2016 Giants Rotation: A Game Score Breakdown

Last winter, I discovered and fell in love with Bill James’ Game Score metric. In hopes that you’ve read some of my previous posts on the subject, I’m going to spare the basics this time around and jump right into my analysis of the Giants rotation this year.

First, a few quick things to note. With the immense help of a Play Index subscription via Baseball-Reference.com, I spent the majority of the 2016 season not only tracking Game Scores on a day-to-day basis, but also diving into the all-time greats (both individual seasons and careers), comparing different generations and adjusting what I had previously thought to be the benchmark scores for Game Score decisions (W-L-ND). I wanted to provide insights to a few of those adjustments here.

The cheat sheet:

In the current MLB Game Score era (roughly 2012-present), the pitching decisions are as follows.

National League

Win = 58 GmSc or higher | Loss = 46 GmSc or lower | ND = GmSc 47-57

American League

W = 56 or higher | L = 43 or lower | ND = 44-54

I will frequently mention Game Score averages and advantages (ADV) alongside pitcher records. As a baseline, an Avg GmSc of 50 is average; 60 is generally All-Star caliber, while anything under 47 should be considered replacement level. Advantage scores are fairly self-explanatory, but I consider a score of +100 above average, +200 very good, +300 great (All-Star), and +400 or better as Cy Young worthy. Anything below -50 is considered replacement level.

In 2016, Max Scherzer (+459 ADV) & Justin Verlander (+394) were the respective league Advantage leaders. Patrick Corbin (-180) and James Shields (-196) were worst among the two circuits. Essentially, the difference between the best and worst starter in each league was 600 Advantage points in 2016. It has become one of my favorite metrics to work with, and proves quite well as a ranking tool.

I’ll spare you the dirty details, but just know that GmSc Wins average out to about a .750 team win% league wide; GmSc Losses amount to a .250 team win%, and as you’d expect, GmSc No Decisions correlate to a .500 win%. This holds true for both leagues.

National League SP earned a cumulative Game Score record of 925-899-604 this year (.381), with an average score of 50.9. AL starters amassed a 1025-855-548 (.422) record, with an average score of 50.3. While it might seem odd that AL pitchers had such a higher GmSc Win% despite a lower overall average score, it’s worth noting that AL teams collectively won 50.6% of their games in 2016, while NL clubs won 49.3% of theirs. While the NL’s top arms (Scherzer, Kershaw, etc) were completely unmatched statistically by anyone in the AL, if you take the Cubs & Nationals out of the picture, the NL didn’t have a very good year, at least as far as W-L records are concerned.



So, how about our Gigantes and the promised “super rotation?” At least, that’s what I thought they could be coming into the season. While it turns out that designation went to the Cubs (117-45 projected GmSc record), our gang ended up fairing pretty well themselves.

First, let’s take a quick look back at what we had to deal with in 2015. Madison Bumgarner (18-4-10, +374 Advantage) was the truly only thing that held that group together. The other 8 starters on that club? They combined for a whopping -2 Advantage. Peavy (7-6-6, +72) & Heston (14-12-5, +50) were the only others with respectable lines, while Hudson (7-11-4, -51) & Cain (3-7-1, -61)  brought the overall record down with poor performances.

Together, the 2015 rotation logged a 66-61-35 GmSc record. The 66 GmSc Wins were their worst as a staff in the current era (2012-present). Ironically enough, their projected record of 84-78 matched their on-field record last season. Did I mention I love Game Score as a measuring stick?

In 2016, Bumgarner and Johnny Cueto were among the elite starters in all of MLB, and Game Score affirmed that notion. Bumgarner (22-5-7, +398) ranked 3rd in the NL (and MLB) in both GmSc Wins and Advantage. His 22 Wins were most by a Giants starter since 2012. Cueto (as I hoped he would) proved to be the biggest bargain among free agent signings last winter. He finished 20-7-5 with a +320 ADV (7th in the NL). John Lackey also had 20 GmSc Wins, but his +249 ADV was significantly below Cueto’s.

So, your two horses check in at a combined 42-12-12, +718 ADV. That’s the best 1-2 combo in the bigs this year – even better than the combined +707 score of Jon Lester and Kyle Hendricks. Unfortunately for the Giants, Jake Arrieta (+291) and Lackey also finished top 10 in the NL. The rest of the Giants rotation combined for and Advantage of only +29. The Cubs as a group finished at +1385. Unbelievable. The Giants? Their +747 still checked in at #3 in all of MLB.

