What A Relief!

Quick, name the top two relievers in terms of WPA/LI… no, it isn’t Francisco Rodriguez. Not Brad Lidge, either. Brad Ziegler is a good guess, but sadly, no, he is not in the top two. If you said Mariano Rivera, kudos, because you’re getting warmer, but he comes in at spot number three. The top two relievers this year have been Hong-Chih Kuo of the Dodgers and Scott Downs of the Blue Jays. How many of you honestly guessed them? And, for the record, a guess does not involve hovering over the leaders tab, clicking ‘majors’, clicking ‘win probability’, clicking ‘relievers’, and then sorting by WPA/LI.

Kuo, a Dodgers lefty, leads all of baseball with a 2.35 WPA/LI, which basically means he has been worth almost two and a half wins more than an average relief pitcher. Downs, a Toronto lefty, has a 2.19 WPA/LI, close to Kuo, but still distanced by a good margin. To further clarify, Lidge ranks 13th at 1.36, Ziegler 17th at 1.34, and K-Rod 34th at 0.99.

While both pitchers have some statistical similarities, it seems they are getting their jobs done in different ways. Kuo has pitched in 40 games, three of which were starts, amassing 78.1 innings. He has allowed just 55 hits, only four of which left the yard. Additionally, his very low walk-rate—19 total in his 78.1 innings—has combined with the lack of hits to produce a 0.94 WHIP. He also strikes out a ton of hitters to the tune of 94 total and a 10.80 K/9.

Essentially, Kuo barely allows balls to leave the yard, barely allows baserunners, and strikes out many. Yep, I would say that is a formula for a tremendous relief pitcher. On the year, his 2.07 ERA is largely supported by a 2.20 FIP. His 79.8% LOB is great but not earth-shattering and is definitely sustainable, as we have seen elite relievers sustain very high strand rates. Kuo, who throws his 93 mph fastball over 80% of the time, is also more of a groundball pitcher, evidenced by his 1.48 GB/FB.

Downs, however, is MUCH more of a groundball pitcher. In his 61 appearances and 66 innings, just 11.9% of balls in play have been line drives, with 65% being hit on the ground. His GB/FB comes in at 2.88, almost double the rate of Kuo. Downs has surrendered just 47 hits, only two of which are home runs, but his walk and strikeout rates are not as dominant as Kuo’s. With 23 walks and 56 strikeouts, his 1.23 ERA vastly outdoes his 3.02 FIP.

Downs’ WHIP is higher as well, at 1.06, however he has stranded 89.9% of the runners that reach base; even though he allows more, he has stranded a much higher percentage than Kuo. I have heard and read Blue Jays fans hoping he could become a starter, but to do that he will really need to develop another pitch. At this juncture, he literally seems to use only a fastball and curveball, as both have accounted for just about 97% of his deliveries.

Kuo definitely seems to have the more sustainable and talent-driven line, but that should not take anything away from Downs’ effectiveness in relief this year. Both have been outstanding, and neither seem to garner the publicity they are due. Relievers are a fickle bunch, so who knows if they will be anywhere near the answer to this same question next year, but they have been rocks for their respective teams this season.



I am a 22-yr old baseball freak from Philadelphia currently writing for Statistically Speaking with occasional contributions to Baseball Prospectus and The Hardball Times. I am also the Magic & Performance Expert at eHow.com as well as an award-winning screenwriter.

6 Comments »

  1. JB said,

    September 4, 2008 @ 8:56 pm

    A general question:
    How helpful is it to normalize stats like WPA to leverage index in determining a player’s value? A player has little control over the situations in which he finds himself. For example, a manager may use a dominating reliever only in low leverage situations. The pitcher’s performance may be outstanding, but do to his usage will have a low WPA/LI rating.

    JB (fortheloveofthecubs.blogspot.com)

  2. Eric Seidman said,

    September 4, 2008 @ 9:16 pm

    JB, it really depends what you are looking for. Closers will usually be in the most critical situations and so Lidge will have a very high WPA but a good, not tremendous WPA/LI… which tells us that relative to the situation in which he pitched, he performed well, but his overall performance was not as good as the WPA would lead us to believe.

    It also depends on the position. For instance, it might make sense to evaluate closers using both WPA and WPA/LI since you’ll then be getting performance in situation as well as performance when all PAs count as 1 PA, unlike WPA.

    I personally like to use WPA/LI when evaluating players to learn how they fare in relation to the average player, but WPA does tell us something interesting. If I want to know which player was worth more to his team, I’ll use WPA/LI…. but if I want to incorporate some type of clutch factor in there, WPA would be good.

  3. JB said,

    September 4, 2008 @ 9:39 pm

    Makes perfect sense. Thanks!

    JB (fortheloveofthecubs.blogspot.com)

  4. Eric Seidman said,

    September 4, 2008 @ 9:56 pm

    No problem. To further clarify, the major difference is that WPA/LI, referred to as context-neutral wins, treats each plate appearance as 1 plate appearance. WPA, on the other hand, could treat a crucial 8th or 9th inning PA as 8 PAs, meaning that if Burrell hits a home run in that situation, his WPA for the game would skyrocket, whereas his WPA/LI would treat the event as a home run.

  5. Matt said,

    September 5, 2008 @ 10:58 am

    I don’t know much about Kuo, but Downs is a lefty like he is, not a righty like the article says.

  6. Eric Seidman said,

    September 5, 2008 @ 11:03 am

    Good catch, thanks. Meant to make that correction last night.

RSS feed for comments on this post

Leave a Comment


Contact Us - - Terms of Service - Privacy Policy