wRC+, which I went over thoroughly in my previous article, has become the benchmark overall batting statistic. Because new data comes to light and new ideas are subsequently presented, baseball metrics are everchanging and constantly innovated upon. New formulas using an array of different underlying statistics are developed, then tested against other metrics using historical player/team statistical data, and sometimes eventually released to the public if proven to be reliable. This is essentially how every metric, including those we use on a regular basis, is created.
Having acknowledged that better metrics are released periodically, we know that one day a new one will come along to challenge, and eventually surpass, wRC+ as the best publicly available batting metric. Sadly, there haven’t been any worthy running mates to wRC+ since its release.
Until now, that is.
During the first week of December, Jonathan Judge and the statistics team at Baseball Prospectus rolled out a new batting metric called DRC+ (Deserved Runs Created Plus), which, just like other ‘plus’ metrics, is scaled to 100 for league average.
For a concise definition of DRC+, let’s turn to the creators of the metric themselves, Baseball Prospectus:
“Deserved Runs Created Plus (DRC+) is BP’s new all-encompassing metric for batters.
DRC+ differs from other (public) hitting metrics in that it focuses on each hitter’s expected contribution, rather than merely averaging the result of hitting PAs.”
Expected contribution – now, that’s intriguing! I wonder how that’s calculated, I immediately thought to myself.
Though the formula has not been released for public consumption, BP’s Bryan Grosnick offers a little bit of insight into how DRC+ works:
“DRC+ uses a mixed-model approach to deal with several contextual variables that affect the hitter’s performance, and assign an expected value to the player’s performance that neutralizes those factors. All of this calculation takes place “behind the scenes”—though the process and the different DRC+ components are publicly available—and leaves us with a number of Deserved Runs Created (DRC). The final step is to scale that DRC number to league average, where 100 is average … at least for non-pitchers.”
To phrase that differently, DRC+ endeavours to award the hitter for their deserved skill. A multitude of variables can affect the outcome of a hitting event, so DRC+ only credits the hitter for what’s deemed to be their responsibility. Home runs, walks, and strikeouts are generally more often attributable to a player’s skill than singles or triples are, and these outcomes are weighed accordingly. In a nutshell, the metric attempts to eliminate factors beyond the skill of the hitter that contribute to the outcome of the hitting event.
DRC+ correlates closely to wRC+ and OPS+, which is a promising indicator for the new stat. However, it needs to differentiate itself from, and improve upon, wRC+ or it can be cast aside with relative ease as we await the next great baseball metric. Luckily for us, DRC+ appears to not only distinguish itself from wRC+ rather significantly, but it may also improve upon it in a couple of key areas.
The first is that it adjusts for the quality of the opposing pitcher, which sets it apart from the other all-encompassing hitting metrics. To me, this only makes sense to account for. I mean, why shouldn’t a hitter be credited more for homering off Max Scherzer than for hitting one out of the park against Andrew Cashner?
Another important underlying component of DRC+ is that it incorporates ballpark data differently than other public hitting metrics – it uses single-year park factors while other metrics prefer multi-year park factors. While there are pros and cons to using single-year data over multi-year data, using one-year data is more beneficial to DRC+ due to some subtle changes made to the ball a few years ago. The ball now carries further on average than it did before, leading to a spike in home runs relative to seasons past. Because many parks now ‘play’ differently than they used to as a result, isolating park factors to a single year provides DRC+ with more reliable ballpark data than other metrics.
Something that I find very interesting about DRC+ is that it requires the hitter to exhibit an ability to repeat the skill before he’s awarded more credit for the results. If the skill has not yet proven to be repeatable – in the case of a small sample, for example – his contributions are reduced toward the mean until it’s proven that he’s deserving of more credit. With that in mind, small sample size caveats apply. I’m not certain at what point this metric stabilizes, but it’s probably at least a couple hundred plate appearances before you can get anything remotely worthwhile out of it.
“That’s great, but how does it actually perform vs. other metrics?” This is a fantastic question, astute imaginary reader. Allow me to borrow a table from Baseball Prospectus to illustrate the purported effectiveness of DRC+:
That’s right – DRC+ claims to be more reliable, predictive, and descriptive than the best hitting metrics ever released for public consumption.
To concisely summarize what these three descriptors measure, I'm going to use this explanation from Baseball Prospectus:
Reliability is the extent to which the player’s estimated past contributions are similarly estimated by the same metric during future plays.
Predictiveness is the extent to which the player’s estimated past contributions correspond to the outcomes of future plays involving that player.
Descriptiveness is the extent to which the player’s estimated past contributions to plays corresponds to the outcomes of those plays.
For a more thorough explanation, direct your attention to this article here.
You can find the above table at Baseball Prospectus here, and I implore you to read the entire article for a deeper understanding of DRC+. There are many other free articles pertaining to DRC+ that one can find over at Baseball Prospectus, and I’d recommend to any of you curious readers out there to waltz on over and give each one of the articles a thorough gander. If DRC+ is something that you’re interested in taking a deep dive on, you’ll be glad you took the time to read about it – trust me.
