Cockcroft: Time to replace Wins as a fantasy category


I feel for Scott Baker.

What must it be like to pitch your heart out, yet come out on the losing side of a 1-0 score not once, but twice in a 13-day span? As noted in my Monday Out of the Box, Baker has actually started three 1-0 team losses this season, most in the majors.

No matter what Baker might say publicly about his plight, it's hard not to imagine him thinking something along the lines of "boy, if only my teammates could've scored me a run or two." And even if not, you know his fantasy owners certainly felt that way.

When are we going to accept the fact that wins and losses are largely a function of team performance, not the individual? In today's ever-changing game, it's unfair to put as much importance on a pitcher's win-loss record as we do. Those are team statistics.

As I migrate that thought back to our grand fantasy game, the solution seems simple: We need to do away with wins as a Rotisserie category.

I don't suggest it casually; this has been a thought a long time coming. It's important that every so often, we take a step back from the daily grind of managing our fantasy teams, and evaluate the structure of our game as a whole. And as I do that, what I see are nine Rotisserie categories that, while maybe not perfect, at least offer a decent interpretation of a player's level of success on the playing field. Then I see one that is far from perfect.

I remember hearing all the past cases that stolen bases are on the decline and therefore need to go. Or that saves is a silly statistic. Or that on-base percentage is a far superior choice among hitting categories than batting average. For the record, I'm on board with that last one, but even that isn't the most problematic of the 10 standard categories.

Consider this: For all the debates against other categories, only with wins does there exist a scenario in which the player has positively zero control over his statistical outcome. Hit a home run and you've got your homer, RBI or run scored, and you've boosted your batting average. Strike a hitter out and you've got your K, as well as lowered your ERA and WHIP. Come into a "save situation," and you can earn yourself a save. Stolen bases are self-explanatory, and even if you want to make the ridiculous case that a manager could give all of his players the red light, can he really force them to listen?

But only with wins can you throw a no-hit, no-walk masterpiece, yet lose the game 1-0 if you're so unfortunate as to suffer some poorly timed errors.

Here's the other problem: There are simply fewer big winners than there were a quarter-century ago. In 1980, the year commonly accepted as the first year of Rotisserie Baseball, there were seven 20-game winners. In the first five full seasons in the "Rotisserie Era" (1980 and 1982-1985), there were 21 total. By comparison, only one pitcher won as many as 20 games in either of the past two seasons (Josh Beckett in 2007). This season, only two pitchers are on pace to win 20 -- Cliff Lee and Brandon Webb -- and neither is on a pace for better than 21 total, leaving little margin for error.

Plus, the category as a whole has certainly changed since 1980. Some comparisons:

Those might seem like small amounts, but when you're talking about hundreds upon thousands of starts, it adds up.

Picking a replacement

But ditching wins as a category isn't the tough part. I doubt many people would mourn its loss. It's picking a category to replace it that is a challenge.

I've heard many suggestions: Strikeouts per nine innings (K/9) is an option, though that'd tie too much into strikeouts. Walks per nine (BB/9) is another, though that's too closely linked to WHIP. Batting average allowed could work, though the problem is that it would mean three ratio categories on the pitching side compared to one hitting, further increasing the chances of teams employing the all-reliever strategy to dominate ERA, WHIP, saves and BAA. Besides, the fact that the pitchers who are going to post the lowest BAA are the same ones who excel at ERA and WHIP: Relievers.

Innings pitched is an option, but frankly, that one's no fun. Sure, it represents the number of outs a pitcher records, and outs are the most important things for a pitcher, but if you want a quick way to make streaming a screaming success, add innings pitched to your scoring. (Much easier to amass innings than it is wins!)

Then there's the quality start. Oh, the debates that surround the quality start. Too many to get into in this space, but when you're thinking about success as it pertains to fantasy baseball, consider this: The minimum qualification for a quality start still results in the pitcher logging a 4.50 ERA (six innings, three earned runs).

That's simply not what I'd call a "win-worthy" start.

So here's my proposal: We tweak the quality start, instead call it a winnable start, and use that to replace wins. How we do that: Instead of the six-inning minimum, we raise the cut-off to seven frames. Then, while we keep the runs maximum at three, we go by runs, rather than earned runs.

