The fantasy alphabet, part two: E-I

E is for…extrapolate

It’s something we all think about but rarely practice statistically. While some people cry small sample size, one can make an educated guess, in certain instances, as to what a player’s full-season stats can be if injuries or midseason call-ups (and the lack of at-bats because of that) are ignored.

Take, for example, Dee Gordon, who has been mentioned in this Alphabet series already under “C is for cheap speed.” Gordon was called up at two different points during the season—once because of a Rafael Furcal injury and another because of the Furcal trade—and he also hit the DL once. He logged only 233 plate appearances for the aforementioned reasons, but still stole 24 bases.

Known as a counting stat, we can (somewhat) fairly assume that Gordon would have kept some semblance of that pace if he had gotten 600 PA. You can dispute or not that he could have truly kept such a torrid pace, but if we extrapolate his totals to reach 600 plate appearances, Gordon would have stolen about 62 bases last year.

Extrapolating is used for counting stats only, and is especially helpful with speedsters. If there’s opportunity mixed with talent, extrapolation can sometimes tell you what that equation might produce. It can be misleading, though.

Jason Bourgeois stole 31 bases in 238 AB for the Astros this year, and extrapolation to 600 AB would yield a predicted steals total of 78 bases. But if you don’t know the situation—that Bourgeois hit .400 for a month before hitting the DL, gaining a full-time job when he got back, suffering offensively when he was given that job, and eventually being relegated to the duties of a fourth outfielder—you can be misled.

Bourgeois doesn’t have a true talent level like Gordon, who is a top prospect, the shortstop of the future in Los Angeles, and will get ample at-bats to try his hand at 80 steals.

Here’s a look at some notable extrapolation totals. Assume 600 AB, and do what you will with these stats. Also, I rounded up to the nearest whole number when applicable.

Name                   Ext. RBI       Ext. HR       Ext. SB
Hanley Ramirez               80            18            36
Carlos Gonzalez             115            32            25
Desmond Jennings             61            24            49
Jemile Weeks                 47             3            33
Eduardo Nunez                58            10            43
Brett Lawrie                100            36            28

F is for…FIP

Baseball Prospectus does a good job of summarizing FIP, so I’ll borrow their definition:

Fielding Independent Pitching converts a pitcher’s three true outcomes into an ERA-like number. The formula is (13*HR+3*BB-2*K)/IP, plus a constant (usually around 3.2) to put it on the same scale as ERA.

That said, it’s not the best predictive tool in the universe. It can better tell (in my opinion) what a pitcher’s true talent level over that span was, though—what factors did he control, and how well did he do controlling those factors?

xFIP is known to be a better predictive tool, as it stabilizes home run rate to roughly the league average, sifting out outliers in HR/FB rate by making it a stable 10.5 percent. We’ll use some combination of both to find some names who will not be called elite, and maybe rightfully so, but did find their name on a list of elite pitchers in 2011.

Brandon McCarthy, he of a 2.86 FIP and a 3.30 xFIP, was fifth in the league in the former. He isn’t a strikeout artist and pitches for the A’s, which means fewer opportunities to win (at least in theory) and less publicity, but he was quietly really, really effective with 4.7 WAR and a 3.32 plain-old ERA.

Matt Garza, he of a 2.95 FIP and a 3.19 xFIP, was McCarthy-esque in Chi-Town after the Cubs gave up a king’s ransom to get him. He exceeded his thought-to-be-an-outlier 8.38 K/9 put up in 2009 (he tallied 6-7 K/9 the four previous seasons, though only two were full seasons) by over half a strikeout per nine, and beat his career BB/9 rate (3.11), recording 2.86 BB/9. Though his HR/9 was seemingly fluky (0.64 HR/9 in 2011 vs. 0.97 career), he took a step forward this year, and should serve well as a No. 3 starter in fantasy next year.

Doug Fister, he of a 3.02 FIP, a 3.61 xFIP, and a sparkling 2.83 ERA, was superb in the Tigers uniform after being quietly productive in Seattle for a year and a half. After winning three of his first 21 starters as a Mariner despite his 3.33 ERA and 3.27 FIP, he was shipped to the Tigers, where he won eight of his 10 starts to the tune of a 1.79/2.49/2.75 triple slash.

His entire 5.6 WAR season was generally in line with his career BABIP, LOB%, and HR/FB rate, though the HR may come more often against Fister next year. He, like McCarthy, isn’t a strikeout artist, which may explain why his name isn’t mentioned among fantasy’s elite starters, but he can be excellent for ratios (1.06 WHIP this year, 1.18 career WHIP).

Also interesting to see is ERA-FIP, which is a stat that simply subtracts one’s ERA from his FIP to see how lucky he’s been. The list is often telling in terms of who is a regression candidate. For example, Trevor Cahill’s 2010 ERA-FIP was third worst in the league at -1.21, and his 4.19 FIP was matched almost exactly in 2011 with a 4.16 ERA, which is purely coincidence, but does exemplify the power of FIP to tell us who overperformed.

A Hardball Times Update
Goodbye for now.

The three lowest (bad) and highest (good) ERA-FIPs in 2011:

Lowest

Jeremy Hellickson  (-1.49)
Ricky Romero       (-1.28)
Joe Saunders       (-1.08)

Highest

Derek Lowe          (1.35)
Ricky Nolasco       (1.14)
Brandon Morrow      (1.08)

This isn’t to say that Lowe, Nolasco, and Morrow will be studly in 2012, but it’s more than likely they will return to respectability. Their ERAs, respectively, were 5.05, 4.67 and 4.72, which doesn’t represent their performance rightly; or at least, that’s what FIP is here to argue.

