What We Can Learn From Three Week Samples

Quite often I’ll be doing research for a certain topic, then I’ll notice something that’s unrelated, yet more interesting than what I’m researching and writing about. To wit, this morning when I noticed that the season is exactly six weeks old today, it drew my attention away from what I was doing. The idea that we can divide the season into two separate three-week samples, and address some of the mob’s (fans and media) proclamations thus far in 2022 may tell us a few things – but probably not what you think.

Put a pin in that, we’ll come back to me climbing on my high horse in just a minute. For now, let’s look at the Yankees’ batters’ wRC+ from Opening Day through April 28th (exactly three weeks) and then from April 29th through May 20th (exactly three weeks). If you’re unfamiliar with wRC+, I’ll go into more detail below, but for now, all you need to know is that a) it’s a damn effective measuring stick of hitter performance, and b) it’s set that 100 is the league average in 2022, so 125 is 25 percent better than league average and 80 is 20 percent worse.

Enough of the preamble. Here are the Yankee’s batters’ wRC+ during the first three weeks of the season.

Rizzo 216
Judge 177
LeMahieu 157
Hicks 116
Kiner-Falefa 105
Donaldson 101
Stanton 93
Torres 83
Trevino 66
Gallo 61
Higashioka 2

A few things jump out at me. First, Anthony Rizzo played like Lou Gehrig, but that doesn’t mean he turned into a reincarnated Gehrig, regardless of the reaction from fans and Yankees media at the time. It was great to see DJ get off to a great start, and Hicks played pretty well in spite of how his performance was largely characterized. Also, Higgy reminded us that spring training performances mean absolutely nothing.

Then from April 29^th through today:

Stanton 231
Judge 230
Donaldson 186
Gallo 131
Torres 130
Kiner-Falefa 86
LeMahieu 84
Rizzo 80
Higashioka 68
Trevino 65
Hicks 50

Gallo has been better in May than every Yankee except the obviously scalding hot Stanton, Judge, and Donaldson. DJ and Rizzo have been “not” good for as long as they were good, but you would never know it based on their treatment from fans and media, and Hicks has been awful in May. (Only making his critics from April who were wrong about his performance then, louder now).

What can we learn from the above? From a baseball perspective, absolutely nothing – unless you weren’t aware that you shouldn’t base your evaluations on a small sample size and three weeks is a sneeze in baseball terms. Each of those samples represents only about 12 percent of the season, so all of them will continue to fluctuate to varying degrees. To be fair, look at the players’ total career baselines and more specific numbers that aren’t based on results to see if luck has been a factor in each player’s performance when evaluating them.

Also, although this is highly unscientific, I think we can agree that some players are treated better than others whether they’re playing well or not – which brings me back to my high horse: If you don’t like a player or prefer watching one player more than another, that’s great – we all have preferences and that’s what makes us fans and makes so much of these conversations interesting. But don’t conflate that with a player’s actual performance and contributions to wins and losses, as that’s a completely different discussion.

Did I miss anything? Let me know.

(wRC+ is “weighted runs created plus”. It takes OBP and SLG, as getting on base is the most important aspect of run-scoring, and advancing runners is the second most important. It gives a little more weight to OBP, then makes adjustments for park factors – because it’s easier to hit in Denver than in Oakland – and the particular season, because it was a lot easier to hit in 1999 than in 1967. That makes it a great stat to compare players of different eras, which is irrelevant to today’s chat.)