Your 2018 Giants: Offense Greatly Improved

After the Giants early offensive struggles (albeit, against one of the top rotations in the majors in one of the more extreme pitcher's parks in the majors), many were shouting, "The Giants Offense sucks!  The offense sucks!"  I disagree.

ogc thoughts

And this post won't really assuage them, as they probably don't care to know that adding two good hitters (Longoria and McCutchen), and having another two return (Belt and Crawford), makes for a really good lineup.

I like using the Lineup Calculator.  Some saber long ago did a linear regression of lineups by lineup position, with OBP and SLG (during the so-called steroids era), and I've been using it for over the past ten years, and when I test it against the majors overall numbers, it still comes out pretty close.  Apparently each lineup position contributes a certain percentage of runs depending on the OBP and SLG that hitter attains.   So I believe in the numbers the calculator generates.

Giants Lineup:  Calculated Using ZiPS

For this lineup, which I had done before Spring Training, soon after ZiPS' projection was released, the Giants averaged 4.73 runs scored with a lineup of (that's the way I would bat them if I had to stick with one lineup):

1. Panik
2. Belt
3. Posey
4. McCutchen
5. Longoria
6. Pence
7. Crawford
8. Duggar

In 2017, the average NL team averaged 4.58 runs scored per game.  4.73 runs scored would have ranked the Giants 7th in 2017, basically in the middle, but slightly above average.  This used overall projections for 2018.  Using Fangraph's Depth Chart projections by playing position, and inserting unoptimized into the above lineup, the team still averages 4.53 runs per game (for example, Tomlinson playing 2B, and batting leadoff as above or Blanco playing RF and batting clean-up, very unoptimized), which is basically the NL average.

Giants Lineup:  Calculated Using 3 Year vs. LHP

Now I'm going to look at the lineup using each players' 3 year (2015-2017)  batting line  against LHP:

1. Jackson:        .360/.453/.813
2. Panik:           .335/.342/.677
3. McCutchen:  .394/.545/.939
4. Posey:           .407/.517/.926
5. Longoria:      .333/.462/.795
6. Pence:           .326/.441/.767
7. Belt:              .355/.454/.809
8. Crawford:     .308/.391/.699

Two things pop out from this.  First, Panik should not be batting second based on his historic 3- year numbers.  Second, Belt should not be batting 7th based on his 3-year batting line.  If anything, they should be swapped, and then look at the OBP for the first four hitters!

Still, even with the obviously bad mis-alignment, this lineup averages 4.93 runs scored per game, which would rank 5th in the NL last season.  If the Giants had the same runs allowed as last season's disaster, 4.79 runs allowed, with this offense, it would win at a 83 win rate.  But it only needs to reach 4.55 runs allowed to reach 87 win rate (Colorado's win total for second Wild Card spot) and 4.20 runs allowed to reach 93 win rate (D-backs win total for first Wild Card spot).  In 2016, the Giants allowed 3.90 runs on average, and that would result in a 98 win rate.  The average runs allowed rate for 2016-2017 was 4.34, and that would result in a 90 win rate.

Giants Lineup:  Calculated Using 3 Year vs. RHP

1. Panik:             .348/.445/.793
2. Belt:               .377/.483/.860
3. McCutchen:   .360/.447/.807
4. Posey:            .369/.432/.801
5. Longoria:       .315/.461/.776
6. Crawford:      .330/.449/.779
7. Pence:            .336/.421/.757
8. Blanco:          .349/.374/.723

Here Belt is where he belongs, batting second in the lineup, no matter who is pitching, and if anything, he should be batting 4th (but showing how little changing the lineup means to overall run production, switching Belt and Posey would only add one run per season, and thus average the same per game;  however, swapping Panik and Belt vs. LHP would reap an addition four runs, raising to 4.96 run scored per game average).  This seems actually pretty ideal, as research has found that placing your best hitter batting 2nd is good for the offense, which is good because roughly 75% of all AB's are against RHP, overall.

