I've always found the best starting point for creating my annual draft lists is to determine exactly how much value a player actually contributed to a fantasy team in a standard 5x5 rotisserie league in the previous season, and then adjust from there based upon how likely each individual player was to equal, or even improve upon, those totals in the upcoming year. However, while this method had always worked quite nicely in the analysis of hitters, when it came to evaluating pitchers, the numbers never seemed to come out right.
Most of the problem lies in the fact that the categories most leagues use to value pitchers don't properly map onto an individual pitcher's skill set. For example, let's take National League Cy Young Award winner Tim Lincecum and compare him to teammate Matt Cain.
Both pitchers had the same number of starts, and clearly nobody can argue that with a better ERA, a lower WHIP and more strikeouts, Lincecum was the better pitcher for your fantasy team. The problem is that this makes up only three of the five pitching categories in a standard 5x5 rotisserie league. Neither will do anything to help you at all in saves, and when it comes to wins, there's more there than meets the eye. Yes, Lincecum won 10 more games than Cain, but if you look at his run support, the Giants scored nearly a run and a half more per game in which Lincecum pitched. How huge a difference is that? Quite a bit, actually.
If we look at the number of games Matt Cain started in which the Giants lost by either one or two runs, we see that on 11 occasions, just a little bit of Lincecum-esque run support could have turned losses or no-decisions into victories. Similarly, if Tim Lincecum's appearances were greeted with the same offensive apathy his teammate faced on a daily basis, 12 of his victories or no-decisions could easily have ended up in the loss column. How much differently would we look at this duo's 2008 season (and therefore, their prospects for 2009) if Cain were 19-9 and Lincecum were 11-10, in spite of the fact that nothing would have changed in the value of their contributions to the three categories that rely solely on their own efforts, rather than the support of their teammates at the plate?
Now let's compare Matt Cain with another celebrated fantasy performer from last season, Gavin Floyd. Who was truly better?
Our own ESPN Player Rater ranked Floyd at No. 37 among all pitchers for the season, with Matt Cain at No. 75. But is this really an accurate assessment of which pitcher was better? Certainly, Floyd won a lot more games for you in 2008, and therefore was "more valuable;" however, how much of this was based on luck? After all, if you flipped the run support these two pitchers received, and in so doing greatly altered the win-loss records of both, there's no way you'd pass on Cain for Floyd, since Cain had a better ERA, more strikeouts and only a slightly worse WHIP.
So that's where my Universal Pitcher Values (UPV) come in. Rather than rely on a pitcher's actual win total to determine his likelihood of success, we're going to analyze the statistics that make it more likely that a pitcher will earn victories, regardless of the support he may or may not receive from his teammates. Certainly we have to include strikeouts, as well as walks and hits allowed, since K's and WHIP are very real categories. However, we're not going to use ERA as a measure. Instead, we're going to use ARA, or actual run average.
In other words, if a pitcher gives up five runs in the top of the first inning, the chances of him winning that game are no different if those runs were earned or not. He's still down 5-0 regardless of who is to blame. So we'll take defensive errors out of the equation, and rate pitchers on exactly how many runs crossed the plate when they were pitching, not just the earned ones, since they all count the same. Additionally, we're going to reward a pitcher for going deeper into ballgames. After all, a starter can't win a game if he doesn't make it through the fifth inning, and the longer he stays on the mound, the more chances he gives his team to muster up enough offense to earn him a victory. Durability is a trait to be admired.
Taking all these stats and assigning a value to each, based upon league-wide averages, we can create a truer picture of the pitching hierarchy for 2008, and then use that as the jumping-off point for adjustment based on things such as age, injury, new teams and new roles. (Note: We've also kept Greg Maddux and Mike Mussina in the mix for now, so you can see where their 2008 performances would have ranked, but obviously you'll want to cross them off your lists before Draft Day, lest you face extreme ridicule for drafting a retiree.) Here's a sneak peek at the Top 25:
Incidentally, in case you're curious, Matt Cain clocks in at No. 31, a great deal ahead of Gavin Floyd at No. 57. You can check out the complete list here, but first I want to take a moment to address relief pitchers, and where their fates lie in this evaluation process. You'll notice that there isn't a closer to be found in the Top 25, and in fact only Carlos Marmol and Mariano Rivera register at all in the top 40.
The fact is that if it weren't for saves, relief pitchers probably wouldn't be on fantasy rosters at all. Most of them don't pitch enough innings to accumulate as many strikeouts as starters do or have a positive impact on your team's ERA or WHIP. However, with less room for error, one bad outing can severely, and quite possibly permanently, damage your fantasy staff's statistics for the season. And while a middle reliever's vulture wins count just as much as anyone else's, predicting which pitchers will be the beneficiaries of late-inning rallies is a complete roll of the dice.
This is why you need to have a closer on your team in order to not simply give away 20 percent of your potential pitcher-based fantasy points without a fight. However, you should rarely, if ever, draft one early. After all, Brad Lidge had 41 saves for the Phillies last season. Well, so did Brian Wilson of the Giants who, as a team, lost 20 more games. As brilliant as Francisco Rodriguez's 2008 was, Joakim Soria's wasn't too far off, and he probably could have been had about 10 rounds later than K-Rod in most of last year's fantasy drafts. Closers, for the most part, should be treated just as you would a Joey Gathright or a Michael Bourn. They're one-category specialists, and if you're lucky enough that they don't kill you in other areas, then anything else you do manage to get from them is merely icing on the cake.
AJ Mass is a fantasy football, baseball and college basketball analyst for ESPN.com. You can e-mail him here.