The Commish Online                                                                                
Stat Oddities in '07
August 15, 2007

According to an article in the Wall Street Journal (yes, the WSJ), the Arizona Diamondbacks, if their current pace continues, would be the first team to hold their league's best record while being outscored by their opponents.  The article surmises the feat as nearly impossible to continue throughout the season, according to some professional number crunchers. 

In reality, their 91-71 pace is hardly earth shattering and suggests that the entire league as a whole is mediocre if 91 wins ends up being the most in the National League.  Run differential, while proving to be somewhat accurate over the long haul, isn't so much a predictor of future outcomes but rather an attempt to quantify past performance above or below expectations.  In other words, it can be argued equally that wins are the stronger statistic and perhaps the run differential will trend closer to the win percentage rather than the other way around.

Because so much depends on the starting pitcher, it is likely that the D-Backs have experienced clusters of really good starting pitching and clusters of really bad starting pitching.  I could run the analysis, but all you really need to look at are three recent losses:  on July 29, August 2, and August 14, they lost 14-0, 11-0, and 14-5 respectively.  Arizona was outscored by 34 runs in those three losses.  Arizona is only being outscored on the season by 30 runs total, so you can see the lack in meaning in the totals.  14-0 and 14-13 are each just one loss, but they alter the run differential significantly.  If the D-Backs would have put up a fight in any of those shellackings, they would be much closer to a positive run differential, making this column irrelevant.

Even more peculiar is what's happening in Seattle.  According to Steve Phillips of ESPN, he mentioned yesterday that the starters for the Mariners have a LOSING record but (assuming they hold off the Yankees and company) could become just the third team to do so and make the playoffs.  Now that is a stat that defies the odds.  Quite simply, someone on the team must record the win or the loss, and when a team is 15 games over .500 (Seattle is currently 66-51) and the starters are under .500, either the offense is adept at comebacks or the relievers are vulturing some wins thanks to blowing a lead and "earning" the win themselves.  The reality is that both events are happening.

Look no further than young lefty reliever Eric O'Flaherty.  He has the second highest ERA among relievers used regularly by Seattle (20 or more appearances) yet has a sparkling 7-0 record with just 39.7 IP.  Either O'Flaherty has an amazing ability to hold a tie until the offense comes around, or he has been the lucky recipient of a few comebacks. 

With the relief corps boasting a record of 23-7 as a whole, it is safe to assume that the late inning pitchers are doing their job by preventing the game from "getting away from them."  In turn, they are receiving a hefty treatment of wins.  Oddly enough, the offense scores the most runs per plate appearance in innings 4 through 6, but the LEAST amount in innings 7 and beyond.  It would be expected to see a jump in scoring later in the game for a team with so many relief wins, but the reality is that Seattle's relievers are good (but not great, ranking in the top third in the AL in ERA) and VERY lucky.  Timing is everything, and Seattle will need a lot more of it in the next month to hold off New York.

The fact that Arizona tends to let a few games "get away from them" shouldn't hurt them in the long run - anomalies rarely do, especially come October when those pitchers contributing to the anomalies are watching from the dugout.  Seattle, on the other hand, better stay in prayer mode because timely pitching AND hitting won't last forever.