2016 Giants: September and Final PQS

This post has the Giants Pure Quality Start scores for the month of September 2016, as well as final PQS, PQS as defined in Ron Shandler's Baseball Forecaster annual book and they published the details here (unfortunately, they removed the article; this link gets you at least to the PQS definition, read down to middle for details). I wrote on this first in 2006 (wow, 11th year of this!  10th anniversary!) and have compiled their stats on a regular basis, so I'm continuing it this season for continuity and historical comparison (there is the "PQS" label that you can click to see the old posts on this). Regular readers can skip to the section on "Why PQS?".

This is the Quality Start with a sabermetric DIPS twist, and it gets really easy to calculate once you get used to it. I don't think it's the end all or be all, but then nothing really is that. It is, as I like to say, another piece of the puzzle. A dominating start is scored a 4 or 5 and a disaster start is scored a 0 or 1. DOM% is the percentage of starts that are dominating, DIS% is the percentage of starts that are disasters (any start under 5.0 IP is automatically a 0, or disaster).


What's Good and What's Not

From my observations, a DOM at or above the 40% mark is indicative of good pitching; above 50% is great; above 70% is elite. A low DIS is also indicative of good pitching, just look at the table in the link above showing DOM% and DIS% on the axes.

Basically, you want to see a pitcher's DOM% to be over 40% and ideally over 50%, and you want their DIS to be under 20% and ideally under 10%. For example, Johan Santana has a 76% DOM and 3% DIS in 2006 (2.77 ERA), whereas Orlando Hernandez had a 52% DOM and 28% DIS (4.66 ERA), and Adam Eaton had a 31% DOM and 31% DIS (5.12 ERA). Read the link (unfortunately, they removed the article and thus the table is no longer available, sorry), as I noted, there's a nice chart there showing the combination of high DOM% and low DIS%, and there you can see particularly how a low DIS% is so important to a low ERA.

If you had to chose a high DOM% or a low DIS%, pitchers tend to have a lower ERA when you have a low DIS% vs. a high DOM% (obviously if you combine both, you have a much better chance of having an elite pitcher).  But I think when the DOM% is high enough, you win more by choosing a high DOM% over a low DIS%, as there are more high quality games pitched overall, which is the obviously true key to winning.

I wholeheartedly recommend buying Baseball Forecaster and learning more about their methods of analyzing baseball. It has been greatly illuminating for me, and if you want to get a taste for it without paying full price, they used to sell their old editions of their annuals on their website for half price or less (plus shipping); but that was before he sold the company off, and I haven't checked recently.

Why PQS?

Everyone knows that a well pitched game is key to winning, that's nothing new.  What most analysts don't seem to acknowledge is that there are pitchers who, year in, year out, are good at throwing quality games regularly.  Or maybe it's just accepted by some, but they are unhappy with the traditional definition of a quality game, which, way in the past, when starters were expected to throw complete games, it was giving up 3 runs or less in 7 or more innings, but with the bullpen revolution, is only 3 runs or less in 6 or more innings, which works out to a 4.50 ERA if 3 in 6 IP, which is why people had problems with this definition.  It should really be 2 or less in 6 or 3 or less in 7, but that complicates what used to be a dimple definition.

Another tool that some use is GameScore, which was invented by the legendary Bill James.  I've used it to look at prospects before, because PQS punishes the rating when the pitcher does not go at least 5 IP and key to getting a good PQS is pitching at least 6 innings:  prospects are often pulled before reaching 6 innings.  While it appears to be a better tool, and in particular, more precise tool (since it can vary, theoretically, from 0 to 100 - it has gone over 100 before though), it was created long ago, before many of the newer saber facts were discovered.  What hurts it's usability, in my opinion, is that it takes into account when runs are scored on the pitcher, valuing results over process.  

What is PQS?

