clock menu more-arrow no yes mobile

Filed under:

If the Colorado Rockies had an excellent bullpen, they would still be a bad team

The Colorado Rockies bullpen is baaaaad, but it hasn't had as much of a negative impact on their season as you might think.

Doug Pensinger/Getty Images

Recently, Patrick Saunders of the Denver Post wrote that the Rockies would do well to address the bullpen problem if they want to succeed in the near future. Purple Row's Eric Garcia McKinley responded and claimed that the bullpen should be the last, rather than the first, priority of a rebuilding team. These articles led to some Twitter exchanges, the core of which return to the ultimate value of the bullpen. The Twitter conversation stimulated a discussion among Purple Row's staff. This led us consider thought experiments and research that might help us determine just how many games the Rockies' bullpen has cost the team in 2015.

We settled on two basic questions. The first is: what would the Rockies' record look like if it had a perfect bullpen—a bullpen with a 0.00 ERA? The second is more grounded in reality: how many losses has the bullpen been responsible for this season?

Let's start with the alternate world—the perfect world. The criteria for turning losses into wins for this one is simple. We identified games in which the Rockies lost and the starting pitcher allowed fewer runs than the opposition at the time he departed. Tie games don't pose a problem because the perfect bullpen would just keep chugging along until the Rockies scored a run. Such a scenario has taken place in 28 of the Rockies 69 (nice) losses.

Assuming a bullpen with a collective 0.00 ERA, the Rockies would have a 76-41 record. That's a great record! But it's still within the realm of possibility, whereas we might expect such an impossible feat to lead something similarly unattainable. The Cardinals are 76-43 right now. That means that the Rockies would have a realistically great record if their bullpen were impossibly good. Since 1980, roughly the advent of the modern bullpen, the best single-season bullpen ERA was posted by the 1981 New York Yankees, who had a 2.26 mark. The 2015 St. Louis Cardinals aren't far behind with a 2.22 ERA so far this season.

But we don't have to maintain the 0.00 ERA fantasy. What if the bullpen had held on to each win in which it entered the game with a lead, but it lost each game in which it entered during a tie? That way, we don't credit the offense for something they didn't do. That accounts for 12 of the 28 losses noted above. The result here would be a record of 64-53. That's also a good record! It would place the Rockies in the thick of the Wild Card race. It wouldn't demand the bullpen have a perfect record in terms of earned runs allowed, but it would demand perfection with regard to holding every lead. Something like this might be possible, but it would require both an outstanding squad and a mess of luck.


Now, let's turn from the macro to the micro and find out just how many losses the 2015 bullpen has been responsible for. We used the following criteria to find games in which the bullpen was "responsible" for a loss:

1. The starting pitcher lasted a minimum of 5⅓ innings, so that the bullpen isn't tasked with picking up 4 plus innings.

2. The offense scored a minimum of three runs.

3. Walt Weiss did not mismanage the bullpen and set them up to fail (subjective, but it highlights that context matters in bullpen usage).

Now that we've taken care of that, let's start the process of elimination. Using the starter innings pitched criteria, we eliminate 52 games, leaving us with 65 remaining. Of those 65 games, the Rockies won 35, so we're down to 30. Moving on to our second criteria, we can eliminate 18 of our remaining 30 games due to the Rockies scoring two runs or fewer. Just like that, we're down to only 12 games. More of these can be eliminated as well, but we'll have to look at them a little more closely.

Games in which the starting pitcher allowed more runs than the Rockies scored

April 18 - Rockies 3, Dodgers 6 (Jordan Lyles allowed four runs)

May 18 - Phillies 4,  Rockies 3 (Jordan Lyles allowed four runs)

June 1 - Dodgers 11, Rockies 4 (Kyle Kendrick allowed six runs)

June 28 - Rockies 3, Giants 6 (Kyle Kendrick allowed five runs)

There goes four more games, taking us down to just eight games that the bullpen might be responsible for the Rockies losing. Even if the bullpen really was at fault for all eight of these losses and even if a better bullpen would've guaranteed a Rockies win in all eight of these games, the club would still be just 56-61 on the season and not in playoff contention. Let's dive deeper into these eight games, though, to see just how many the bullpen has really cost the team.

April 12 - Chicago Cubs 6, Colorado Rockies 5

Jordan Lyles started and pitched well for the Rockies in this one, allowing three runs (one earned) in six innings of work. The offense was solid as well as they scored five runs on 11 hits, though they also committed four errors in the game (none of which led to unearned runs against the bullpen). Ultimately, the game was decided on the three runs allowed by then-closer LaTroy Hawkins in the ninth inning.

Verdict: The bullpen blew it

May 22 - San Francisco Giants 11, Colorado Rockies 8

This one was a classic Coors Field slugfest with a total of 19 runs scored on 27 hits. A mediocre Kyle Kendrick got through 6⅓ innings and handed the ball off with the Rockies trailing 3-0 (though a fourth run would later score and be charged to him). The bullpen was pretty bad in this one, allowing seven runs in just 2⅔ innings, but they're not entirely to blame. When they took over in the seventh inning, the Rockies were losing, had managed just one hit (a single) in the first six innings, and had only a 10 percent win probability. Had the game been going a different way, manager Walt Weiss also may have chosen to go to a more experienced reliever than Ken Roberts, which could have produced better results. While the bullpen certainly exacerbated the situation, I have a hard time saying they were responsible for the loss here.