What about Jeff Samardzija? Nobody is going to mistake his first season in San Francisco for something amazing, but consider this: his 16 GmSc Wins were #13 in the NL, and his +134 ADV was 20th best in the league. On the flip side, his 12 GmSc Losses were tied for 11th in the NL, so there’s no doubt he took his lumps. After surviving his midseason disaster (2-8-1), he actually finished the season with a mark of 7-2-2. If he can just find even a smidge more consistency next year, there’s definitely some major helium potential for the Shark. It’s on him to even things out, though.

For the year, Samardzija’s Average Game Score finished at a 54.2. That mark was actually 24th among all NL starters with at least 10 GS, putting him squarely in the middle of the 2nd tier for the league. His closest NL comparisons this season (by GmSc Avg) were Dan Straily (54.5), Jerad Eickhoff (53.9), Jeremy Hellickson (53.8), and Matt Moore (53.8 in 12 GS for the Giants). That list certainly isn’t a who’s-who among MLB pitchers, but all of those guys provided pretty good value for their clubs this year.

Here’s how the rest of the Giants staff fared, according to Game Score.

Matt Moore: 12 GS; 7-5-0, +46 ADV

He was 15-10-8, +95 between SF & TB, with a 52.9 Avg. We’ve seen some good and some bad from Moore already, but frankly, his healthy presence could make the Giants a serious contender for the NL West next year.

Jake Peavy: 21 GS; 7-12-2, -69 ADV.

Statistically, Peavy was among the worst SP in the Majors for the first month or so of the season. He leveled out a bit, but his -69 ADV was still 76th among qualified NL pitchers (10 GS or more), putting him right on the line between “below average” & “far below average.”

Is Peavy’s career as a starter over? That’s very hard to know, but if it is, it was a very good one. I calculated career Advantage scores for active SP, and Peavy’s +2,036 is 7th best. Walter Johnson’s estimated +9,029 is the highest career ADV in MLB history, for comparison. Peavy’s adjusted Advantage per 32 starts (A measure I created and call ADV32) is a +173, meaning he is a well above average pitcher for his career. When you consider he’s made 377 starts, that’s pretty darn impressive.

Matt Cain: 17 GS; 5-10-2, -99 ADV

There’s really no sugar-coating how bad Cain was this year. 97 pitchers made 10 or more starts in the NL, and Cain’s -99 ADV was 88th among them. For this last two seasons, he is a combined 8-17-3, with a -160 ADV. That’s not quite the equivalent of a full season’s worth of starts, and he’s earned enough negative Advantage to be among the worst pitchers in either league in any given year.

Amazingly enough, Cain’s career average GmSc of 55.4 is identical to that of Jake Peavy, giving them matching +173 ADV32 scores. Cain’s made about 70 starts fewer than Peavy, and it really doesn’t appear that he’ll make it to +2,000 Advantage for his career. However, his current mark of +1,663 is top 15 among active pitchers (no surprise). These days though, he really doesn’t look to hold any value as a SP.

Albert Suarez: 12 GS; 0-4-8, -7 ADV

Suarez was a 50/50 pitcher if there ever was one in 2016. He failed to earn a GmSc Win in any of his 12 starts, yet still gave the Giants a fair shot to win 2 out of 3 times he took the ball. There’s value in earning ND’s, and Suarez’ Advantage score places him 50th in the NL. That’s Tier 4, or league average (though his 49.4 Avg GmSc is about 1.5 points below the actual NL Avg).

Ty Blach: 2 GS; 1-1-0, +24 ADV

Blach essentially saved the season with his performance on the final weekend of the year, matching up against Clayton Kershaw no less. Blach’s 81 Game Score that day was the 7th best performance by a Giants starter all season (they had 11 outings of 80 or higher GmSc this year). It’s a very small sample, but if you’re like me, you’d really like to see Ty settle into a more permanent role with the club next season.

As a staff, the Giants earned a Game Score record of 78-56-28 in 2016. Their 78 Wins were 3rd in the NL, their best as a team in the past 5 seasons, and a 12 game improvement on 2015. Their projected W-L record for the year was 92-70. With the addition of Matt Moore, I can see that mark improving next season. Now the questions are, can they find a 9th inning presence, and can they get more consistent production at the plate? If so, this team has a chance to be a real contender again going forward.

Thanks for reading, and please feel free to post any thoughts, questions, or concerns in the comments.



15 thoughts on “2016 Giants Rotation: A Game Score Breakdown”

  1. Great stuff.As someone that keeps score at every game and writes game score into my scorebook every night,It’s nice to see someone that loves this stat as much as I…

    1. Thanks for checking in! Game Score is just so simple, and yet it can give you a very good picture of how a pitcher has fared over the course of a game, a season, and even a career. I don’t know why it hasn’t caught on for most, but I really have enjoyed studying it. It truly has changed the way I look at starting pitching.

