March 22, 1999
Pitcher Usage and Result Patterns: New York Mets
Focus on 1998 Mets starting pitchingThe use of a four-man starting rotation was firmly established in the major leagues by 1925. In the 1970s, teams started tinkering with five-man rotations, and by about 1980 all clubs had switched over to it. Last year, Mets' manager Bobby Valentine implemented a six-man rotation in late-July, which he used up until the last week of the season. Was this some type of revolutionary forward thinking on Valentine's part? As it turns out, probably not. It was more of a change forced by circumstance. Mother Nature had wreaked havoc with the Mets schedule the first six weeks of the season, forcing eight postponements. The doubleheaders required to
Al Leiter escaped last year's atrocities in Pro Player Stadium when he was acquired by the Mets after the '97 season. A regular visitor to surgery tables around the country, Leiter flourished under the decreased workload, with an incredible 25 QS+BQS in 28 starts, including a perfect six-for-six QS when he had at least six days' rest between starts. Despite Leiter's brilliant pitching, Valentine resisted temptation and never shifted him up in the rotation, until going to the whip for the last two weeks of the season. The extra off-days and a three-week respite in July for a torn patella tendon (knee) helped Leiter endure the high pitch counts that Valentine placed on his fragile left arm. Ironically, the four starts Leiter missed because of his knee injury may have both cost him a shot at the Cy Young Award and kept him in contention for it.
In 1998, Rick Reed established himself as the best #2 starter in the Senior Circuit outside of the state of Georgia, posting an excellent 81% QS+BQS. Likely due to his pinpoint control, which in turn reduces the number of pitches thrown which preserves his good health, Reed was occasionally tabbed to move up in the rotation for an injured or ineffective teammate. Even though the effects of this additional work were minimized because of the early season rainouts and the six-man rotation, Reed faded in August and September, logging only four QS in his last nine starts.
Although he was on the mound for the Mets on opening day, Bobby Jones' performance was as average as his surname. He finished the season with a 9-9 record and a Support Neutral Record of 10.6-10.6. Jones pitched 195 innings for the third time in the last four years (the other year he totaled 193). His %QS+BQS was essentially unaffected by the number of days between outings. His longest streak of quality starts was three; his longest streak without a quality start was three. Yawn... zzzz.
Thirty-three year-old Masato Yoshii left the Yakult Swallows and joined the Mets for the 1998 season. A little background on the Japanese Leagues is in order, because it sheds some light on Yoshii's results. The Japanese League season is only a week shorter than the Major League campaign, but about 25 fewer games are played. Usually, a strict five-man rotation is used, which means that a starter pitches about once a week. Because of the rainouts and Valentines' six-man rotation, the Mets' season was divided into three general usage patterns. The first pattern lasted through the end of May, and most of the starts were made on five or six days' rest due to the rainouts. The second pattern lasted until late-July and the majority of the starts in this period were on four days' rest, as cancelled games were being replayed. The six-man rotation began in late-July and was used until the last week of the season, creating the third usage pattern, where five or six days of rest again were the norm. Since Valentine always kept Yoshii in his spot in the rotation (never moving him up), Yoshii's usage patterns mimicked the team's. During the first and third usage patterns, which are similar to what Yoshii was used to in the Japanese Leagues, he pitched well, logging 15 QS+BQS in 20 starts. However, during that second usage pattern in June and July, he had only two QS+BQS in nine trips to the mound. It appears that Valentine was wise to drop Yoshii from the rotation when he changed to a four-man staff the last week of the season.
The careers of Dave Mlicki and Hideo Nomo crossed paths, presumably somewhere over Kansas, when they were traded for each other in early June. After a promising 1997 season, Mlicki opened the year as the fourth starter in the Mets' rotation, whereupon he self-destructed with only two QS in ten tries. His failure certainly wasn't due to overwork: Valentine never asked him to pitch on less than five days' rest. When Nomo arrived, he assumed Mlicki's spot in the rotation. Unfortunately, he also imitated Mlicki's performance on the hill, with seven QS in 16 starts. Valentine handled Nomo's surgically repaired elbow with care by keeping his pitch counts low, but he may have been better served by starting Nomo more frequently, as he was 5 of 6 in quality starts on four days' rest. With the Mets in a three-way battle for the wild card, Valentine dropped Nomo from the rotation in mid-September.