PQS As I noted above, PQS is quality starts with a sabermetric twist.  It's really easy to do once you learn the system, and even easier if you pull the game data and have a spreadsheet calculate everything for you.   It looks at the process of pitching and rewards a quality start based on how the pitcher was able to control the inputs - IP, hits, walks, strikeouts, homeruns - and not on the outputs - runs - which the pitcher does not always have control over, as it could just be random luck of when the hit happens and/or random luck of whether the reliever keeps the inherited runners from scoring.  

For example, you could have two starting pitchers with the exact same pitched game, exact same order of ball, strikes, hits, everything exactly the same.  However, while the reliever for the first pitcher was able to stop the rally by stranding the bases loaded, the reliever for the second pitcher gave up a grand slam homer.   This reduces the game score drastically for the second starter:  drops it 12 for Bill James version, 9 in Game Score 2.0 that Fangraphs uses.  So for the first pitcher, who, say, had a nice 59 Game Score, which is good, the second pitcher got 47 with Bill James, 50 with 2.0, which is a roughly average score.  Per a detailed analysis of Game Score, a 59 game resulted in a win 62% of the time, which results in a 100 win season if the staff can do that every start, whereas a 47 game resulted in a win 51% of the time, which is an 82-83 win season.  Huge difference, but each pitcher had the exact same start.

So I like PQS more than I like Game Score.  Not that PQS is perfect, but this has been my stance with the Giants for many years now,  I would rather focus more on the process than the results.  Perhaps the trade or signed player did not work out.  But the key thing to me is whether those were the best choices/decisions facing Sabean at the time he made it.  Sometimes you just have to sign the best player available, or trade for the best guy other teams are willing to give up for your non-prospect (since the Giants under Sabean has rarely traded away a prospect that they really like).  But results are important too, at some point, and three championships are pretty good results, showing that his process has been pretty good over the years.  

And I like the analysis of games versus overall results more because each game started by a pitcher is different, but the results within a game, are often related.  Perhaps he woke up wrong that day.  Perhaps everything clicked.  And do more starts click or more starts suck?  Or are just mediocre.  

That helps to understand a prospect starting out in the majors.  Brad Hennessey came up with pretty good minor league ERA, in spite of a pretty poor strikeout rate, and was able to maintain an OK average-ish ERA in the majors for a number of years.  But that was mainly because he was good enough to avoid disaster starts.  Meanwhile, he was unable to generate many dominant starts.  That showed that he was at best a mediocre major league pitcher unless he could figure out how to strike out more batters.  He wasn't able to.

Like I try to convey, trying to understand players is like putting together a puzzle.  PQS is a piece of the puzzle, albeit an important piece, in my opinion.  So is strikeout:walk ratio and strikeout rate (which should lead to a high PQS score often), plus scouting reports on what the pitchers can do and which are their best pitches, and which are better than average.  I've liked quality start as a measure since following Fergie Jenkins' career, and PQS really clicked with me since it was sabermetrically derived.  

It is not a well-known measure, but that's OK, just like there is a Baseball-Reference.com that focuses on results and Fangraphs that focuses on process, Game Score is focused on results and PQS on process.  To be perfectly frank, I'm more of a results guy, because I think Sabers think that they know more than they really do about how baseball works, and that hubris leads to too many poor conclusions (like the disavowal of clutch performances), but I like the simplicity and ease of calculation that PQS provides, and the results from the metric over the years have convinced me of the utility of this metric.

Giants Starters' Final PQS for 2016 Season

Madison Bumgarner- (76% DOM, 0% DIS; 26:0/34):  2, 5, 2, 5, 5/5, 5, 4, 5, 5, 5/5, 5, 5, 5, 5, 2/5, 5, 3, 5, 5/2, 5, 5, 3, 3, 5/5, 4, 4, 5, 4, 3/

Matt Cain- (24% DOM, 41% DIS; 4:7/17):  4, 0, 0, 3, 3/0, 5, 5, 5, X/0/0, 2, 3/3, 0, 0//

Johnny Cueto - (66% DOM, 6% DIS; 21:2/32):  4, 4, 5, 3, 5/3, 4, 4, 5, 4, 5/3, 5, 5, 5, 2/4, 5, 1, 5, 2/4, 2, 3, 3, 5, 0/5, 5, 5, 3, 4/

