Quick All-Star Break Study #3: Month-to-Month Correlation for ERA and Related Stats

Over the break, I’m going to run a few quick studies. If you want to request, add it to this Twitter thread.

I am running a few quick fantasy baseball studies over the next few days. I'll take suggestions. The two I have planned:

1. If a hitter's AVG is below his projection, how much rebound is expected?

2. How much does a month's ERA/FIP/xFIP matter when looking forward?

— Jeff Zimmerman (@jeffwzimmerman) July 16, 2024

The final study I’ll run is looking at how much stock should a person have in a month of pitching stats. I wanted to understand these correlations since I quote them a ton. The usual suspects top the list with an interesting find on WHIP projections.

Limits

• From 2021 to 2024
• Pitched at least 20 IP in the first month and 10 IP in the second.

Looking at all instances, here are the month-to-month r-squares. Included is the average monthly increase on ERA. I wanted to see how much of a change should be expected as the months warm up. Since I had the information set up, I went ahead and collected the information.

Note: The xStats and SIERA aren’t available in our database to enable easy querying.

First off, none of the stats come close to projecting the next month’s stats with K%-BB% being the closest in projecting itself at .20. With a month of information, no one knows what will happen next. On that note, ERA should never be used with just a month’s of info. WHIP is more predictive of the next month’s ERA than ERA. Treat ERA as a descriptive stat.

Second, if someone is looking into the future with just a small sample, the key stats to focus on for future production are xFIP or K%-BB%. They are the top two projecting ERA and in the top three for WHIP (with WHIP being the third).

One interesting note is that pitchers see a .21 increase in their ERA from month to month. I expected some increase with the months warming up, but the trend continues throughout the season (besides staying even from June to July). Here are the monthly changes in ERA.

It’s a steady drive upward.

To look at it another way, I compared a pitcher’s March and April numbers to his September and October numbers.

Seeing the 0.68 increase in ERA from the season’s start to the end is eye-opening. It’s probably not at the 1.00 ERA increase since some struggling pitchers will be demoted before the entire increase happens. My guess as to why the difference happens is that:

• Pitchers are ahead of hitters in their preseason prep.
• Pitchers wear down (i.e. lose velocity, get hurt) as the season goes on and see their performance decline.

The other major point is that K%-BB% and WHIP are the best at predicting future talent. With xFIP not as effective, I’d only use K%-BB% for any in-season analysis.

Overall, there are two main takeaways. K%-BB% should be in the discussion more with its short-term predictive values for stats that matter in the fantasy game (WHIP and ERA). Second, expect pitchers to keep getting worse as the season goes on. The actionable part here is to load up on two-start guys for two months and then back off.