For Armando Reynoso and Brian Bohanon, Shea Stadium's turf was covered with four-leaf clovers in 1998. Both hurlers were free agents following the season and parlayed a handful of good outings into multi-year, big-money contracts with free-spending teams in the Mountain time zone. After missing the first three and a half months of the year with a strained elbow, Reynoso and his battered right arm emerged from the disabled list and joined the club for his annual twelve-start cameo in late-July. It was Reynoso's improved health and impressive pitching (four QS in his first five starts) that cemented Valentine's switch to a six-man rotation. Consistently working on at least five days' rest, Reynoso kept pitching fairly well until Valentine included him in the quartet selected to work the last week of the season. In the four-man rotation, Reynoso got lit up like Ted Kennedy on St. Patrick's Day, including a 48-pitch disaster on the last day of the season, which eliminated the Mets from the playoffs. Bohanon did a commendable job as the emergency starter/long reliever before he was dealt in mid-season to an even better pitcher's park (Chavez Ravine) in which to continue his trek to the pot of gold.
Even though he already had six starters, the Mets picked up Willie Blair from the D-Backs at the trading deadline. Valentine probably should have immediately plugged Blair into Nomo's slot in the rotation, but instead he primarily used Blair in long relief until mid-September. Bill Pulsipher was rewarded with what is known as a "courtesy start" for being named International League Pitcher of the Week in early June, and for working his way back from a myriad of arm injuries. At the trading deadline, the Mets further rewarded him by shipping him to Milwaukee for a lead-gloved third baseman and a case of PBR.
Bobby Valentine headed into the '98 season with a capable group of experienced starting pitchers. Because the Mets' middle relievers were erratic, Valentine often had the starters work deep into games, which led to 12 BQS, including four for each of the staff workhorses, Leiter and Reed. It also caused some high pitch counts (especially for Leiter), but that didn't prove to be a problem for a staff whose average age was 31. The early season rainouts and Valentine's unconventional six-man rotation provided plenty of recuperation time between starts, and probably was at least partially responsible for the fact that no starter suffered an arm injury. While two-thirds of the team's starts were on five or more days' rest, the rotation was actually more successful on four days' rest, with 76% QS+BQS. In fact, for the ten teams examined so far, the average starter had 56% QS+BQS on four days rest and 52% QS+BQS with at least five days rest. So, based on the success of the Mets' 1998 rotation, are other teams going to follow Valentine's lead and institute a six-man rotation? Probably not anytime soon. It's difficult enough to assemble five decent starters, and there is no clear evidence that there is a performance benefit from the additional down time between starts.
A. Leiter Days rest Reed Days rest 4 5 6+ Totals 4 5 6+ Totals Starts 10 12 6 28 Starts 11 16 4 31 QS 6 9 6 21 QS 9 9 3 21 %QS .60 .75 1.00 .75 %QS .82 .56 .75 .68 BQS 3 1 0 4 BQS 1 3 0 4 %QS+BQS 90 .83 1.00 .89 %QS+BQS .91 .75 .75 .81 Avg # Pitches 116 110 121 114 Avg # pitches 97 100 96 99 B. Jones Days rest Yoshii Days rest 3 4 5 6+ Totals 4 5 6+ Totals Starts 1 10 11 8 30 Starts 10 11 8 29 QS 0 6 7 4 17 QS 5 5 4 14 %QS .00 .60 .64 .50 .57 %QS .50 .45 .50 .48 BQS 0 1 0 0 1 BQS 0 3 0 3 %QS+BQS .00 .70 .64 .50 .60 %QS+BQS .50 .73 .50 .59 Avg # pitches 101 103 99 101 101 Avg # pitches 89 91 86 89 Nomo Days rest Reynoso Days rest 4 5 6+ Totals 4 5 6+ Totals Starts 6 6 4 16 Starts 1 5 5 11 QS 5 2 0 7 QS 0 4 3 7 %QS .83 .33 .00 .44 %QS .00 .80 .60 .64 BQS 0 0 0 0 BQS 0 0 0 0 %QS+BQS .83 .33 .00 .44 %QS+BQS .00 .80 .60 .64 Avg # pitches 102 86 81 91 Avg # pitches 48 110 102 100 Mlicki Days rest Bohanon Days rest 5 6+ Totals 2 4 6+ Totals Starts 7 3 10 Starts 1 1 2 4 QS 1 1 2 QS 0 1 1 2 %QS .14 .33 .20 %QS .00 1.00 .50 .50 BQS 0 0 0 BQS 0 0 0 0 %QS+BQS .14 .33 .20 %QS+BQS .00 1.00 .50 .50 Avg # pitches 97 93 96 Avg # pitches 85 81 76 80 Blair Days rest Pulsipher Days rest 3 4 Totals 2 Totals Starts 1 1 2 Starts 1 1 QS 1 1 2 QS 0 0 %QS 1.00 1.00 1.00 %QS .00 .00 BQS 0 0 0 BQS 0 0 %QS+BQS 1.00 1.00 1.00 %QS+BQS .00 .00 Avg # pitches 78 92 85 Avg # pitches 83 83