Though not as potent, this lineup still averages 4.75 runs scored per game, which would rank tied for 6th in the NL last season, with the Dodgers.   The implication of this is that since that's LA's overall batting line, and both lineups above are at or above LA's and thus is comparable in strength with the Dodger's offense (assuming some downward effect of using the bench, as what happened above with the ZiPS projections of starters vs. positional totals. 

If the Giants had the same runs allowed as last season's disaster, 4.79 runs allowed, with this offense, it would win at a 80 win rate (only time with this lineup analysis where the Giants end up under .500).  Presumably, the Giants have better pitching and fielding than they did in 2017.  But the team only needs to reach 4.40 runs allowed to reach 87 win rate (Colorado's win total for second Wild Card spot) and 4.05 runs allowed to reach 93 win rate (D-backs win total for first Wild Card spot).  In 2016, the Giants allowed 3.90 runs on average, and that would result in a 95 win rate.  The average runs allowed rate for 2016-2017 was 4.34, and that would result in a 88 win rate.

Offense Good Enough to Win Wild Card Spot

From the above, the Giants look like they are going to easily be an over .500 team, assuming health and production.  And look like they will probably earn a wild card spot, and perhaps even the NL West Division Title.

Obviously, the Giant starting pitching should be better than 2017, as there is no obviously bad starter like Cain or Matt Moore were, both Stratton and Blach look like they can keep things under 5.00 easily (Blach, for example, had 4.78 ERA in 24 starts and 10 relief appearances, 163.2 IP), plus Samardzija was hurt by BABIP in 2017, his peripherals were much better than his ERA would suggest.   But obviously 2018 should not be as good as 2016, as both Bumgarner and Cueto were aces to the core, though one should also noted that Moore and Cain in 2017 did a pretty good imitation of Peavy and Cain in 2018.  Plus the bullpen looks better than 2017 as well.  Overall, one should be able to easily say that the 2018 Giants should be somewhere in-between, and perhaps closer to 2016 than 2017.

If you say that the 2018 is exactly in-between, that would be 4.34 runs allowed, and works out to roughly 89-90 wins, which should be enough to win the second Wild Card spot.  If the Giants average 4.12 runs allowed, that gets them to 94 wins, and the first Wild Card, and potentially the NL West Division title, as the Dodger's starters averaged 0.52 runs lower ERA relative to their FIP.   Adding that many runs to the 2017 Dodgers would push their Pythagorean Wins to 92 wins.

How 2018 Giants Compares with 2016 Giants

How might the 2018 Giants be closer to the 2016 Giants, especially considering Bumgarner is out to June, sometime?  First off, as mentioned above, neither Peavy and Cain were that good in 2016, both in the mid-5 ERA range.  Second, Cueto looks like he's back to 2016 form.  Third, Blach did pretty well last year, much better than either Peavy/Cain in 2016, lessening the need for the other starters to match 2016 performances.  Fourth, Stratton has demonstrated a great curveball and, even better, good results, was killer for 9 starts at the end of 2017, and his first start of 2018 would have been a 4 PQS DOM start (old definition; 3 newPQS DEC or decent start, but even that is actually pretty good). 

Holland is the unknown factor here.  It depends on how long the Giants allows him to stink up the rotation before he is moved into the long relief role.  As I noted in recent analysis, Holland has actually had a very good ERA for good parts of the start of the past few seasons, but his arm strength would eventually give way and he's horrible.  Kind of like Blach last season, he was decent for 19 starts (122.1 IP in those starts, 128 total IP including relief) in 2017, after being placed in the rotation, with a 4.12 ERA/3.79 FIP, and then his arm gave out and he had 5 bad starts, 7.90 ERA. 

Holland's ERA was okay to the middle of 2017, and probably will give way again sometime around the All-Star Break.  I'm hoping that either he gets made the long reliever when Bumgarner comes back, or the Giants recognize the fatigue and put Blach back in, or better maybe have Holland and Blach take turns in the rotation, pitching every 10-11 days, and being the long reliever for the other in-between.   But even if he's bad, as long as he's in the same ballpark as Peavy/Cain in 2016, we are okay as long as the other starters are doing as well as expected. 

