Beware of negative regression

Today, I am going to focus on the concept of regression to the mean – a principle I have found valuable in fantasy sports in choosing which plays I want to draft, and which players I want to avoid drafting. It would also apply to players I want to add from the waiver wire (or avoid adding).

Regression towards the mean is a statistical principle that, assuming all else being equal, if your first sample you take is an outlier (either positively away from the mean or negatively away from the mean) then the next sample you take is likely to be closer to the mean than the first sample.

Regression to the mean can be either negative, or positive, although it is used more often in the negative sense. I apply the principle both ways, but probably more often in evaluating the potential for negative regression in players who are highly ranked and are being drafted early.

Consider Fernando Tatis, Jr., the Padres’ shortstop, who currently has an ADP of 21. This is very high for such a young player with so little track record, but his HR/SB numbers in 2019 are drawing many to take him in the second round. He seldom stays undrafted far into the third round.

While some analysts have expressed fears of negative regression with Tatis, the sample size is too small to determine this. While the advanced metrics would give me pause (I didn’t draft him on either of my teams), the chance of him being an impact player is just as high as negative regression.

Now, let’s look at another shortstop – Marcus Semien of the Athletics. This is a much better example because the sample size is larger than Tatis. Semien, who had a tremendous 2019, is starting his eighth year in the majors, and he has a career batting average of .256 and a career OPS of .752.

Compare those career numbers to what Semien’s put up last year, when he hit .285, with an OPS of .892. Now there’s a negative regression candidate if I’ve ever seen one. Semien hit a career-high 33 bombs last year. There was only one other season when he hit more than 20.

Semien’s otherworldly 2019 could very well prove to be an outlier in what has otherwise been a fairly pedestrian career for the 29-year-old veteran. If you’re looking for something in his past that may explain his out-of-nowhere success, don’t bother—you won’t find any.

The red light is going off on the dashboard, and the voice on the loudspeaker is shouting negative regression with Semien, but his ADP is currently 80. In my opinion, people drafting Semien as early as the sixth round this year will be disappointed when reality hits them like a ton of bricks.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s