Which is Better - Diamondback Starters or Relievers?

“It depends how you look at it.” Let’s look at it three ways:

• Directly compare stats for starters vs relievers.
• Compare stats to league average for starters and relievers. Determine which is better compared to league averages.
• Find a team with similar pitching characteristics. Determine what they do better as a possible path forward towards improving Diamondback pitching.

An important question is, “These comparisons will look at which stats?” To make this comparison simpler, let’s compare the 4 true outcomes (K%, BB%, HBP%, and HR%) plus BABIP. I tip my hat to Jack Sommers who wrote that there are 4 true outcomes, not just 3.

The statistics were downloaded on 29 April (games through 28 April). Because the statistics are updated every day, the comparison could be revisited later in the season.

How do starters and relievers compare?

The following chart is a direct comparison of starters and relievers.

Observations follow:

• Starters have a slightly better BABIP (.278 vs .289).
• Starters have better BB% (7.2% vs 10.3%) and HR% (3.5% vs 4.3%).
• Relievers have better K% (23.2% vs 20.0%) and HBP% (0.5% vs 1.5%).
• Relievers have a higher percentage of true outcomes (38.4% vs 32.2%), leading me to conclude starters pitch-to-contact more often than relievers.

This comparison is mixed - who is best depends on how you look at it.

How do starters and relievers compare to league averages?

Let’s look at a bar chart showing the comparisons.

Games thru 28 April. Data from Baseball Reference and FanGraphs.

Diamondback starters have stats that are better than league averages (BB% and BABIP). Their weaknesses appears to be HBP%, K% and HR%.

Starters have a 4 true outcome rate lower than the league average (32.2% vs 36.6%) widening my previous conclusion; Diamondbacks starters pitch-to-contact more often than starters in general.

Diamondback relievers have stats that are worse than league averages (BB%, BABIP, K%, and HR%). Their strength is low HBP%. Their weakness is high HR% (4.32% compared to league average of 2.9%).

The starters’ lower strikeout rate (20.0% vs 24.2%) is bad, but relievers’ higher HR rate (4.3% vs 2.9%) is a worse problem. Although neither starters nor relievers fared well compared to league averages, this comparison favors starters.

How do starters and relievers compare to a team with similar pitching characteristics?

After looking at many pitching characteristics, my intuition told me that, at this point in the season, the Astros pitchers are the more similar to the Diamondbacks pitchers than any other team.

What do the Astro pitchers do better than the Diamondbacks pitchers? Perhaps those differences are ways the Diamondback pitchers could improve, although how to execute those changes is unknown. Let’s look at a bar chart showing the comparisons.

D-backs thru 28 April, Astros thru 29 April. Data from Baseball Reference and FanGraphs.

Looking beyond the data shown, a big difference between the two teams is that the Astros starters had a lower homers-per-fly-ball (10.0% vs 12.8%). Could Diamondback starters reduce their HR/FB rate? Perhaps.

The Astros and Diamondbacks relievers had HR rates higher than the league average. I concluded that although the HR rate for Diamondback relievers is higher than the Diamondback starters, it is about what I might expect for reliever pitching.

The Astro relievers had a lower BB rate. Could Diamondback relievers reduce their walk rate? Yes. And it would have more impact than a similar improvement in starters because Diamondback relievers have a higher percentage of true outcomes.

Which improvement is easier - improving HR/FB or improving BB rate? My intuition is that reducing the BB rate is easier. Therefore, this comparison indicates that the relievers have an easier path to improvement.

An additional comparison: What does a comparison of ERA vs FIP vs xFIP tell us about starters and relievers?

The following chart shows the comparison:

Conclusions follow:

The starters were hurt by defense while the relievers were helped by defense.

The starters’ expected FIP (xFIP) was .200 lower than their ERA. They had little difference between FIP and xFIP. Improvement is expected in the future.

The relievers’ expected FIP (xFIP) was .170 lower than their ERA and .540 lower than their FIP. Relievers had an xFIP much lower than FIP. The quality of relief pitches is definitely better than it appears, and improvement is expected in the future.

This comparison shows expected improvement by both starters and relievers, but the relievers’ xFIP is lower than the starters.

Summary.

My opening statement remains true. When comparing starters and relievers, “It depends on how you look at it.”

Comparisons are mixed for the Diamondbacks. The starters fared better better compared to league averages. What tipped the scales towards the relievers is that they are more likely to improve as the season progresses. The relievers’ xFIP is lower than the starters. With the higher percentage of true outcomes, the relievers have an easier path to improvement - improve their walk rate. Therefore, my conclusion was that the relievers are better than the starters.