Running the Gauntlet: The Diamondbacks continue to defy the odds

Back in March I took a look at the Gauntlet the DBacks faced in their early season schedule.

For quick review:

• 28 of 41 games, 68%, are vs. teams that won 90 or more games in 2018
• 25 of 41 games, 61%, are vs. teams that made the post season in 2018
• Accumulative 2018 record of first 41 games opponents : 3559-3092, .535 W%
• Projected record of opponents (FG Depth Chart) through first 41 G is .521 W%

There have been a few wrinkles. The Padres are indeed an improved team and have played the DBacks tough, winning four of the seven games. On the other hand, the DBacks caught the Red Sox at a good time, winning two of three from them. And they swept the injury depleted Yankees 2-0.

Still, any way you slice it, the DBacks have faced one of if not THE toughest schedules in MLB so far. We can confirm this a number of ways:

They’ve played 31 games against teams with a winning percentage over .500 and gone 16-15. They’ve played six games against team under .500 and gone 5-1. The next most games played vs. +.500 teams is by the Astros with 27. The Cubs, notably, have gone 15-8 vs. teams over .500 in their 23 games. Report Link Here

The second way we can see this is by looking at Baseball Reference SOS ranking, or Strength of Schedule. That ranking is based on the run differentials of opponents, not their won loss record. Here the DBacks are tied with the Astros for the highest SOS ranking in MLB.

So the Diamondbacks, with their 21-16 record are assured of being at least one game over .500 through their first 41 games, even in the very worst case scenario. (They start a four game series vs. the Braves tonight) In the March “Guantlet” article linked above, I wrote:

.......if they manage to navigate the first 6 weeks of the season at .500 or better, the schedule does level out a bit going forward and they might just be looking to add some pieces by the all star break or trade deadline.

While it would be premature to come to that conclusion today, I think it’s pretty clear that the team has done much better than at least I thought they would through this first difficult stretch. Especially considering the losses of Souza and Lamb. Some will say that was addition by subtraction, but nobody could know for sure that Christian Walker would be able to fill the breach, let alone contend for Rookie of the Year. And Adam Jones, while regressing mightily of late, both at the plate and in the field, was instrumental in helping the team get off to a hot start and stabilizing the clubhouse.

Much of all that I just wrote is a way to say......wow, this team has been pretty darn good and far exceeded my expectations. And they have been fun to watch most nights and days.

So is all of this “real” ? Is the over performance against expectations , (whether it be mine, Las Vegas book makers, various projection sights, and most fans, except for the most optimistic among us), something we should expect going forward ? Or is the team destined to regress into a heap of mediocrity ?

Nobody knows the answer to that of course. That’s why they play the games. But there are a few tools we can use to try to get an idea if this is likely to stick or not. I will say up front, it’s not conclusive. It’s a mixed bag. But please allow me to take you along for the ride of what I’m looking at.

1.) Baseball Prospectus Adjusted Standings:

I’m not sure if that link will work for those without a subscription. However here is a brief explanation of the 4 types of “record” they show

Actual Record/ W%: Self Explanatory

1st Order W%: This is based on runs scored and allowed, using Pythagenpat which is Baseball Prospectus’ version of Pythagorean WL , similar to what you’’ find at Baseball-Reference.

2nd Order W%: This is based on substituting projected runs scored and allowed for actual runs scored and allowed. The idea being that the underlying peripherals give a better idea sometimes of the real performance levels, taking out some of the random variance of sequencing. (i.e. getting that big hit in a close situation matters a lot to runs and wins, but is not as predictive or indicative of underlying talent and performance)

3rd Order W%: This take the 2nd order W% based on peripherals instead of runs, and then adjusts it for schedule.

I hope you followed all that, because it’s kind of important when we look at the DBacks:

So what we see here is that when you look at the runs scored and allowed, the adjusted W% drops from .568 to .544. When we look at the underlying peripherals, the W% drops further to .532. That’s still above .500 and pretty good. But the difference between .568 and .532 is probably the difference of a playoff spot. HOWEVER.....when the strength of schedule is taken into account, the adjusted W% pops right back up to .561. So IF the team were to play just as well as they’ve played through the first 37 games, then this reports states you should expect a .561 W% the rest of the season, which would land them at 91-71 record, and give them a really good chance at being in the wild card. So this is clearly one measure that is a good indicator that how they’ve played till now at least, has not been flukely.

2.) Baseball Savant Expected Stats:

Explained in the Link:

Expected Outcome stats help to remove defense and ballpark from the equation to express the skill shown at the moment of batted ball contact. By looking at the exit velocity and launch angle of each batted ball, a Hit Probability is assigned based on the outcomes of comparable historic balls in play. By accumulating the expected outcomes of each batted ball with actual strikeouts, walks and hit by pitches, Expected Batting Average (xBA), Expected Slugging (xSLG), and (most importantly) Expected Weighted On-Base Average (xwOBA) tell the story of a player’s season based on quality of and amount of contact, not outcomes.

Please do click through to the link to see full summary table, and sort up and down by clicking on column headings, and also flip back and forth between 2019 and previous seasons using the drop down file at the top. But here is a screed shot of the top 5 “over performing” teams at the plate so far this year.