In addition, the bullpen had a 3.65 ERA in 2016, only 4.34 ERA in 2017, and as I noted in my bullpen post, the bullpen is looking like it can be in the low 3 ERA range, and thus matching the 3.65 ERA of 2016 looks easy as long as they are healthy enough (plus, Moronta, Johnson, and Osich look like they might be okay, Okert too).  The main thing is the bulk of the reliever innings will go to Strickland, Watson, Dyson, Gearrin, and probably Smith, so the bullpen ERA should be pretty good (their 3.65 ERA in 2016 would have ranked second in 2017, and these guys overall should beat that, while others will bring back towards 2016 average).

Finally, the goal here is to be closer to 2016 than 2017, not to beat 2016.  The above suggest that the 2018 Giants are closer to 2016 than 2017, and in some ways, could be better than 2016.  Even exactly in between would bring the team to the upper 80's in wins.  I think the Giants are in good shape for the playoffs because of their offense helping to bring the pitching over the hump, and then hopefully all the key pitchers are healthy and performing at year's end, and we should have a good chance in the playoffs.

Belt (and Giants) Kershawed:  How Giants Can Beat Kershaw

A final addendum to the above about Belt.  As noted, Belt's actually a pretty good hitter vs. LHP, but I learned something new over Twitter in the past week:  we all know that he sucks against Kershaw but never really understood what that means overall.  Taking Belt's 3-year vs. LHP batting stats, and taking out his putrid results against Kershaw (oh-fer-24, with a HBP; yeah, it's that bad), I ended up with Belt hitting vs. LHP not named Kershaw of .273/.372/.481/.853.  His OPS vs. RHP is .860!!!

So if Bochy had just sat Belt for every Kershaw start (and, say, bat Hundley, lifetime .260/.315/.300/.615, which is not good, but way better than the zero Belt has, and play him at 1B, so we got Posey's defense behind the plate, as we need every edge we can get), Belt has been hitting as well vs. LHP as he did against RHP the past three years, and .850+ OPS is very  good.

And this could be significant in Kershaw starts, with Belt being such an automatic out, maybe that is why the Giants have lost all those games to him during his career.   Dropping Belt's bat out of the lineup, and replacing it with zero, ends up costing the Giants around 1 run scored per Kershaw start.

If you take his 22-10 record (.688 winning percentage) and his 1.87 runs allowed average per 9 inning games, you get the Dodgers scoring 2.90 runs per game on average in his starts against the Giants using Pythagorean.   Apparently our best pitcher was up against Kershaw a lot (and that would be Bumgarner). 

Adding in Hundley's career numbers against Kershaw into the lineup in the 7th spot adds 0.83 runs over what Belt was providing (and assuming Belt was in every Kershaw start vs. the Giants; he was roughly in only 7 of the last 15 over the past 3 years, though), that would have changed Kershaw's record from 22-10 (.688) to 17-15 (.532).   Adjusting for half time usage would drop it only to 19-13 or so, though.

And the Giants apparently noticed that, as well, because Belt did not start one of the Kershaw's starts in 2017, missing all five (of course, he was injured in the latter two months, and probably miss 2-3 there, but still).  And that didn't help the Giants all that much, as he went 4-1 against them in 2017 (though I would note here that the Giants offense was crippled by a lot of poor performances besides Belt, in 2017).

Still, I think the concept holds, for the most part.  Just give Belt a rest on the day that Kershaw starts, and his numbers are great, whether LHP or RHP.

Though, oddly, he was not shut out in 2011-12, just not very good, still worse than Hundley.  And actually got a hit this year, and has his best OPS, .667, against him in any season (albeit, 1 for 3);  if he can do at least that much each time, it is better to have him in there than Hundley, it is just the zeros that really hurt the Giants offense.   But for the most part, Kershaw has had a ton of ownage on Belt over his career, and it is better for the Giants to give Belt a rest on Kershaw starts.

Final note:  I wouldn't sign anyone just for ownage on one player, but McCutchen has a lifetime .333/.375/.533/.908 batting line against Kershaw in 32 PA.  Just noting.