Matt Moore - (75% DOM, 25% DIS; 8:2/12):  ////4, 4, 4, 3, 5, 4/0, 4, 2, 0, 5, 5/

Jake Peavy - (48% DOM, 29% DIS; 10:6/21):  3, 0, 2, 4, 0/3, 3, 5, 0, 4, 4/4, 4, 5, 4, 0/5, 0, 4, 3, 0//

Jeff Samardzija - (56% DOM, 16% DIS; 18:5/32):  3, 4, 3, 4, 4/5, 4, 4, 5, 5, 3/1, 0, 4, 0, 2/3, 5, 2, 1, 2/4, 2, 5, 5, 5/0, 5, 4, 3, 5, 4/

Albert Suarez - (25% DOM, 42% DIS; 3:5/12):  //4, 3, 0, 4, 3/5/0/3, 1, 1, 3, 0/

Ty Blach - (50% DOM, 50% DIS; 1:1/2):  /////0,5/

X - Cain had a disaster start, but was dealing when hamstring went out, not counting in PQS stats

Giants Season overall - 57% DOM, 17% DIS out of 161 games counted (91:28/161)
Giants Month of April - 48% DOM, 16% DIS out of 25 games counted (12:4/25)
Giants Month of May - 79% DOM, 7% DIS out of 28 games counted (22:2/28)
Giants Month of June - 56% DOM, 22% DIS out of 27 games counted (15:6/27)
Giants Month of July - 46% DOM, 21% DIS out of 24 games counted (11:5/24)
Giants Month of August - 52% DOM, 14% DIS out of 27 games counted (14:4/27)
Giants Month of September - 57% DOM, 23% DIS out of games counted (17:7/30)

September (really, Sept/Oct) had its ups and downs.  Ultimately, while it was one of the best months for DOM%, it was also the worse for DIS%, with 7 in the 30 starts.  Had the Giants had, say, only 4 DIS instead, that could have gotten them to 89-90 wins.  And with the 9 blown saves in September by relievers, just changing that (there were only 2 in August, 1 in July, 6 in June, 3 in May, 3 in April) should have won the division for the Giants, and then they face the Nats instead of the Cubs.  And given how closely the Giants played the Cubs in a five game series, who knows how they would have fared in a seven game series instead in the NLCS.

As noted, it was one of the better months for DOM.  Bumgarner led the way again with 5 DOM starts, followed by Cueto and Samardzija with 4 DOM starts each.  Moore had 3 DOM and Blach had his brilliant DOM start as well.

As also noted, it was the worse month for DIS.  Suarez, unfortunately, was not up to the task, and had 3 DIS starts out of five starts.  Moore was also shaky with 2 DIS starts out of 6.  Blach and Samardzija had 1 DIS start each.

Now, this is a good month showing how sometimes the DOM/DIS profile does not always translate into ERA results.  In spite of all the DIS starts, overall, the starters still had a good 3.40 ERA.  This was mostly because Suarez, in spite of 3 DIS starts, only had a 4.13 ERA.  Normally, his ERA should be much worse, for example, Moore with 2 DIS, had a 5.17 ERA in spite of his 3 DOM starts.  In fact, Suarez's ERA was not much worse than Bumgarner's 3.92 ERA for the month, in spite of Bumgarner's 5 DOM starts, showing the other end of the spectrum, where DOM does not lead to good ERA.  Dominance does not always end up with good results, hence why my playoff studies showed an 80%, roughly, win rate when you have DOM starts, because even if you pitch really well, the ball sometimes bounces the wrong way for you.

Still, most of the time, DOM = good ERA.  Cueto with his 4 DOM/0 DIS month had a 1.78 ERA.  Samardzija with his 4 DOM/1 DIS month had a 2.95 ERA.

