Sean Smith's interesting 2nd half predictions

Over at the Hardball Times, Sean Smith crunches a bunch of numbers about the Angels, tackles the Great Pythagorean Riddle, and comes up with a 42-37 kicker to the rest of the season. I think he's too pessimistic by a great deal (I think we'll make it to 97 wins; he sez 91), but it's an interesting analysis.

about 1 year ago mattwelch Comment 8 comments 1 recs |

Story-email Email Printer Print

Comments

Yep, I'm not buying this one...

...they’ll pitch much better than that (and probably bat worse).

The ‘5000 miles away’ finger-in-the-air estimate says 95 wins.

I see red people

by The Limey on Jul 4, 2008 11:59 AM PDT reply actions 0 recs

Anytime a writer uses the phrase, "regression to the mean," it makes my skin crawl.

How can anybody predict a player or team will regress to the mean until it actually happens ? It is the most over used and trendy catch phrase installed in so many baseball analysis to make the writer appear knowledgeable and speaking with authority.

It’s not just this article, I find the miss use of “regression to the mean” all over the blogosphere. It’s miss use is even more bothersome when fans of other teams say the Rays or Angels second half season will be a regression to the mean because they are doing so well now they can’t possibly continue to do well the rest of the season. “Pythagorean says this team can only win X number games so enjoy your regression to the mean the remainder of the season.”

by 44FAN on Jul 4, 2008 2:04 PM PDT reply actions 0 recs

Regression to the mean

is a statistical term that refers to the fact that a random set of events (those that can be described by a normal distribution aka a “bell curve”) will tend towards the underlying mean with enough data. Its use in baseball suggests the writer thinks that they know what a player or team’s true capabilities are (usually based on something like their lifetime numbers, which ignore growth or decline as a player, or the Pythagorean abortion, that is only useful after the fact).

I agree that anyone who (mis)uses this term is showing off their ignorance of the term’s true meaning, and should be shunned.

K-brothers rule! (Kendrick & Kotch)

by halofan91 on Jul 4, 2008 5:12 PM PDT up reply actions 0 recs

Regression to the mean...

...is all about denying an individuals ability to excel beyond what is expected or predicted, even when the results are staring you in the face.

It’s the anti-sport.

I see red people

by The Limey on Jul 4, 2008 6:01 PM PDT up reply actions 0 recs

What are these numbers based on, I wonder?

I mean damn are some of those predictions pessimistic. Howie hits .298 the rest of the way, Kotch hits .280, Chone hits .285, Vlad hits .306 the rest of the way… K-Rod with a 3.05 ERA the rest of the way-

I mean, I appreciate regression as much as the next person, but when you predict nearly everyone on the team to play well below their career averages for the remainder of the season, there’s something wrong with your crystal ball, Nostradamus.

~Till the Halo burns out...

by Zu Long on Jul 4, 2008 5:30 PM PDT reply actions 0 recs

Nobody ever talks about PROGRESSION to the mean

When you suck for a while and then suddenly are totally on FIYAH FIYAH FIYAH you progress back to where you should be.

It happens as often as the regression, but stat-fucks are inherently dreamkillers who want to piss on your parade. They save the hope in life for the hot algebra teachers.

by Rev Halofan on Jul 5, 2008 2:19 AM PDT reply actions 0 recs

Nobody?

As a proud, card-carying “stat-fuck” I have to disagree with the use of the word “nobody” when you write “Nobody talks about progression”. This is the basis for my feeling that the Angels shouldn’t go out and make a big trade for someone like Holliday, I feel this team will trend back to their mean and start hitting. Most have put a fork in GA (including me), but there looks to be some life left in his 36 year old body. GMJ on the other hand never had much life except for his career year in Texas and there’s not much of a progression to be had. The Angels will hit, they’ll score runs and win more than 95 games because they have the ability.

And now I’m going to use the most hated term on this site…Fantasy Baseball. I really enjoy it. I think it makes me a better fan of baseball because it makes me look “deeper” than surface stats that casual fans rely on, and I appreciate other teams’ players more than I previously did. Everybody looks for a progression in fantasy baseball, its what wins leagues.

I hope Vlad leads the Angels and my fantasy team to a championship (in that order) because he’ll regress (progress) to his mean.

I was uncool before uncool was cool.

by WiHaloFan on Jul 5, 2008 9:27 AM PDT up reply actions 0 recs

I'm going on a limb here and predicting the Angels win the division.

That being said, I don’t really care how many games they end up winning as long as they win more than the other division leaders.

The greater concern is whether the hitting will get hot when we go into October. Will Vladdy be beaten up from getting beaned or swinging in the home run derby et. al? Will GA be batting blind? Will Kotch mysteriously dissappear? Will side salad be laboring on a gimpy ankle?

None of these are outcomes are predictable except that after 162 games, it’s likely that players may banged up, fatigued, or mentally drained. How does this play out in a playoff series? No one knows. That’s why they take they damn field! Follow the games and cheer for your team.

Gamblers and fantasy owners can disregard this post…

by melvintoast on Jul 6, 2008 7:37 PM PDT reply actions 0 recs