Is the Rays’ Experiment Working?

Well here we are at the “midway” point of the season (we’re actually 60% through). On Wednesday, after all of the All-Star Game festivities have ended, we will enter into the darkest two days of the summer. For two days, there will be no football, no basketball, no hockey, not even any more World Cup soccer. And, of course, most importantly there will be no baseball.

But all is not lost because this gives everyone ample time to make second “half” predictions. The All-Star break can be a turning point for many teams, and many hopeful fans will wonder if their team can pick it up after these four days without any official games. For the Tampa Bay Rays, though, that turning point came much earlier in the season.

Way back in the middle of May, the Rays were in a series against the Angels. On the 19th, they won the third game of the series 5 – 3. If we go back to that game on MLB.com and click on the summary, this is the headline we find:

At first, nothing seems suspect, except maybe the Rays winning 6 straight games. Any baseball fan acquainted with the Rays’ roster might see that headline and reasonably believe that Ryan Yarbrough started the game, had a solid outing, and then passed the ball onto Sergio Romo, who pitched excellently in relief.

That baseball fan would be wrong.

On May 19th, Sergio Romo started the first game of his career, after 588 relief appearances in the majors. He pitched for one inning, struck out the side, then handed the game off to Yarbrough, who pitched the next 6 1/3 innings. This marked the beginning of the Rays’ turning point. This has been a regular occurrence since that mid-May night, not just with Romo, but with several of their relievers. The outlandish decision to start a relief pitcher who had never seen the mound in the first inning was demonstrative of the grand experiment brewing in Tampa Bay.

The logic of using what the Rays are calling “openers” (i.e. relief pitchers throwing for the first couple innings before being replaced by the true “starters”) is as follows:

• The manager can now play matchup games in the first inning, instead of waiting until late in the game.
• Pitchers who are able to dominate for an inning or two can pitch in the first inning to lineups that are particularly top heavy (this was the case with Romo against the Angels).
• The “starters” will typically not face a lineup more than two times before being replaced late in the game by a closer or set-up man. This doesn’t allow the hitters to get comfortable with one pitcher.

Some old-school baseball thinkers were very quick to assert that this “reliever game” experiment would never work, and that it will surely fail. (One such thinker is Dennis Eckersley, which may not surprise some readers given his past opinions.)

But here at Rising Fastball we don’t have to guess at the effectiveness of what the Rays are doing. We have the data! So let’s see what it tells us.

The Numbers

Here are the rankings of the 15 teams with the best FIPs from the beginning of the season to May 19th, the date of Sergio Romo’s strange one-inning appearance. Also included is the team’s ERA, WHIP, and Opponent OPS:

The Rays are, unsurprisingly, at the bottom. This would put them exactly in the middle of all of Major League Baseball in FIP, nothing to write home about. Note that the rankings would be slightly different if we ordered it by ERA, but I believe that FIP is a better metric for our purposes since it doesn’t concern us with defensive performance. However, the Rays would also be in 15th place on this chart in ERA anyway.

To be fair, a 4.12 team FIP isn’t bad. The league average FIP is usually somewhere around 4.20 (*cue snickering from high school students*) so the Rays are still above average here. However, having the 15th best pitching staff doesn’t exactly scream playoff material.

Here are the rankings of the same metrics, but from May 19th onward:

Now that’s more like it! An 11 place increase. And they’re in good company. It’s never bad to see your team amongst the likes of the Astros, Dodgers, Red Sox, and Yankees. In fact, if we ordered this ranking by any of the other three metrics, the Rays would be in first place. The picture that all four metrics paint together is even more important. There was a significant decrease in each area, showing overall improvement, not just a strange statistical fluke:

That’s over an 11% decrease in every category. This may not sound like much, but clearly it’s enough to push the Rays into a field of playoff teams.

But there’s one problem: the Rays aren’t a playoff team.

No October Run in Sight

The Rays are currently 49 – 47. Being above .500 is a good start to having a decent team, but after all of this praise to the Rays’ pitching staff and their wacky thinking, why are they so far away from the postseason?

The most obvious answer is their division. They play in the same division as two of the best teams in baseball, who seem to plausibly be on their way to 100+ win seasons. There’s no chance that the Rays will sniff first place in the AL East. They’re improving, but they’re not that good yet. And the fact that the Red Sox and Yankees are battling for first place means that the rest of the league is essentially competing for a singular wild-card spot. Right now the Rays are 9 games back in the AL wild-card race. If the Mariners fade, which they probably will, this could leave them in a tight race with the Rays, Angels, and A’s for the open spot.

A deeper answer, though, comes from their run differential. As good as the Rays’ pitching has become, they still only have a +20 run differential. The graph below helps illustrate this problem. The red line represents how the average number of runs allowed per game has changed for the Rays as the season has gone on. The blue line, on the other hand, represents the average number of runs scored per game:

The dashed line shows the week that the remarkable Sergio Romo start occurred. Impressively, the Rays’ average runs allowed per game since the May 19th Romo game is over one run lower (3.04) than it was before that game (4.38). Their runs allowed per game has essentially trended steadily downward since the season began. Unfortunately, so has their runs scored per game.

As the Rays’ pitching has improved, their offense has gotten almost equally bad. They are allowing fewer runs, but are not getting the run support to make a difference. Until the last couple of weeks, their average runs scored per game was lower than their average runs allowed per game all season.

This is simply not a formula for a good run differential, and thus not a formula for a playoff-bound winning percentage. It does not appear to be in the cards for the Rays this year, but their new way of treating their pitching staff could have a marked impact on the game of baseball.

Is This All for the Better?

The Los Angeles Dodgers are often credited with the line of thinking that a starting pitcher should not face a lineup three times. If they do, the hitters will have gotten too comfortable, and they’ll be able to predict what the pitcher will throw too easily. Whether or not this idea truly originated in LA, the Rays have taken it to heart with their new approach.

To front offices and analytics departments, this makes perfect sense. In fact, if a team could feasibly throw a different pitcher on the mound for every hitter of every inning, they probably would. No two hitters would get the same look at a pitcher. No information could be passed along to the team from the previous hitter. However, to the fans, this would be detrimental.

In the ridiculous hypothetical above, and in the Rays’ case, there are simply more pitching changes. This means significantly longer games. This means more commercial breaks. This means more watching guys stand around an infield. This means more fans just milling around the concourse looking for a beer under $11. This means less viewers, and possibly less fans of baseball as a whole.

In the age of trying to increase the “pace of play” of baseball, it seems that what the Rays are doing would infuriate the commissioner’s office. One idea that’s been tossed around to save baseball viewership is limiting the number of pitchers a team is allowed to carry on a roster. This would force teams to take one pitcher deeper into a game, and change pitchers less often. I don’t think that’s such a radical idea. It could potentially help the length of games get under control, without changing the nature of the game itself. If somehow the players’ union allowed that to happen (and that’s a big if) would the Rays’ new model of pitching be able to survive?

I’m not saying the Rays are ruining baseball. Without the Rays, I’m not sure how retired Red Sox and Yankees fans in Florida would ever be able to see their team play. And I’m not saying they should change their philosophy. As long as the rules allow it, they should do whatever they can to get an edge and find an inefficiency in the game. Luckily for them, they play in a market that will allow them to make radical decisions like this without much real backlash. For the sake of their fans, I hope it works.

Will this idea catch on with more teams in the league? I’d be lying if I said I wasn’t worried about the impact that would have on the length of the game. But, as the saying goes, stranger things have happened.