Baseball Musings: Clemens Report

January 28, 2008

Clemens Report

The Hendricks Brothers, agents for Roger Clemens, released a statistical report showing that Roger's longevity was due to adjustments, not steroids. Unfortunately, the AP does not link to the report. Has anyone seen it?

Update: The Boston Globe provides the link. You can find the report here.

Update: I just started reading, but there's some evidence the people writing the report know what they are talking about. Unfortunately, the report does not allow copying. At the bottom of page 3, however, the report discusses how ERA is a much better indicator of success than W-L record, noting that wins and losses are also dependent on run support, fielding support and bullpen support.

Update: Page 18 does a good job of refuting the idea that Clemens was washed up after the 1996 season. The writers note Clemens's won-lost record was due more to run support than poor pitching. They also note that Clemens's ERA margin that year was similar to Schilling's in 2004, a year Curt went 21-6 with great run support.

This is actually one of the things I find wrong with the reporting on Clemens. People are giving Dan Duquette more credit now for realizing Clemens was done in 1996. But according to McNamee, Clemens didn't start using steroids until 1998, after he won a Cy Young award in 1997. I don't think Duquette is off the hook for his statement, especially since the only thing wrong with Clemens's 1996 season was his W-L record.

Update: Starting on page 30, the report shows that while Clemens's quality of pitching didn't decline much with age (it jumped around a lot), his quantity of pitching did, both in terms of innings per game and pitches thrown per game.

Update: I finished a quick perusal of the report. It makes a number of good points, mentioned above. However, it does appear to cherry pick comparisons, showing similarities between Clemens, Ryan, Randy Johnson and Schilling to show that Clemens wasn't unusual. Of course, all four of those pitchers are unusual.

The most interesting graphs to me, however, were the ones showing the yearly fluctuations in Rogers ERA margin compared to Johnson and Schilling. Roger's bounces up and down throughout his career. Both Johnson and Schilling start off below their career averages, have a long steady period above their averages, then fall and don't recover. The fact that Clemens bounces around a lot means he suffers years of unexpected poor performance that Schilling and Johnson don't. Those might be the times Clemens is tempted to use steroids.

I'm interested to see what others think of the tables. For example, there is a table comparing the years Clemens, Johnson and Schilling finish in the top five in K per 9. It shows they all do it at about the same ages. If you extend that to top ten, does the table look different? There's a lot here for sabermetricians to chew on.

Posted by David Pinto at 07:39 AM | Cheating

Comments

No matter what the report, um, reports it only challenges those that mistakenly thought that steroids was, and is, a miracle drug that saved clemens career. Those people were wrong to begin with and are easy to target.

If one believes that steroids simply allowed clemens to keep up his workout schedule as he got older (recovery time) which is what happened if he did in fact take steroids then the report is useless.

Posted by: matt d at January 28, 2008 08:19 AM

Still, look at his ERA and peripherals. The strikeouts were always there, but he walked a lot of people in his last four years with the Sox. When you see a guy who turned 33 and his last four seasons include his three worst ERAs and one good season, I don't fault Duquette for thinking he was done. Then all of a sudden Clemens turns 34, strikes out more people than he ever has, brings his walks down near his career low, and gives up a remarkably low HR rate (lowest of his career). Much as I'd love to blame Duquette, you just can't see that coming.

The stats indicate Roger changed SOMETHING in 1997. Maybe he worked out harder, maybe he began juicing. In his defense, why would he take steroids in 1998 after posting his best season ever in 1997? And going against him... well, if we're to believe he was talking about using in 1998, I'd bet based on the numbers that he started in 1997.

Posted by: Mike at January 28, 2008 09:09 AM

I'm with Mike on this one. It really makes no sense that Clemens would use steroids in 98 after coming off the best season of his career in 97. Steroids or not, something happened in 97. That season was historic. A 221 ERA+ in 264 innings...which led the league, and 9 complete games, which led the league....clemens was by far and away the best pitcher in baseball that year, and better than he ever was before. He did something in 97 that enabled him to pitch like a god...im not saying it was steroids...but I wouldn't doubt if it was steroids either.

Posted by: fgfg at January 28, 2008 09:22 AM

I'm not sure I give any report credibility where the only comparisons are to other pitchers within the steroid era. How about comparisons to Seaver, Drysdale and other hard throwing RHs. Too many players and pitchers in the steroids era are posting late career numbers the like of which has never been seen before. A better understanding of the human body, year-round conditioning, better surgical techniques can explain small parts of the differences but not the sea change.

I'm also not sure why the report points out the fallacy of wins and losses as a useful measure then ignores the multitude of sabermetric data that would point out the similar fallacy of using ERA as an especially useful measure. XERA and BPV and other tools are available and familiar to most knowledgeable baseball fans.

Bill

Posted by: Bill McKinley at January 28, 2008 09:44 AM

Dave,

I'm surprised to see you fall into the Duquette fallacy re Clemens and "twilight of his career."

Duquette never said Clemens was in the twilight of his career in 96; instead, he expected that "twilight" to occur sometime during the life of the contract Clemens was offered with the Sox.

Posted by: Edward at January 28, 2008 12:02 PM

Dave,

I'm surprised to see you fall into the Duquette fallacy re Clemens and "twilight of his career."

Duquette didn't mean that Clemens was in the twilight of his career at that moment '96; instead, Duquette expected that "twilight" to occur sometime during the life of the contract Clemens was offered with the Sox.

Here is the actual quote:
"The Red Sox and our fans were fortunate to see Roger Clemens play in his prime and we had hoped to keep him in Boston during the twilight of his career."

It's a subtle difference but it changes the meaning quite a bit.

Posted by: Edward at January 28, 2008 12:06 PM

Edward,

Thanks for the quote. I don't remember seeing that. I just remember the fuss that Duquette thought Roger was done.

And, just to be clear, I think letting Roger go and trading for Pedro was a great move. Better to let a player go too soon than too late. But if the Red Sox had managed to keep Clemens and trade for Pedro, Boston might have broken the WS drought much earlier.

Posted by: David Pinto at January 28, 2008 12:27 PM