Hi everyone. This is my first post since September 2015 when I was filling in for John while frantically writing a dissertation and popping on to post minor league recaps written by some guy named "Michael Ernst." Whatever happened to him? (Jokes aside, my inability to keep up with the minor leagues those days turned into one of the biggest breaks CubsDen ever had.) I'm hoping this is the first of many statistically based articles on the site to help honor the man that started it. I'd like to think that some of my less pleasant personality traits of the old days have dissipated with age and reduced stress.
There's been some debate about how the Cubs offense performed in 2017 vs 2016. One line of logic holds that, since they scored more runs in 2017, they 2017 version was better.
I'd argue that, if we dig a little deeper, the 2017 offense wasn't as consistent as 2016. Yes, the 2017 team scored more runs, but a lot of that was based on some high scoring gamaes. The figure below shows an estimated distribution of their runs per game in 2016 vs 2017. (Technically, I'm cheating a bit using this method here but it is a good way to show the issue.) The horizontal axis shows the total number of runs scored in a game. The height of the line reflects how likely they were to score that many runs in 2016 or 2017.
A couple things jump out here. First of all, the line in 2017 is higher than 2016 at the low and high end of runs per game. This reflects the feast or famine nature of that offense. The 2017 version of the Cubs had more high scoring games than the 2016 version. 12.1% of the 2017 Cubs games had them score 10 runs or more as opposed to 6.2% of the 2016 Cubs. However, the 2017 team also struggled to score more often. The 2017 Cubs were shut out 10 times as opposed to 6 for the 2016 team and, in all, scored zero to three runs 40.1% of the time as opposed to 34.6% for the 2016 team.
That implies that the 2016 team was much better at scoring moderate amounts of runs -- putting up between 4 and 9 runs with some consistency. If we take the median (the midpoint of all runs scored), it was 4 for the 2016 team and 5 for the 2017 team. To give an idea of what that means, I simulated 20,000 games between the two teams where I randomly selected a regular season game for each season and used the Cubs' run total in those games as the score for my theoretical game. The 2016 team won 47% of the games, the 2017 team won 44% of the games, and the other 9% were ties. That doesn't take into consideration that runs per game were up for everyone in 2017 relative to 2016.
Trying to explain the difference gets more difficult. And obvious starting point is the lack of Dexter Fowler. Dexter was an elite leadoff man in 2016 with a .393 OBP and fourth in all of baseball with 4.40 pitches per place appearance. There are two ways this could have helped. The obvious is that 40% of the time Bryant and Rizzo were coming up with a speedy runner on on base. Singles by both or a double by either scored a run. Replacing Dexter with Schwarber or Zobrist or Rizzo or Heyward (all guys who spent time there in 2017) could easily explain some of the issues scoring runs. But it seems unlikely it explains a full run per game difference by itself and certainly doesn't explain the feast or famine nature of the offense.
Another explanation may come in the at bats that Fowler and Zobrist (4.17 PPPA) were putting up. There was talk at the time about how that exhausted the starter and got a team into middle relief. That's possibly some of it but I'd argue it also helped in letting other hitters get a good look at a starter's pitches. In key situations, there was a good chance that players had gotten a good look at the pitches the starter was going to use. And since Bryant and Rizzo followed Dexter, the best hitters on the team may have gotten an edge against the relievers, as well. Even a small change in odds in high leverage moments could lead to a big difference over a full season. (If you really want to dig into the rabbit hole, that may be some of the reason Bryant saw his RBIs drop even though his raw numbers remained similar.)
The nature of the lineup itself may also explain some of the difference. The 2016 team had Fowler, Zobrist, Rizzo, and Bryant as starters with OBPs of .385 or higher. When Contreras (.357) got regular playing time and LaStella (.407) and Coghlan (.391) come off the bench, the offense could string together walks and hits -- the key to scoring without a lot of home runs. That contributed to the "relentless" aspect of that 2016 offense. Guys were always getting on base.
In 2017, Rizzo and Bryant were the only starters with an OBP above .380. LaStella was the only bench member to do so, though Jay was close (.375) and Rene Rivera put up a very high OBP (.408) in minimal at bats against September bullpens. The 2017 team traded high OBP for high power, with Schwarber, Baez, and Happ putting up 20+ home runs with OBPs under .330. Every now and then, the team would cluster home runs and get the big run totals that 2017 showed. More often though, they got zero to one home run a game (0-4 runs).
The big question, of course, is what this means going forward. Zobrist appears to be moving into the latter phase of his career and any expectation of a repeat of 2016 appears to be optimistic to the point of foolishness. Fowler is long gone but Happ is intriguing. He is getting the lion's share of time at lead off in spring training and has suggested that he would like to be more like Fowler. If he can trade off some power to get on base, take pitches, and set up Bryant and Rizzo, he could really help the offense. Of course, that requires getting him in the lineup. With the current log jam in the outfield, that is going to mean Heyward, Schwarber, or Almora sits. For many reasons (such as attracting future free agents), sitting the highest paid player in franchise history would be a difficult choice but it might be in the best interest of the team. I can't say I expect Heyward to sit though.
Replacing Zobrist is probably going to fall on Russell. (The talk of Russell for Machado earlier this winter almost certainly took this into account.) He's going to have to turn into the guy scouts thought he could be with a good batting eye and solid batting average to go with the defense. I'm hopeful but not optimistic that can happen. Almora and Heyward are also possibilities but I'm less optimistic on both.
Schwarber becoming the unholy monster who terrified pitchers in the 2016 playoffs would also help lengthen the lineup. While this is possible I don't think it's fair to expect it.
The obvious conclusion is that the Cubs need to get back to having good at bats up and down the lineup. Happ morphing into Fowler would be a big help. I would imagine Chili Davis has been tasked with improving at bats, as well. Either way, a return to the monster 2016 offense is going to require a lot of things to break the Cubs' way. This team is going to lean a lot more on the rebuilt pitching staff.