Similarly with DOM/DIS, K/W and K/9 are key indicators, generally, of good ERA.  Bumgarner led the way with 8.80 K/W and 10.2 K/9, but his ERA was average.  Still, Cueto was second in K/W with 5.00 and he had a very high 8.9 K/9 and Samardzija was third with 4.20 K/W and first with 10.3 K/9.  Moore was actually pretty good overall with a 3.00 K/W and 9.5 K/9.  This was a very dominant top 4 starters and why I am very enthused about our chances in 2017.

Ultimately, I think Suarez's poor DIS performance with OK ERA reflected more a managerial decision to shut him down before things got bad.  His K/W was horrible, only 1.29, and that was because his K/9 was horrible, only 3.4.

2016 Comments

Overall, the staff did well with 57% DOM/17% DIS.  Remember, 60% DOM is very good for a starting pitcher and our staff almost did that as a group.  And 17% DIS is OK too, you want it under 20%.

Bumgarner led the way with 76% DOM/0% DIS.  The Giants were 20-14 in his starts.  A pattern for him during his career is a slow start, and this season was no different, as he had two MID starts out of three to start.  And I know that many laud his ability to throw many innings, especially the Giants, but once he got warmed up, he had 15 DOM (all but 1 were 5 PQS too, mind you) out of 17 starts (88% DOM/12% MID), but then ended the season's last two months with only 8 DOM out of 12 starts (67% DOM/33% MID, with 3 of 8 being 4 PQS DOM starts).  So he was still very good, but qualitatively, he went from Superman to Superiorman.  And he ended up with good but not great 2.74 ERA, though great 4.65 K/W and 10.0 K/9.

Cueto was not far behind with 66% DOM/6% DIS.  Even better, the Giants were 23-9 in his starts, as he pitched better than his stats often.  He also wore down during the season in terms of performance.  In the first half, he had 13 DOM starts in 16 starts (81% DOM), similar to Bumgarner , with a 2.42 ERA.  But in his second half, Cueto had 8 DOM/2 DIS starts in 16 (50% DOM/13% DIS), which is good, but just barely there good, though his ERA was still good at 3.19.  So he was still better than many pitchers and overall, a 2.79 ERA with 4.40 K/W and 8.1 K/9, but his second half was worse.

This is why I've been advocating for a 6th starter during the month of August.  Generally, there are only two days off in that month (oddly, the Giants were scheduled with few days off early, and ended up with 4 days off in August this year; but they only had two days off in 2015, 2013, 2012, & 2011, three days in 2014 & 2010) and the starting pitchers get slammed with less rest during the month.  Especially if we already have a long reliever on the roster, you put him in so that there is an extra rest day during the week, whether from an actual day off or the day the long reliever starts.  That only takes away one or two starts from the starters (depending on how many off days there are in the month), though from Bumgarner and Cueto, the #1 and #2 starters (since there are 162 starts in a season).  It gets harder if a starter has to be brought up from the minors, which means someone gets sent down, but, again, it helps ease off some of the work our top starters are getting, and hopefully leads to less of a drop in the second half, in terms of performance.

Samardzija was the next step down with 56% DOM/16% DIS.  The Giants were only 15-17 in his starts though, and that was reflected by his average 3.81 ERA overall.  He had two halves as well, as the Giants went 10-6 in his first half, 5-11 in his second half.  And for him, it wasn't really halves, he had a poor 13 start streak there in the middle (only 2 DOM/3 DIS), sandwiched between a 8 DOM out of 10 starts and a 7 DOM/1 DIS out of 9 starts, both dominant streaks.  He was basically lost for a couple of months there, before turning things around.  That's ERA of 2.54, 5.83, and 2.70, covering the three parts of the season for him.  He had a good 3.09 K/W and a nice but not good 7.4 K/9.

