This past Sunday the Nationals deployed an unusual strategy with their pitching. After Jake Irvin pitched five and a third strong innings, Mitchell Parker pitched a full three and two-third innings to close out the game for the Nationals and secure a 3–2 win against the high-powered Mets offense. Even though Parker is finishing the season as a reliever, the deployment of Irvin and Parker in this manner makes sense as a consistent strategy for the Nationals moving forward.
The Nationals should intentionally plan on using a “piggyback” pitching strategy to begin next season. Instead of a conventional strategy that relies on five starting pitchers to pitch about five innings to begin a game and a set of eight relievers to cover the remaining innings, the Nationals should designate a couple days each week where two starters will piggyback on top of each other. Instead of relying on one starter and then multiple relievers, the manager will pair two starting pitchers together to cover most of the game by pitching three to five innings each. The strategy would ideally reduce pitcher workloads and increase effectiveness.
The starting pitcher’s role in baseball has dramatically changed in recent decades. From the 1950s to the 1970s, starters averaged 6.5 innings per start. Managers could rely on their pitchers to eat innings and did not need to rely on a deep bullpen to get them through games. Starters now only average about 5.1 innings per start. Over the course of a 162-game season, this 1.4 inning drop amounts to 226.8 more innings for the bullpen to cover. Moreover, since starters are only pitching about half the game, managers often ask their relievers to work multiple days in a row to finish games. The combination of having to pitch more innings and having fewer days off undoubtedly contributes to reliever fatigue and injury.
A piggyback strategy could potentially alleviate this problem. By deploying two bulk pitchers in these games, the bullpen will have fewer innings to pitch in those games, potentially reducing relievers' overall workload and giving them more days off to recover. However, there is a drawback with this strategy. The team is pairing two pitchers together who will only pitch once every five games. That means the team functionally has an extra starting pitcher, leaving room for fewer relievers on the roster. For instance, if a team decided to use two piggyback games through a traditional five-man turn through the rotation, they would need to carry three starters and four piggyback pitchers, leaving space for only six relievers. If the strategy is not deployed correctly, there will be fewer men in the bullpen, meaning the remaining relievers will have to cover more innings.
To discover whether the piggyback strategy would harm or help relievers, I looked at the workload per reliever in a five-man turn of the rotation. Throughout this period, a team will need to get a maximum of 27 outs per game, for a total of 135 outs. Starters, s, who average 15.3 outs per game, can cover a large portion of these outs. In piggyback games, p, the two combined pitchers will record a certain number of outs, y, and the number of relievers is eight, the traditional number of relievers, minus the one extra pitcher that is used for a piggyback game. The equation below expresses this:
Reliever Workload (p)= {135 - [(5 - p) · 15.3 + y · p]} / {8 - p}
With this equation in mind, I determined the number of outs that both piggyback relievers need to cover for the strategy to reduce pitcher workload. Finding the first derivative of the equation with regard to p allowed me to find the value of y where the piggyback strategy goes from increasing individual reliever workload to decreasing reliever workload. The first derivative is written below:
Reliever Workload ' (p)= {180.8 - 8y} / {(8 - p)^2}
By setting this equation equal to zero, I found the point where the rate of change in workload per reliever goes from increasing to decreasing. At this point, y equals 22.6. This means the combination piggyback starters need to cover at least 22.6 outs in their outings to decrease reliever workload; each pitcher would need to record about 11 outs, or three and two-third innings. The table below shows the workload for relievers for differing values of p, the number of games with a pair of piggyback pitchers, and y, the number of combined outs covered by the two pitchers. For instance, if a team had four conventional starters and one set of piggyback pitchers who combined to record 21 outs, the seven relievers in the bullpen would be responsible for recording 7.54 outs every five games.
Piggybacks = p | y = 21 | y = 22 | y = 23 | y = 24 |
|---|---|---|---|---|
0 | 7.31 | 7.31 | 7.31 | 7.31 |
1 | 7.54 | 7.40 | 7.26 | 7.11 |
2 | 7.85 | 7.52 | 7.18 | 6.85 |
3 | 8.28 | 7.68 | 7.08 | 6.48 |
4 | 8.93 | 7.93 | 6.93 | 5.93 |
When piggyback relievers get less than 22.6 outs, more piggyback games only lead to more outs for each reliever to cover. When piggyback relievers get more than 22.6 outs, more of these games only reduce the overall workload for relievers. Against the Mets, Irvin and Parker, who have been two of the least effective pitchers in baseball this year, combined for 27 outs. If these two can do it, the Nationals can conceivably implement this strategy properly and reduce their bullpen workload over the course of a season.
That being said, the Nationals should not go all in on the piggyback strategy. They should keep Mackenzie Gore and a few other starters on traditional pitching schedules. Math cannot account for the unpredictability of baseball, and it would be unwise for the Nats to go all in on this strategy and have all of their games be piggyback games.
The strategy could also reduce the workload for the Nationals’ younger and promising arms. By working in piggyback roles instead of traditional starting roles, pitchers like Irvin, Parker, Cade Cavalli, Brad Lord, and Josiah Gray returning from injury could be successfully limited to only around four innings per outing instead of five innings. Over the course of a full season, this can reduce their workload by 20 to 30 innings. As the Nationals rebuild, they can slowly build up their starters and not overtax them with a full season's workload.
Nationals pitchers could also be more effective under this strategy. The deeper pitchers go into games, the less effective they become. They become more fatigued, and hitters are more familiar with pitchers the second and third time through the order. By having two pitchers pitch only four innings in a piggyback instead of five to six innings in a conventional start, these bulk pitchers will hopefully avoid the penalties of pitching deeper into games.
The Nationals are under new leadership and rebuilding. Now is the time to experiment, and the piggyback strategy could give the team an advantage in run prevention throughout the season. Testing a piggyback strategy for one to two rotation spots could reduce fatigue and increase effectiveness for all pitchers on the staff.