  2. I really appreciate Game Score as well. I don’t know if any metric can account for stressful, high leverage situations. If we’d had more offense? Would our pitchers have performed even better? That sort of thing is impossible to know. And then, with Game Score averages… Maybe one out of 8-9 starts, Bum will spin a doozy. I’m not one for cherry-picking/bias, so it obviously still counts. But it severely skews the average GSc. Maybe all others average 70+ GSc. In my mind, this is something I’ve never resolved.

    I remember back to when Peavy pitched for SD, but mostly his time for CHW. My goodness, he was fantastic. That lat separation injury seemed awful. Sometimes I can barely believe he got through it. It’s a sad way for some of our Giants to go out. No heartwarming farewell. Some bitter memories right at the end. Welp. Baseball.

    I know how much work you must’ve put into this, Cove. Wow. The way you’ve written and presented so many stats and facts and opinions and conclusions is fantastic. It’s not easy to make it all flow and track for a reader. And be interesting. You’re great at all of this. I learn a lot. Cheers. Dream big…

    1. Thank you for all the compliments Carmot. I have done so much digging with Game Score, I really thought it was time to share some of the findings. Organizing the data and trying to tell a story with it is my biggest struggle, but I will continue to work on that aspect, as I feel I have so much to share.

      I do know what you mean about skewed samples, especially with Bumgarner. This was certainly Bum’s most complete season as a starter, but he did still have that stretch midseason where he lost 3 times in 5 starts (Scores of 25, 42, 35). His two wins during that stretch were scores of 76 & 79, further proving your point that when Bum is good, he’s GOOD.

  3. 1. Glad I keep checking your blog.

    2. Wrenzie deleted his blog a while back. So you might want to update your side-bar.

    3. Do you have something similar for our relievers? Because I felt the Giants were a 96 win team under the assumption that everything else remained, relatively speaking, within the norms of 2014/2015 on offense. But the relievers… They cost us at least 10, maybe as many as 14, wins this year thanks to their abysmal save rate.

    1. Hey Moses. Wrenzie shut down his blog? Bummer! I enjoyed reading his posts. I don’t know of any game score for relievers, but I’m sure we could come up with something similar with a bit of tweaking.

      1. Cove and Mozes: His twitter is still active, and Wrenzie closed his old blog down by migrating it to a new one:

        I can never figure out how to do paragraph breaks in your comments here. Wouldn’t it be possible to use the same Game Score formulations for relievers- I mean, kinda sorta? Wouldn’t work too well, because of the innings pitched per appearance, so never mind, I guess not. I’ve been reworking my own metrics, that I’ve had a little project for the past 3-4 years. It’s a crazy project. I guess that fits me pretty well then!

        Anybody ready for Shohei Otani to come stateside? I feel like it’s going to happen. Cheers.

  4. You ever consider hosting live chats on your blog once in a while, like fangraphs does? They use Jotcast for their chats. It would be cool to have a live back and forth on stuff like this.

    1. You know, I never really considered it before. I’m not sure how much interest it would gain since I haven’t been actively blogging for quite a while (until just this week), but I agree that it would be pretty sweet to set something up like that. Maybe I can look into it. Thanks for the suggestion.

  5. Rotation is really good right now.Blach showed a lot of guts in that LA start.He,Stratton,and Blackburn should have a good camp battle for fifth spot.I’m comfortable with whoever wins that spot.Beede should be ready by the All-Star break.Giants have done great job drafting,trading,and signing pitchers(3 WS).Stratton,Blackburn,and Blach providing depth and versatility.Coonrod,Suarez,Krook,and Crick big league starters in my opinion.Bullpen has a lot of talent too.Law,Strickland,Osich,Okert,Smith.It’s time to take the kid gloves off with these guys.They’re ready for the challenge.We should see Gregorio closing in Sacramento.It would be nice to see C.Johnson as his set up guy.I think both of those guys are going to have good years.Peace Kyle.

  6. I’m sorry to be dim, but how does one figure Advantage? And are you using Tango’s game scores, with 40 to start with, or James’s, with 50 to start?

    1. Hey Campanari! Not dim at all. I use James’ Game Score, but would like to do more comparisons with Tango’s (I really like his work). Advantage is a pitcher’s total Gam Score minus 50 for each start.

  7. Starting line up projections 2017

    Sacramento Rivercats

    C-J. Arnold



    Closer & Set up


    Richmond Flying Squirrels




    Closer & Set up


    San Jose Giants




    Closer & Set up


    Augusta Greenjackets

    SS-Van Horn



    Closer & Set up


I'd love your feedback!

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s