As I noted in the pre-season, I figured that this would be a learning year for him and the Giants coaching staff, and it turned out that way.  In that first part, he had a nice 61 K in 71.0 IP, but when he was lost, only 45 K in 75.2 IP.  More importantly, he ended with a strong 60 K in 56.2 IP.  Anything above 9 K/9 is excellent.  So it worked out the way I was hoping, Samardzija was dominant at times, but he had his learning moments as well, so having Cueto to buffer him there in the middle of the rotation was perfect on the part of the Giants.  And had the offense been better in the second half - Giants only went 3-6 in the Shark's last 9 starts, even though he had a 2.70 ERA, which should have been more a 6-3 stretch and would have put us one game behind the Dodgers instead of four, plus we would have won other games - we probably would have won the division.

Peavy was the fourth starter and pitched liked it, though his results were more like a fifth starter.   He had a 48% DOM/29% DIS for a horrible 5.47 ERA, though good 2.73 K/W and OK 7.5 K/9.  In spite of that bad ERA, the Giants were 11-10 in his starts.  Peavy started out poorly, but got it kicked up nicely, in the middle, but ended poorly, and why the Giants had to trade for Moore to take over his spot.

He's probably a fifth or spot starter at this point of his career.  I like him as a pitcher, but as I noted when the Giants signed him, I didn't think that the Giants should have signed him back, I would have been happier giving our prospects the opportunity to win and hold a spot or signing another pitcher to a 1-2 year contract.  Heston did well in 2015, and we could have signed a Wellemeyer/Fister type either year to cover other contingencies, or even brought up Blach sooner in 2016, or brought up Blackburn in 2015 while he was doing well in AAA, maybe his poor 2016 was due to a letdown getting sent back down to AAA or to a new stadium to pitch in.  But, still, he was a good Giant, and I wish him well trying to make back the money he lost due to his advisor stealing $35M or so from him, it looks like the Padres might sign him for a second time around with them.

Those were the starters who got most of the starts (at least 20 starts).  Out of the rest, you got a lot of mixed but mostly poor results.

• Cain had a 25% DOM/44% DIS for a poor 5.81 ERA, 2.13 K/W, and 7.1 K/9.  The Giants were 7-10 in his starts.  
• Suarez had a 25% DOM/42% DIS, but as noted, was pulled before things got too bad, so he had an OK 4.28 ERA, but his poor 2.05 K/W and 5.8 K/9 reflected his poor PQS results.  The Giants were only 3-9 in his starts, which was reflective of his poor PQS.
• Blach had a 50% DOM/50% DIS, but great 1.64 ERA as he had that great start against the Dodgers, while getting pulled quickly in his poor first start (well, more middling than poor, could have gotten a 2 PQS if Bochy had let him continue, but he was probably amped up in his first MLB start).  Still, decent 2.25 K/W and 7.4 K/9.  

Overall, these 5th starters were a collective 8:13/31 for a 26% DOM/42% DIS, 4.91 ERA, and 11-20 Giants record.  If the Giants were 16-15 in those games instead, they would have won the division.

Last but not least, I wanted to cover Moore with more than a bullet.  Moore replaced Peavy as the fourth starter and had extremes in his starts, resulting in no MID starts and a wonderful 75% DOM and poor 25% DIS.  In spite of the 75% DOM, he had an average 4.08 ERA, and that poor result, relatively, was reflected in his average 2.16 K/W.   His great 9.1 K/9, though, means that poor K/W was due to very high walk rate.  Still, the Giants were 7-5 in his starts, which might not seem like much, but over a season that's a 94 win rate or a 19-13 record over 32 starts.

I think this demonstrates how stats can be skewed when you look at less than a full season of starts, where one or two bad starts can really mar the overall record.  In particular, he had a horrible start in Colorado.  Removing that outlier would reduce his ERA to a very nice 3.43 ERA and better 2.43 K/W.  Of course, somebody has to pitch in Colorado, and his result is his result, but the point of this exercise is to get a better view of how well Moore actually did, as that one very bad Colorado start colored his 12 starts greatly, whereas had he pitched that one bad Colorado start out of 32 starts instead, it would not have skewed his results as badly.

