Categorical improvements

Going to a 6x6 Rotisserie format can really improve fantasy baseball

Updated: February 22, 2013, 12:24 AM ET
By Tristan H. Cockcroft | ESPN.com

Baseball has undergone quite a statistical revolution in the past 30-plus years, yet remarkably, Rotisserie Baseball, a game based upon the real game's statistics, has remained largely unchanged during that time.

Sure, since 1980, many Rotisserie leagues have migrated from 4x4 to 5x5 scoring, adding runs scored and (pitcher) strikeouts, but the same core categories -- including categories many sabermetricians regard flawed -- remain.

[+] EnlargeCliff Lee
AP Photo/Marcio Jose SanchezCliff Lee's six wins in 2012 was nowhere close to showing how good a pitcher he actually was.

For example, during the history of fantasy baseball, we have learned valuable lessons about two of Rotisserie's longtime staples:

• That batting average, most notably, both fails to credit a hitter's ability to avoid making outs and is subject to random fluctuation. It has come under the most criticism for discounting walks, a hitting skill that has skyrocketed in appreciation the past 30-plus years.

• That wins, while always a questionable measure of individual pitching skill, has become increasingly so as team pitching staffs grow more specialized. Its value has diminished as starting pitchers' workloads have lessened and bullpens have taken on greater importance.

That's just a quick glance, as we've also learned things about other longtime Rotisserie categories: That RBIs and runs scored, like wins, are context-dependent, too reliant on teammates and not a good enough measure of individual skill; that saves have too broad a definition and aren't necessarily the best indicator of relief success; and that earned run average (ERA), like batting average, isn't the ideal barometer for pitching skill because of its variance depending upon ballpark, team defense and, frankly, official scorer's decisions.

But the goal of this column is not to nitpick all 10 traditional Rotisserie categories. We continue to learn new things about the game every year, so it's possible we have not yet arrived at the ideal mix of statistics to measure player skill.

The goal of this column is to make our game better. It is to inspire forward thinking, to generate ideas how to change with changing times. It is to give you one big idea that might inspire you to take the next big leap.

That big idea in which Rotisserie Baseball can evolve involves removal of both the batting average and wins categories, while migrating to a 6x6 scoring system that incorporates some stronger indicators of player skill. It is a long-overdue change, and one that should increase both the levels of knowledge and enjoyment of this generation of fantasy baseball players.

This new Rotisserie 6x6 scoring, six categories apiece for hitters and pitchers -- one first mentioned nearly three years ago -- uses the following statistics:

In addition to removing the aforementioned batting average (AVG) and wins (W) from the equation, it also replaces strikeouts (K) with strikeouts-per-nine-innings ratio (K/9).

Let's discuss the rationale behind the hitting changes first.

[+] EnlargeAdam Dunn
Brace Hemmelgarn/US PresswireAdam Dunn's ability to draw walks is often ignored in many fantasy leagues, but adding on-base percentage as a category helps his value.

By adding on-base percentage, we finally reward players for walks, the most glaring omission on the hitting side in Rotisserie Baseball history. Using 2012 statistics, there's perhaps no stronger argument for migrating from batting average to on-base percentage than this: Alex Rios, who batted .304 in 605 at-bats, was rated one of the 20 most valuable in terms of batting average on our 2012 Player Rater. Meanwhile, Chicago White Sox teammate Adam Dunn's .204 batting average in 539 at-bats rated second-worst in the category. Dunn, however, reached base via hit, walk or hit by pitch two more times than Rios (216-214), had a near-identical on-base percentage (.333 to Rios' .334) and, per Baseball-Reference.com, committed eight fewer outs (442-450).

Slugging percentage, meanwhile, credits players for their ability to generate extra-base hits, as doubles and triples were entirely ignored in Rotisserie 5x5, while providing the necessary six-category balance on the hitting side. Alex Gordon is an outstanding example of a player who doesn't receive enough credit for these valuable contributions. In the past two seasons combined he has the fifth-most extra-base hits (142) and a slugging percentage (.478) that ranks 36th out of 124 qualifiers; his 37 home runs during that span, however, rank only 54th.

Now, let's analyze the pitching changes.

By removing wins, we eliminate the potential of frustration over poor-fortune seasons such as Cliff Lee's 6-win 2012. Lee was the year's poster boy for bad luck in the category; he managed 21 quality starts (17th-most in the majors), 12 games of at least seven innings and two earned runs or fewer (14th-most), and five games of at least eight innings and zero or one earned run (10th-most). Yet Lee's wins total was low by historic proportions; he set a new all-time record for fewest wins by any pitcher who made at least 30 starts, struck out 200 hitters and had an ERA below 3.25. In fact, only two times in 399 seasons all-time that met those criteria did a pitcher win fewer than 10: Nolan Ryan's 1987 (8 wins) was the only other time.

It's the addition of quality starts that most often raises fantasy owners' eyebrows. I'm not lauding it as a change resulting in perfection; I am suggesting it as a change to better our measures of starting-pitching skill.

Most commonly, critics of the quality start claim that, at its minimum qualification -- a starter must pitch at least six innings while affording no more than three earned runs -- a pitcher's ERA is 4.50. Perhaps those minimums should be more rigid, but criticizing the minimum qualification for a quality start isn't a compelling argument if you're also a proponent of Wins. Consider that 318 out of the 2,484 quality starts accrued by all pitchers in 2012, or 13 percent, resulted in a single-game ERA of 4.00 or greater. By comparison, 292 out of the 1,738 wins tallied by starting pitchers in 2012, or 17 percent, resulted in single-game ERAs of 4.00 or greater. In fact, 186 of those wins -- a whopping 11 percent of all starters' wins -- resulted in single-game ERAs higher than 4.50 -- the worst ERA a pitcher could have while qualifying for a quality start. Heck, 24 times last season a starting pitcher won despite sporting a single-game ERA of 9.00 or higher!