Not that I'm opposed to using earned runs, but to trust that category, I'd first require a stingier standard in which to judge what runs are earned or unearned. Something about the two-out, none-on error that leads to the six-unearned-run inning doesn't sit well with me. Shouldn't a pitcher be held at least somewhat accountable for his actions, even if they might have been the result of his being inconvenienced by his fielders?

A few more statistical reasons to go with seven and three: The collective group of relief pitchers tends to manage an ERA right around four, so leaving them two innings of work means tacking an extra eight-ninths of a run onto the starter's three, and if you count the typical reliever's unearned runs, it's probably closer to a full run. That's a four-run night allowed by the pitching staff, and the average MLB team usually averages four and a half, which I'd call enough of a winning margin. And remember, we're talking the minimum cutoff here; a half-run win is the worst to qualify.

Sure enough, the statistics seem to back up the idea. Through Tuesday, 767 "winnable starts" have been recorded this season; they resulted in 471 actual wins, or 61.4 percent. I'd call anything that provided better than a 60-percent chance of your team winning the game a success. For one, the 1993 San Francisco Giants represent the last team to win at least 60 percent of their games yet fail to make the postseason. It's pretty rare.

By comparison, the chart above showed only 53.6 percent of quality starts so far this season resulted in a win for the pitcher. Sorry, but that's simply not a high enough success rate; four MLB teams actually won at least that often in 2007 yet failed to reach the playoffs. That suggests that the quality-start standard simply isn't strict enough for our purposes. It might be fine for the real game, but for fantasy, remember, we're replacing a 29-year-old categorical staple. The replacement needs to be restrictive.

So to sum up my proposal to every commissioner in a 4x4 or 5x5 standard Rotisserie league: eliminate wins, and replace it with winnable starts (seven-plus IP, three runs or fewer). We'll call it "WS" for short (fitting abbreviation, isn't it).

The benefits of "WS"

Obviously, the primary benefit of shifting your league from wins to winnable starts is no longer having to fret when your starting pitcher departs a game in the ninth with a 3-2 lead if you have no faith in his team's closer. Pitchers earn their reward in the category for turning in a strong start; their offense and their bullpen can't take it away.

All this leads me to think two things: Finally, I no longer have to pretend Mike Mussina is a top-25 fantasy starting pitcher and wow, it's about time Cole Hamels got a little more credit for how remarkably dominant he has been this season.

I don't toss those two names out casually. Hamels, through Tuesday, did lead the league in winnable starts. Mussina, meanwhile, had 12 wins … but two winnable starts. Here are the leaders in either category, with comparisons between the two:

Another benefit of the winnable start is that there are fewer to go around than there are wins. That makes the value of an ace pitcher stronger, and it decreases the appeal of the streaming-starters strategy. It also increases the value of a quality middle reliever, because it'd almost always make more sense to pick a reliever to bolster your ERA and WHIP than to take a middling-to-poor starting pitcher who's a long shot bet at best for a WS.

And isn't that what fantasy baseball is supposed to be about anyway, demonstrating your ability to pick quality performances?

Making the transition

Now here's where it gets a bit tricky, because at nearly two-thirds of the season in the books, wholesale change is not going to prove a welcome thing in any fantasy league.

If you're in line with my proposal, great. Just make sure you do it in 2009, not 2008.

Rule changes have no business being discussed this deep into a fantasy season. Those that have to do with the general structure of a fantasy league especially -- scoring, rosters, etc. -- should never be entertained in-season. Save such things for the offseason, when a formal league vote can be held (if needed), and your league's owners can take more time to adapt to the inevitable changes that come with altering a league's scoring system.

To help you ease the transition, next season we'll offer you the option of this new category -- "winnable starts" -- as a replacement for wins. I'm that serious about we, the fantasy community, updating our game for the new millennium. We at ESPN will be happy to offer you the choice of change. Make the switch if you agree with me on wins being a stale, outdated, crusty old category. Or, if you disagree, it's perfectly OK to stand pat. [Editor's note, March 31, 2009: We were unable to implement this category addition this season, but it will be considered for a future release]

But just do me one favor: If you pick the latter, don't come crying when Johan Santana gets another win yanked from under his nose by his bullpen.

Tristan H. Cockcroft covers fantasy sports for ESPN.com. You can e-mail him here.