G is for…ghost teams

For commissioners and competitive Rotoheads alike, nothing is worse than a ghost team. A familiar situation may be this: You find yourself bummed after another anemic offensive day and, slipping in the standings, you panic and look for a change. You scour the waiver wire and free agent pool and you find nothing suitable; you want shakeup, and you want it now. You have your mind set on a certain player, you decide—let’s say Pablo Sandoval—and you find he belongs to the ghost team.

Whether that becomes obvious right away or after the ghost team’s owner doesn’t respond in the next week, it’s a frustrating situation, no doubt.

In leagues in a professional setting, corresponding over e-mail to spitball trade ideas before proposing them through the system is a generally good idea. You have the assurance that they will check their professional system/company e-mail (with a few exceptions), and you can guarantee yourself an answer.

But say you don’t have such a e-mail setup, and your owner has truly “gone rogue.” Well, frankly, there is no solution I can see. I’ve suffered through it several times and, unfortunately, with the same guy in one of those cases, and my advice is to seriously consider who you put in your league. If you already have an indifferent owner, replace him or her before the season starts.

And what happens in redraft leagues, where the owner can freely stop paying attention because “they have no chance?” In such a case, they’re in the right, as frustrating as it can be.

They would be doing the whole league a disfavor by picking which owner (their best friend, probably) they want to win, and trading their stars to him or her with no tangible reward involved. Or they could just be committing collusion, and selling their players to the highest bidder so they cash out even with a bad fantasy season. So by simply playing in a re-draft league, you are allowing ample opportunities for an owner to screw it up for everyone else. Simple solution? Play in a keeper league.

H is for…HR/FB

Home runs-per-fly ball (HR/FB) is about as self-explanatory as stats get, but here’s an explanation anyway: How many of a given player’s fly balls end up as home runs.

The 2010 MLB average was 10.6 percent, and the highest was 25 percent. Since 2005, only five HR/FB percentages have been above 30 percent, and Ryan Howard has put up three of the five highest HR/FB percentages. His peak season, 2006, had him at 39.5 percent, which is almost eight percentage points higher than the next highest one (31.8 percent, put up by Howard in 2008).

Over the last six years, many players who have put up HR/FB percentages over 25 percent in a single season have experienced some sort of regression, but also interesting are the outliers. Here’s a chart that better exemplifies the regression/outliers:

image

You’ll notice that it’s terribly volatile, random, and full of outliers. Ryan Howard has a career 28.7 HR/FB. Alex Rodriguez has a career HR/FB of 22.5.

This all brings me to the point about Mike Stanton; his power is elite, possibly historically elite. He has a 24.0 percent HR/FB ratio is his young career. Stanton led the league by 2.1 percentage points with a 24.8 percent HR/FB ratio in 2011, and he has 56 home runs before he’s even turned 22. Among all hitters with 500 plate appearances in history, he has the fifth-highest HR/FB percentage; and one of the hitters ahead of him on the list is pitcher Carlos Zambrano Howard, Jim Thome, and Barry Bonds represent the other three.

Howard had the historically good 39.5 percent in 2006, and then dropped considerably to a point where he was still around 20 percent above average. In 2007 he put up that mark, 31.5 percent, and followed it up with 31.8percent. His ridiculous 2006 aside, he improved his numbers from 2007 to 2008 in terms of HR/FB, despite the notion that he was riding on a huge wave of good fortune.

Stanton is one of these outliers. He was assigned an 80 on the 20-80 scouting scale by Keith Law, unsurprisingly, and if you’ve ever seen one of his bombs, you know he has a special level of raw power. His HR/FB may be able to stay at the 25 percent level or even increase, as players with his caliber of power don’t necessarily regress (though the possibility remains), and his outrageous ratio, coupled with his untapped upside, means we could be looking at 50 HR as soon as next year.

I is for…injuries

I is for injuries, or a lack thereof. The following players have played in almost every game over the last three years, or are within 10 games of the 486 possible in the years between 2009-2011. There’s something to be said, especially in a H2H league, about players who are consistently suiting up and giving you the maximum number of games and not disturbing the balance of your team by hitting the DL or giving themselves too many off days.

Consistently on-the-field players
Prince Fielder has played in 485 games.
Matt Kemp, 482.
Nick Markakis, 481.
Robinson Cano, 480.
Adrian Gonzalez, 479.
Dan Uggla, 478.
James Loney, 477.
Billy Butler, 476.

Injury-prone players

Granted, this doesn’t paint the full picture; some of these injuries can be considered “freak” injuries, rather than the accumulation of bumps and bruises. Particularly brittle are Rollins (three DL trips in the last two years; four in the last four years) and Cabrera, who has hit the DL twice in the last three years, and missed 11 games with dings and dents in 2011.

Jason Bay has played in 369.
Stephen Drew, 372.
Chase Utley, 374.
Asdrubal Cabrera, 379.
Hanley Ramirez, 385.
Jimmy Rollins, 385.
Dustin Pedroia, 388.

There’s no cutting edge research here, but those numbers are just something to keep in the back of your mind at a H2H draft. When given the choice between Pedroia and Cano, would you rather have the advantage of 30 more games a year? Are rhetorical questions annoying?


Comments are closed.