luck and fantasy
3 January 2011
“This [Chiefs-Raiders] game features a top-10 fantasy QB (Matt Cassel), two of the top-5 fantasy running backs (Jamaal Charles and Darren McFadden) and the best fantasy receiver (Dwayne Bowe), proving yet again that FANTASY FOOTBALL IS A COMPLETE F——— CRAPSHOOT AND WE ARE IDIOTS FOR DOING FOUR WEEKS OF MOCK DRAFTS TRYING TO FIGURE OUT WHAT THE F—- IS GOING TO HAPPEN BECAUSE THERE IS NO F——— WAY TO KNOW!!!!!!”
Thank you, Bill Simmons. Even though I am very, very serious about fantasy sports, for years I’ve been unable to even read fantasy football analysis – even while pooping at work (which should really tell you something) – just by virtue of the fact that it’s the fantasy equivalent of simply rolling a fucking pair of dice. It’s almost completely impossible to predict, and anyone who tells you otherwise is wrong.
We’re talking here about 16 game sample sizes for players, a number so small that it lends itself to flukes. (And no, I have absolutely no background in statistics.) Further, football is a true “team” sport, in that the performance of each player is heavily affected by the players around him: if a QB’s o-line stinks, he doesn’t have time to throw the ball; if a team’s defense stinks and gives up a lot of points, they’re going to passing a lot to play catch up, thus raising a QB’s numbers and lowering an RB’s; a WR can have all the talent in the world, but if the QB blows, well, you get it.
(And I’m sure there are better ways to articulate these points, but give me a break – it’s 10am on a Monday morning.)
Baseball, on the other hand…oh, baseball. First, we’re talking 162 games (or 30+ starts or 200+ innings pitched or 600+ PAs), so there’s a strong sample size from which to base conclusions. Second, of course baseball is a team sport and an individual’s numbers rely on his teammates (a ground-ball pitcher playing in front of a terrible defensive infield would likely take a hit in the ERA dept, etc), but for the most part it’s pitcher vs. hitter. And third, unlike football, baseball has a GINORMOUS amount of nerd stats that, as I said, allow one to fairly accurately predict the performance of an individual player.
A few of my favorite nerd stats (and I don’t want to blow my load here, as we’ll revisit this in March when prepping for the season):
- BABIP (batting average on balls in play). League average in 2009 was about .316. Therefore, if a player hit higher than that in 2009, it’s likely his batting average would be lower in 2010. If his BABIP was lower than .316 in 2009, it is likely that he would raise his batting average in 2010, as this number tends to normalize.
Examples: In 2009, Hanley Ramirez hit .342, with a BABIP of .384. In 2010, he hit .300, because his BABIP returned to a more normal .327. Alternatively, Paul Konerko batted .277 in 2009 with a low BABIP of .282. In 2010, his BABIP rose to .326, and he batted .312, nearly 40 points higher. Konerko was targeted as a “like” in my fantasy baseball preview last year, and while no one is ever going to stay away from Hanley Ramirez, the high BABIP was noted.
- For pitching, I mostly focus on three stats: BABIP (in this case, BABIP against), LOB% (left on baseball percentage) and what I call E-F, which is ERA minus FIP, or Fielding Independent Pitching, a fancy way of assigning a number to what a pitcher’s ERA would be minus the effects of defense, ballpark, etc.
In 2009, average BABIP was .297, average and average LOB% was 73.86%. In 2009, Max Scherzer had a 4.12 ERA and 1.34 WHIP, because of his BABIP (.323) and LOB% (68.7%) were far from league averages. In 2010, his ERA and WHIP improved to 3.50 and 1.25, respectively, because his BABIP (.295) and LOB% (74.9%) normalized. I “liked” him in my 2010 preview.
As for E-F, let’s take the case of Cole Hamels. In 2009, Cole posted an ERA and 4.32 and WHIP of 1.29. His FIP was 3.72, meaning his E-F was 0.61. This doesn’t seem like a lot, but it’s a red flag that means that this pitcher is targeted for a rise the following year. And what happened in 2010? Hamels ERA was 3.06 and his WHIP 1.18. Once more, I ”liked” him in my 2010 preview.