2017 Rotation Thoughts

I'm very excited about the Giants 2017 rotation.  We potentially could have four ace-type starters with Bumgarner, Cueto, Moore, and Samardzija taking four of every five starts.  Bumgarner and Cueto just needs to do what they can do, Moore and Samardzija need to do what they were doing for us in the last two months.  They could have a collective ERA of under 3 if they can put together a full season of what they showed that they can do.

Of course, there will always be glitches.  I would be surprised if Bumgarner and Cueto don't have good seasons overall.  But Bumgarner will probably have his poor start again, plus wear down at the end.  And Cueto will have his Jekyll and Hyde first and second halves that he has had during his career.  Samardzija seems to be a Lincecum type:  dominating most of the time, but will just lose it for stretches of time.  Hopefully, like Lincecum, Samardzija can learn to minimize the stretch of games where he is lost, and hopefully the Giants coaching can help him get straightened out quickly too.  That's the key to a good season from him.  Moore looks like he will do great for us, and having him makes it less likely that the Giants will try to re-sign Cueto after the 2017 season.

I think it all depends on what happens with the 5th starter.  Cain is expected to compete with Blach for the 5th starting spot, with Suarez, Beede, Stratton, and Blackburn providing some additional competition.  Perhaps the other Suarez in our system as well, Andrew Suarez appears to be a Rueter type of starter, similar to Blach as well.

I think Cain can put it together for a number of starts, like he was doing just before his hamstring went:  he had 3 straight 5-PQS DOM starts (you don't do that by accident) before his hammy broke the streak.  In those three starts, against good competition in Toronto, D-backs (in AZ), and Cubs, he had a 1.71 ERA, 21.0 IP (7.0 IP in each start), 17 K vs. only 4 BB, only 2 HR and 19 hits, and a .253/.284/.387/.671 composite batting line against him.  Basically, because he missed much of spring training, his early starts were his spring training, and he got himself straight but then went down.

In 2017, he should have a normal off-season and normal spring training.  So I expect him to get back to Cain-like standards.  But the question is how long can he maintain it, before injury?  The hammy could have been because of the poor off-season and spring training, but it could also been because of his body starting to get old at 31 YO, many players fall down the cliff in their 30's.  Still, most long time pitchers continue to do well enough to be good in their 30's, and that 3 start stretch last season gives a hint that he has not lost it, but rather is close enough to do as he once did, making longevity the issue instead.

If he can get back to his past, the Giants could have the best record in the NL, let alone the NL West.  The Giants have not had such a dominant rotation in recent years other than in 2011, when Zito was replaced by Vogelsong.  However, 2017 could be a very dominant rotation in that every starter has the potential to be above average in K/9 and K/W.  That along with how good the Giants offense has been, could lead to a 100 win season.

As far fetched as that might seem to some, the Giants Pythagorean was 90 wins in 2016.  Just looking at the fifth starters, we got 11-20 from them in 2016, and if we actually got an ace starter, that could easily be a 20-11 record (see Bumgarner's and Cueto's team record, Moore's too), which would put our Pythagorean at 99 wins for 2016 (or high 90's at the least) and thus it would not take much more for the Giants to reach 100 wins in 2017.

2017 Bullpen Thoughts

And in particular, we should have much better results from the closer with Melancon.  As closer, Melancon saved 47 out of 51 in 2016, 51 out of 53 in 2015 for 98 out of 104, or roughly 94% save rate.  Casilla saved 31 of 40 in 2016, 38 out of 44 in 2015, for 69 out of 84, or roughly 82% save rate and a poor 78% in 2016.  Just in the 40 saves opps that Casilla blew, replacing him with what Melancon did in 2016 would have resulted in 6-7 more saves, which would have won the division for us, putting us at 93-94 wins.  Plus there were the save opportunities in September that Casilla did not get since he was removed from the closer role, which was iffy too, and probably would push the Giants to at least 95 wins with Melancon, as those opportunities would have been his to save.

And that don't include the positive contributions that the addition of Melancon would deliver to the bullpen, not just in the improvement in the closer role.  Per the media, there were a lot of grumbling among relievers about the lack of clarity regarding their role.  So having Melancon around should give clarity that should ease the grumbling.

