Poll 2019: Which Group of Pitchers Performs Better? — A Review

As I have at the end of the first half since 2013, I grouped two sets of pitchers together and aggregated their results through the half based on the degree of SIERA outperformance and underperformance. I then asked you which group of pitchers would perform better from an ERA perspective over the second half, and which range each group’s ERA would fall into. This year’s poll and voting results are here.

Let’s first review the poll results:

Which Group Posts a Lower 2nd Half ERA?
Group B – SIERA Underperformers 78.85%
Group A – SIERA Outperformers 16.22%
Neither, each group posts the same ERA, or within .05 in ERA 4.93%

Which Range Will Group A’s 2nd Half ERA Fall Into?
3.75-3.99 27.86%
4.00-4.24 26.39%
3.50-3.74 16.42%
4.25 or above 14.08%
3.25-3.49 8.21%
3.00-3.24 6.45%
2.75-2.99 0.59%
Below 2.75 0%

Which Range Will Group B’s 2nd Half ERA Fall Into?
3.50-3.74 25.84%
3.75-3.99 25.84%
4.00-4.24 18.54%
3.25-3.49 16.11%
3.00-3.24 8.21%
4.25 or above 4.86%
2.75-2.99 0.6%
Below 2.75 0%

The results are quite surprising to me, as I believe this is the first time in my polling history that you readers have voted for Group B to post a better ERA than Group A over the second half. It wasn’t even close either, as Group B earned the vast majority of votes. It certainly makes sense, as group’s aggregate underlying skills were superior, resulting in a 4.00 SIERA, versus a 4.74 mark for Group A. Still, you don’t figure the pitchers would change teams, so they would be pitching their home games in the same park and be supported by the same defense. Would the skills truly win out as SIERA suggested?

Only 14% of you voted for an ERA over 4.25 for Group A, despite a SIERA of 4.74 in the first half, while over half of you voted Group B would post an ERA significantly better than their first half ERA and just below their first half SIERA.

Now let’s find out how the two groups actually performed in the second half.

Surprise! Yet again, Group A outperformed their SIERA in the second half, and also outperformed Group B, which underperformed their SIERA. However, the gaps between ERA and SIERA for both groups narrowed significantly. Interestingly, this time the gaps were only driven by BABIP and LOB%, as HR/FB rate was nearly identical (in the first half, Group A outperformed 10.4% vs 17.3%, a huge difference).

Group B did allow a higher LD%, so they did perhaps deserve a higher BABIP allowed. Some of it also could be that Group A legitimately received better defensive support from better fielders. In addition, Group A may simply benefit from stronger relievers behind them who are more skilled at stranding their baserunners.

Last, there was certainly an injury component hurting Group B, thanks to Chris Sale and Chris Archer, while German Marquez is prone to SIERA underperformance due to his hitter friendly home park.

So what did we learn this time? That these groups do usually end up converging and the gaps between their ERAs and SIERA do narrow as expected. However, some things that don’t change like home park and defensive support could allow SIERA outperformance and underperformance to continue, though not necessarily by the same degree. That said, it does validate that the average outperformer makes for an excellent sell high, while the average underperformer makes for excellent an buy low after the first half.