Verdict: Not the bullpen's fault

June 2 - Los Angeles Dodgers 9, Colorado Rockies 8

We all remember this one. The Alex Guerrero grand slam game. David Hale started the game and allowed four runs in six innings and the bullpen took over with a 5-4 lead. That lead had expanded to 8-5 in the ninth inning before Guerrero happened. His grand slam off Rafael Betancourt gave the Dodgers a 9-8 lead that would eventually be the final score. This one is debatable, but we think this is a case of bullpen mismanagement. Recall that this was game two of a doubleheader and Betancourt had already pitched in game one. That may have been okay 10 years ago, but it's probably not smart to try to use a 40-year old arm twice in one day. The pitcher that relieved Betancourt -- Tommy Kahnle -- had not been used that day and was likely a better option.

Verdict: Bullpen mismanagement

June 27 - Colorado Rockies 5, San Francisco Giants 7

This is a tough one. Chris Rusin was the starting pitcher in the game and he allowed five runs (just two earned) in 6⅓ innings. A costly error from Wilin Rosario led to three unearned runs scoring in the third inning and helped the Giants to tie the game, 3-3. This is another case of what we believe to be poor bullpen management, as Betancourt had just returned from the disabled list with vertigo symptoms and had already pitched an inning the day before. He entered the game and allowed two runs that ended up deciding the game. As a general rule, it's wise to avoid using fresh-off-the-disabled-list pitchers on back-to-back days, but Betancourt was the first guy Weiss went to in this game. Again, we chalk this one up to poor bullpen management.

Verdict: Bullpen mismanagement

July 3 - Colorado Rockies 3, Arizona Diamondbacks 4

The Rockies got out to a quick start in this game with three runs in the top of the first. They then proceeded to not score for the rest of the game, only managing two hits -- a single and a bunt single -- over the nine innings that followed. Kendrick was the starter in this game and allowed three runs in 5⅔ innings. When the bullpen took over with the game tied 3-3, a combination of Justin Miller, Kahnle, Christian Friedrich, and Scott Oberg combined to throw 3⅓ scoreless innings to send the game to extra innings. Yohan Flande would eventually allow a walk-off sacrifice fly in the bottom of the 10th, but all in all the bullpen allowed just one run in 3⅔ innings. That's a very solid 2.45 ERA, which makes it hard to say that that they're responsible for the loss here. The lack of offense after the first inning is what did the Rockies in.

Verdict: Lack of offense

July 18 - Colorado Rockies 4, San Diego Padres 5

After Chad Bettis' quality start in which he allowed three runs in six innings, the bullpen took over in yet another game that was tied 3-3. Friedrich walked the only batter he faced before Kahnle relieved him, and he promptly allowed a 2-run home run to Clint Barmes that put the Rockies down 5-3. The offense scored four runs on eight hits against Tyson Ross at Petco Park, which isn't superb, but it's certainly adequate given the circumstances. With a quality start (even though it was a minimalist's quality start) and a solid performance from the offense, we have to say the bullpen is at least mostly responsible for this loss.

Verdict: The bullpen blew it

July 22 - Texas Rangers 10, Colorado Rockies 8

Here we have yet another game that was clearly played at Coors Field, with 18 combined runs scored on 19 hits. Jorge De La Rosa allowed six hits in the first six innings before the bullpen took over in the top of the seventh with the Rockies trailing 6-4. This game bears a striking similarity to the game exactly two months before it in that the bullpen did not pitch well -- they allowed four earned runs in three innings of work -- but they also were not entirely to blame. De La Rosa clearly wasn't at his best in this game, and when the bullpen got involved, the Rockies had a win probability of just 20 percent. For those reasons, we can't lay all of the blame on the bullpen here.

Verdict: The bullpen and the starting pitching share the blame

August 14 - San Diego Padres 9, Colorado Rockies 5

Our most recent example was not a pretty one. After a serviceable start from Flande in which he allowed four earned runs in six innings, the bullpen imploded by allowing five runs -- all earned -- in just three innings. The game was tied up, 4-4, when the bullpen took over, so it wasn't like some of the previous games when poor performances from the offense and/or starting pitching had already put the team in a bad spot. This is a game in which the bullpen was likely responsible for the loss.

Verdict: The bullpen blew it

The final tally here comes to three games blown by the bullpen, two blown by bullpen mismanagement, two games where the team was largely out of it and the bullpen's performance wasn't very relevant, and one game with a lack of offense bringing the team down. Do the Rockies have a good bullpen? Absolutely not, but in spite of that, there are only three games from the entire season where we can look at them and say that a loss was primarily their fault. It would be great to see improvement from the bullpen, but it won't matter very much if the Rockies don't improve other areas of their team first.

And, in fact, these thought experiments and game-by-game investigations illustrate that multiple factors -- the offense, starting pitching, and managerial decisions -- interact with and have consequences for the bullpen. Maybe the Rockies' primary objective for the future shouldn't be isolating one or another aspect of the team to take priority, but in making each component of the roster and the coaching staff complementary.