Though it should be noted that whatever their problems might have been, the relievers actually did pretty well, among those who should have roles in 2017:

• Law:  2.13 ERA
• Kontos:  2.53 ERA
• Smith:  2.95 ERA
• Strickland:  3.10 ERA
• Okert:  3.21 ERA
• Gearrin:  4.28 ERA
• Osich:  4.71 ERA

I thought I should point out that Gearrin and Osich should not be dismissed as difference makers in 2017 based on their poor 2016 ERA.

Gearrin was skewed by pitching with his arm troubles that got him DLed in July.  He had a 2.97 ERA in his first 40 appearances.  Taking out that bad appearance that got him DLed drops him to a more palatable 3.59 ERA overall.  Also, he wasn't quite ready when he came back off the DL, with three bad appearances.  If you look only at the appearances after those three, he had a 1.86 ERA.  If you remove just that first bad appearance, as well as the one that caused his DL, his overall performance in 2016 would have been a 3.23 ERA.  That shows his skill level better, and why he has been touted as Romo's replacement since he joined our bullpen.

Osich just had a bad season overall, but even then, some skewing.  He had a really bad appearance in Colorado, which if it were removed, dropped him to a 4.29 ERA.  He also had a bad stretch before getting DLed, but unfortunately, he has a long history of getting injured, so there will probably be stretches where he's bad before he's DLed.  But, as highlighted by Fangraph analysis on the best pitches in the majors in 2016, Osich was on the Top 15 list (15th but still) with his 96 MPH sinker, which has Top Ten movement in both directions.  He also had a problem with his knee that got operated on after the season, so that should be improved as well.  So I would not count him out for 2017.

Of course, as I noted before, there is a crunch in the bullpen in 2017, even with the loss of Romo, Casilla, and Lopez.  With seven spots (usually), Melancon takes one as a the closer, and presumably the loser between Cain and Blach will be the long reliever (assuming no DL for either), that leaves five open spots for seven skilled players.  Smith isn't going anywhere, and Gearrin, Kontos and Strickland are out of options, I believe.  That would leave Law, Okert, and Osich fighting for the last spot.

If Cain is as good as I think he should be, that could give the Giants another option:  keep Blach in AAA, which would allow the Giants to keep two of the last three on the roster.  If everyone is pitching well and thus into the 7th and beyond regularly, the relievers won't be used as much, and the need for a long reliever is lessened, where Bochy could ham and egg the relievers to cover six relief innings, say.  If they go this route, this then probably limits the competition to Okert vs. Osich, as both are lefties, and Law was being treated as a future closer in the waning days of the 2016 season, and in fact, was the first pitcher used in that fateful 9th inning against the Cubs.

It is a good problem to have, and gives us good backup in AAA in case anything should happen to anyone, as most of them have had arm problems before, and many of them had problems just in 2016, with Law, Gearrin, and Osich had issues in 2016.  Plus we have Slania, Gregorio, Stratton, Suarez, Cory Taylor, and Rodolfo Martinez waiting in the wings, potentially.  Slania looks like he could b ready, and Martinez is probably on the fast track because of his 100 MPH talent.  Ray Black also tantalizes with 100 MPH talent, but has been way to wild to hope for the moment, but does deserve some mention.  Some learn to let go of some of that speed and gain enough control to go with high 90's heat to make the majors.

On the starters side, we would have the loser of Cain/Blach as the next starter should anything happen.  Then Beede, per the recent Top 10 Prospect list by Baseball America, the Giants believe that he's ready for the majors in 2017.   Stratton and Blackburn has some recent AAA success to point to, plus there are Gregorio and Suarez probably waiting next, if not passing these two, as both are now Top 10 prospects, 7th and 5th, respectively.  Plus, there are the various former MLB pitchers who the Giants have invited to spring training or signed to minor league deals that could be additional depth (you never know, that's how we got Vogelsong and Petit), like Josh Johnson and Ricky Romero (and I'm sure the Giants will add others along the way to Opening Day).