What we can see here is that the DBacks , at .27, have by far the biggest differential between actual wOBA, and their expected numbers based on criteria above. It should also be noted that no NON COLORADO team has posted an entire season more than .16 better than expected since 2015. It’s important to note that park factors are stripped out here. This is based on contact. The Dbacks DO employ the Humidor of course, which in theory should have some impact on exit velocities in their home games. Last year for the full season the DBacks hit almost exactly to their expected number, (.310 vs. .311). I would have loved to see this by month however. ;)

I think it is pretty safe to say that the DBacks as a team will not continue to outperform this expected metric over a full season. Their level of over performance is unprecedented in the statcast era. Of course if the quality of their contact, and their BB/K ratios improve a great deal, then they could continue to post actual numbers as they have year to date. But barring such contact and BB/k Ratio improvements, it’s just very hard for me to envision the DBacks maintaining their current results at the plate.

Here are the individual players. David Peralta and Eduardo Escobar have been the biggest “over performers” to date. Better them than somebody else, as they have track record. And encouragingly, Christian Walker and Ketel Marte look like they might be mostly for real. Dyson, Kelly, and Ahmed’s wOBA perhaps not quite so sustainable.

On the pitching side of the ledger, as a team, the wOBA against and the expected xwOBA are almost exactly the same, .324 wOBA (21st in MLB) vs. .328. xwOBA (22nd in MLB) Not really much to see there as a team as far as over or under performance vs. expected.

Here are the individual pitchers:

Well there is something. According to this, Archie Bradley and Matt Andriese have been extremely unlucky, mostly on their balls in play. Andrew Chafin as well. The starters are all doing slightly better than expected, but nothing statistically significant there. Yoan Lopez is the one reliever this metric expects to regress, and it’s by a whopping .115 wOBA points. That’s to be expected perhaps, .211 for a full season is not sustainable. .326 is not bad, but it’s barely average. Fingers crossed his BB/K ratio improves as if it doesn’t his regression could hit hard. Greg Holland while showing .21 better than expected, his expected wOBA is still only .213.

I think the removal of Matt Koch from the roster, and the removal of John Ryan Murphy from the pitching tables combine to make this not look too bad. Certainly we can’t say the pitching staff has been lucky as a group, or over performing expectations. But clearly the relievers, except for Lopez and Holland , have been getting worse results than “deserved”, at least according to this metric

What about Fangraphs Run Estimators ? (Link is to team totals, you can click on starters and relievers to get breakouts. My takeaway is there is no significant variance from team ERA and the various run estimators. The middle of the pack, league average results from the pitching staff as a whole are completely “deserved” The difference is FG does not think the bullpen deserves much better than their ERA by FIP and xFIP, but there is a gleam of hope there in their Siera. Here is summary:

For those not familiar, FIP is Fielding Independent pitching, taking balls in play out of the equation, so only using Walks, Strikeouts and Home runs allowed , and then scaling to ERA. xFIP take the Homerun per fly ball rate, and regresses to the league average, to take out any luck with fly balls. Siera, viewed by some as a superior estimator, takes balls in play into account, and looks at deserved or expected results on those as well. Follow links for detailed explanations

FIP

xFIP

Siera

3.) Fangraphs BABIP and Hard/Medium/Soft hit rates:

The team has a .307 BABIP, which ranks 4th best in MLB

But if you click the column headers Hard, Medium and Soft, you see

Hard Hit: 10th Best

Medium Hit: 24th Best

Soft Hit 15th Best

So what that is telling you simply that the quality of contact does not seem to be supportive of the BABIP ranking. When I see these kinds of numbers, I expect the BABIP to drop a bit.

Summary:

The various projection systems, despite the team’s performance year to date against some very stiff competition, still do not project the DBacks to perform at a level one would expect from a playoff team:

Baseball Prospectus Rest of Season Projection has 62-63 , with a below average batting line of .246/.306/.396, .712 OPS, but about where they’ve been at league average ERA 4.27

Fangraphs Rest of Season Projection has them 61-64, just one game worse. (Please spare me the FG outrage. ;) They show a projected Runs Scored Per game of 4.48, which ranks just 22nd in MLB while the 4.61 Projected Runs against ranks 15th

Fivethirtyeight Projects the team to 64-61 the rest of the way and come in at 85 wins, and finish with a +35 Run Differential. (It’s currently +17) But there is no breakout for offense or defense.

The one fairly consistent theme here is the offense not only has been better than expected, which has been the story of the year, but the underlying peripherals and metrics that are available to us all seem to suggest it’s not sustainable with these inputs. In other words, if the quality of contact does not improve, and/or the BB/K rates don’t improve, it will be really hard to maintain current levels of offense.

The pitching on the other hand is getting the results they should be as a result, and there are some reasons for optimism the pitching will actually be better going forward due to certain personnel changes, as well as the depth that is available to them from the minor leagues.

Finally, it’s very difficult if not impossible to quantify the impacts of coaching, scouting, front office input, and especially dugout and clubhouse chemistry. This team FEELS like it’s all rowing in the same direction and feeding off each other. Of course when they are winning more than losing it’s easy to have that impression. They have showed a resilience and bounceback quality that has to be admired. Whether that would be enough to overcome any significant regression from the batters box remains to be seen.

No matter how it pans out, I am extremely entertained. This team is fun and interesting. And for that alone, I am grateful and happy, as I’m sure is everyone else.