The point? If you want to use actual science and make some money, stick to fantasy baseball (I’ve been making about one rent payment a year for the past five years with my winnings). And next time you’re tempted to read a 3,000 word article written by a fantasy football guru about who’s going to perform well in Week 14 match-ups or in the big Monday night game, instead put it down and ask the opinion of your favorite toddler – they’re guess is just about as good as any fantasy football “expert’s.”
Thank you, Bill Simmons. Even though I am very, very serious about fantasy sports, for years I’ve been unable to even read fantasy football analysis – even while pooping at work (which should really tell you something) – just by virtue of the fact that it’s the fantasy equivalent of simply rolling a fucking pair of dice. It’s almost completely impossible to predict, and anyone who tells you otherwise is wrong.
We’re talking here about 16 game sample sizes for players, a number so small that it lends itself to flukes. (And no, I have absolutely no background in statistics.) Further, football is a true “team” sport, in that the performance of each player is heavily affected by the players around him: if a QB’s o-line stinks, he doesn’t have time to throw the ball; if a team’s defense stinks and gives up a lot of points, they’re going to passing a lot to play catch up, thus raising a QB’s numbers and lowering an RB’s; a WR can have all the talent in the world, but if the QB blows, well, you get it.
(And I’m sure there are better ways to articulate these points, but give me a break – it’s 10am on a Monday morning.)
Baseball, on the other hand…oh, baseball. First, we’re talking 162 games (or 30+ starts or 200+ innings pitched or 600+ PAs), so there’s a strong sample size from which to base conclusions. Second, of course baseball is a team sport and an individual’s numbers rely on his teammates (a ground-ball pitcher playing in front of a terrible defensive infield would likely take a hit in the ERA dept, etc), but for the most part it’s pitcher vs. hitter. And third, unlike football, baseball has a GINORMOUS amount of nerd stats that, as I said, allow one to fairly accurately predict the performance of an individual player.
A few of my favorite nerd stats (and I don’t want to blow my load here, as we’ll revisit this in March when prepping for the season):
- BABIP (batting average on balls in play). League average in 2009 was about .316. Therefore, if a player hit higher than that in 2009, it’s likely his batting average would be lower in 2010. If his BABIP was lower than .316 in 2009, it is likely that he would raise his batting average in 2010, as this number tends to normalize.
Examples: In 2009, Hanley Ramirez hit .342, with a BABIP of .384. In 2010, he hit .300, because his BABIP returned to a more normal .327. Alternatively, Paul Konerko batted .277 in 2009 with a low BABIP of .282. In 2010, his BABIP rose to .326, and he batted .312, nearly 40 points higher. Konerko was targeted as a “like” in my fantasy baseball preview last year, and while no one is ever going to stay away from Hanley Ramirez, the high BABIP was noted.
- For pitching, I mostly focus on three stats: BABIP (in this case, BABIP against), LOB% (left on baseball percentage) and what I call E-F, which is ERA minus FIP, or Fielding Independent Pitching, a fancy way of assigning a number to what a pitcher’s ERA would be minus the effects of defense, ballpark, etc.
In 2009, average BABIP was .297, average and average LOB% was 73.86%. In 2009, Max Scherzer had a 4.12 ERA and 1.34 WHIP, because of his BABIP (.323) and LOB% (68.7%) were far from league averages. In 2010, his ERA and WHIP improved to 3.50 and 1.25, respectively, because his BABIP (.295) and LOB% (74.9%) normalized. I “liked” him in my 2010 preview.
As for E-F, let’s take the case of Cole Hamels. In 2009, Cole posted an ERA and 4.32 and WHIP of 1.29. His FIP was 3.72, meaning his E-F was 0.61. This doesn’t seem like a lot, but it’s a red flag that means that this pitcher is targeted for a rise the following year. And what happened in 2010? Hamels ERA was 3.06 and his WHIP 1.18. Once more, I ”liked” him in my 2010 preview.
The point? If you want to use actual science and make some money, stick to fantasy baseball (I’ve been making about one rent payment a year for the past five years with my winnings). And next time you’re tempted to read a 3,000 word article written by a fantasy football guru about who’s going to perform well in Week 14 match-ups or in the big Monday night game, instead put it down and ask the opinion of your favorite toddler – they’re guess is just about as good as any fantasy football “expert’s.”
