outfielders?
Baseball Notes: Lineup Protection
Baseball Notes: The Crux of the Statistical Biscuit
Baseball Notes: Looking Out for Number One
Baseball Notes: Preview

The Best Baseball Research of the Past Year

Posted on by
Reply

Once again, the Society for American Baseball Research has chosen fifteen (non-ALDLAND) finalists for awards in the areas of contemporary and historical baseball analysis and commentary.

My latest post at Banished to the Pen highlights each finalist. The winners will be announced on Sunday.

The full post is available here.