Wins are also inherently fluky when you're talking about those awarded to relief pitchers. Last season, relievers were responsible for 692 total wins. Of those 692, 50 were "earned" by relievers who faced only one batter; 57 were "earned" by relievers who had single-game ERAs of 9.00 or higher; and 38 were "earned" by relievers who both blew a save and allowed at least one run.

The addition of innings pitched is a cap-tip to one of the most increasingly appreciated skills of a pitcher during the fantasy baseball era: generating outs. Remember, innings pitched is a direct calculation of outs recorded by a pitcher, innings pitched times three equaling his number of outs recorded. A pitcher has but two goals when he's on the mound: record outs and prevent runs.

Why pitchers haven't received more credit historically for their innings pitched in fantasy baseball is puzzling, but as the real game evolves, pitching staffs become more specialized and efficiency of outs recorded further drives managerial strategy, individual pitchers' innings totals will greater reflect their contributions. A quarter-century ago, in 1988, 37 pitchers totaled 215 innings or more, and 20 of them had WHIPs higher than 1.20. Last season, 12 pitchers worked at least 215 innings, and only Clayton Richard (1.23) had a WHIP higher than 1.20.

[+] EnlargeCraig Kimbrel
Daniel Shirey/US PresswireCraig Kimbrel is already one of the best fantasy relievers because he gets saves and piles up strikeouts, but using K/9 as a category could boost his value even more.

Finally, the rationale for switching strikeouts from a counting to rate category is twofold: One, it helps counterbalance the loss in value of relief pitchers as a whole due to the addition of a sixth category, and two, it helps diminish somewhat the streamer's strategy. In a six-category system in which the only categories a relief pitcher can influence are saves, ERA, WHIP and strikeouts, relievers as a whole would have their value depressed to the point where it can be argued that only Craig Kimbrel/Kenley Jansen types, legitimate 100-K relievers, would be the ones who warrant a top-100 pick in a draft. Fantasy owners would also load their lineups with starters, diminishing the value of an important part of today's game: bullpens.

Granted, moving to K/9 ratio hurts the low-strikeout relievers: Jim Johnson (5.37 K's per nine) and Jonathan Broxton (6.98) were two of the four closers with at least 25 saves in 2012 to average fewer than eight K's per nine innings. That'd have been true whether K's were counting or ratio, however, and in defense of it being a rate statistic, a high-rate closer like Kimbrel (major league-leading 16.66 K's per nine) actually gains value in this new format. A pitcher like Kimbrel, frankly, deserves more credit than he seems to receive. So does a pitcher like Jansen, whose impact in K's per nine would be greater than in total strikeouts, meaning that even if he doesn't close a single game in 2013, he'd be a more appreciated commodity in a K/9 system.

As for mitigating the streamer's strategy, remember that strategy mostly targets volume -- counting numbers like wins and strikeouts -- or 50 percent of the four categories that a starting pitcher influences in traditional Rotisserie. In a six-category system with three ratio departments, a starting pitcher can influence only five, and 40 percent of those (two of five) are quantitative: quality starts and innings pitched. And since quality starts has a direct connection with ERA and WHIP, it becomes more important to pick quality, rather than quantity, of matchups.

Again, the above Rotisserie 6x6 proposal isn't meant to be a perfect system; rather it is pitched a potential improvement upon traditional Rotisserie 5x5. It is a creative change for fantasy owners seeking a more modernized scoring system, or who have been frustrated by the shortcomings of batting average or wins. There might be -- and presumably is -- an ultimately more ideal scoring system to come in the future, and maybe this is step one towards it. Frankly, if baseball continues to shift its focus from batting average and closer to advanced statistics like wins above replacement (WAR), a first step will soon be mandatory in order to keep fantasy baseball up with the times.

Tradition is understandable in the game of baseball: Look at the past fall's Miguel Cabrera-versus-Mike Trout American League Most Valuable Player debate. Some fantasy owners will prefer to stick with tried-and-true categories in the absence of a proven-perfect alternative. This is all about what statistics you value and what gives you enjoyment. That choice is yours.

I will, for example, play in no fewer than four Rotisserie 5x5 scoring leagues in the 2013 season, and I will garner no less enjoyment in those -- longstanding expert leagues like the League of Alternative Baseball Reality (LABR) and Tout Wars as notable examples -- than in my Rotisserie 6x6 league, entering its second year.

That Rotisserie 6x6 league, however, was well received by its ownership base last season, and has the added advantage of having spawned much open debate about player skills and the value of certain statistics in the game of baseball.

And that's really the fun of all this, isn't it?

Rankings for Rotisserie 6x6/Quality Starts/OBP-SLG leagues

In order to help owners new to either Rotisserie 6x6 leagues, or formats that count quality starts, on-base percentage and/or slugging percentage, listed below are adjusted top 50 hitter and starting pitcher rankings for this column's proposed scoring system. The column to the left ranks hitters if on-base percentage and slugging percentage replaced batting average, while the column to the right ranks starting pitchers if quality starts replaced wins, innings pitched was added and K/9 ratio replaced strikeouts.

Note that starter pitching rankings wouldn't be terribly different in a league that only replaces wins with quality starts; these rankings serve the dual purpose of helping owners seeking help in either format.

ALSO SEE