So now that we have a better understanding of what DRC+ is and how it differs from wRC+, let’s look at what DRC+ thought of the Blue Jays in 2018:
Yikes, that is decidedly not good overall – mind you, that shouldn’t come as a huge surprise, as we know the Jays were not good in 2018. Though our knowledge of the metric is still somewhat limited at this time, I’d like to highlight a few of the players from the above table and discuss them a little bit below.
Kendrys Morales gets a small boost compared to OPS+ and a bigger one compared to wRC+. Without knowing all the components of DRC+, we can assume that he benefits from an above-average BB% and a below-average K%, which is attributable to his skills as a hitter. It’s worth noting that his BABIP was 22 points lower last year than his career mark and 24 points lower than the league average.
Justin Smoak gets downgraded compared to both wRC+ and OPS+ and I have a hunch that it’s at least partly because of his high 26.3 K%, despite a well above-average BB% of 14.
Randal Grichuk is another player that suffers a significant drop from wRC+ to DRC+. Because walks and strikeouts are weighted more heavily than some other hitting outcomes, Grichuk’s well below-average 5.8 BB% and bad 26.4 K% don’t exactly portend sustained success or good times ahead. What Grichuk does do well offensively is that he hits the ball far and he hits the ball hard, so that likely retains some of his offensive value.
Veteran Russell Martin was shipped out of town to Los Angeles last week and Kevin Pillar really needs to be traded, like, yesterday. Please just do it, Atkins – before his defense drops off considerably and he’s left devoid of value.
There probably isn’t a whole lot to glean from the small samples of Danny Jansen, Billy McKinney, Dwight Smith Jr., Brandon Drury, Richard Urena, and Rowdy Tellez, but I would be remiss if I didn’t at least mention the massive discrepancy between Tellez’ DRC+ and wRC+. 63 points! It’s an even wider gap – 66 points – from DRC+ to OPS+. Let’s go over a couple of reasons why I believe that to be, aside from chalking this up to just small statistical sample size noise.
Holy shit, that is a ghastly BB/K rate! This is certainly a significant factor, as there’s no way success can be sustained with such a massive difference between walks and strikeouts.
You probably recall Tellez doubling in what seemed like almost every PA last September. While that’s not quite true, 9 of his 22 hits did go for doubles. This is simply way too many doubles – something that would not sustain itself over a larger data sample. DRC+ is right to be skeptical of this and will make Tellez prove that he deserves to be given credit for this skill – a skill which is undoubtedly non-existent, at least at this rate.
I’m not going to pick apart Tellez’ numbers any further, as it was just too small of a sample to glean anything worthwhile from. And, really, anything that the Jays can get out of Rowdy offensively is a bonus, as he’s never been considered a core piece of the team’s future. But no matter what he becomes, we’ll always have the fun memories of September 2018 to reminisce about.
It’s worth touching on that, like wRC+, DRC+ is a believer in Lourdes Gurriel Jr.’s hitting abilities. Despite an unfavourable BB/K rate, Steamer projects him to be roughly a league-average hitter in 2019, with a 97 wRC+. DRC+ projections do not yet exist, but it’s easy to draw a conclusion from the table above that DRC+ would likely view Lourdes similarly. If he can cut down the swing-and-miss, he’d be almost a lock as an above-average hitter moving forward because of how hard he stings the ball.
One final point of interest that's not depicted in the above table is that we also have at our disposal DRC+ ratings for the minor leagues. Vladimir Guerrero Jr., in 128 PA at AAA, put up a 154 DRC+ rating as a 19-year-old. In 266 PA at AA prior to that, he fared even better with a 170 rating – this was the top mark in all of AA. Barring something unforeseen, we know that Vlad will make his MLB debut in the early part of the season, likely once he's surpassed the service time cut-off that would give the Jays an extra year of team control before he potentially reaches free agency.
As is the case with all other hitting metrics, DRC+ is not an exact all-encompassing measure of offense and I’d advise against using it as such. But it does appear to be, for now, one of the best metrics that we have at our disposal. Whether DRC+ is or isn’t better than wRC+, we do know that it won’t be the last metric released – there will be others in the future that we will be afforded the opportunity to dissect and compare to wRC+ and DRC+.
We’re still in the very early days since its release, so the verdict is very much yet to be determined regarding its merits and purported superiority to wRC+. Having acknowledged that, the early indications are intriguing and we, like the rest of the analytic community, will continue to monitor and review DRC+ throughout the 2019 baseball season.
Thanks for reading Introducing DRC+ and Using It to Evaluate the 2018 Blue Jays by Michael Paul. If you have any questions or comments relating to this article, we encourage you to leave them below. For all general inquiries, we can be reached at the following:
Twitter https://twitter.com/radio_scouts
Comments