SB Nation User Blog: GiantFan

WAR, Postitional adjustments and moving Pablo Sandoval to 1B

GiantFan — Wed, 09 Dec 2009 09:45:04 -0000

I feel it's worth mentioning something that people need to consider when you're using WAR to compare players and also when you're talking about whether to move Sandoval to 1B or not. WAR is calculated using positional adjustments, which are a good way of determining the overall ability of a player or when comparing players who play at different positions. It's very good if you're looking at two players in a vacuum but troubles arise if you'e doing a comparison in a team context.

In the Giants situation that can cause trouble if you're comparing a 1B and a 3B using straight WAR. The reason for this is that 3B get a 1.5 WAR boost over 1B because they can play 3B and so they should. In the Giants situation they have a player who can play 3B (Sandoval) and so the positional effect isn't anywhere near as big.

How big is the effect? Well that depends on how Sandoval's defense rates at the two positions. So far he has a UZR/150 of -3.6 which is only in a small sample size, but seems reasonable. I'll take that as his true level for the moment but you can adjust accordingly if you disagree. In order for the positional 1.5 WAR adjustment to be true, Sandoval would therefore need to be a +11.4 1B. Now that just isn't feasible as no 1B is that good defensively and so the Sandoval will "lose value" at 1B.

Now how good defensively will Sandoval be at 1B? Well he should be pretty decent although i don't think he's going to a league leader defensively. Three examples of well renowned defensive 1B are Adrian Gonzalez, Mark Teixeira and Casey Kotchman, they have career UZR/150 of 0.1, 1.7 and 4.4. They average to around +2 per 150 games and so lets say that Sandoval is above average and can put up a UZR/150 of around 1.4 (as that's a convenient number).

That means that that Sandoval will gain around 5 runs defensively from the move to 1B which is nowhere near the 15 runs the positional adjustment provides. What that means is that if we were to sign a 2B/3B his actual effective WAR (as in how good he'd make the team) is quite possibly 1 WAR less than is actual stated WAR. So a free agent 2B/3B who you think is a 2.5 WAR player might actually only effectively be a 1.5 WAR player when compared to a 2.5 WAR 1B. Or the improvement over Garkokawa might seems like it's worthwile and a 1.5 win improvement when actually it's only a 0.5 win improvement. So a pretty rough rule of thumb is to subtract 1 WAR (or however much you think) from the value of a 2B/3B if you're using WAR, to get their effective value to the team.

Now the numbers aren't exact and you can argue about the specifics but if you're using WAR and the like you've got to consider the effect on the team as a whole and not just on the single players. The Giants may not really understand how moving a player down the defensive spectrum has an effect on the team, and assume that it doesn't matter whether Sandoval plays 1B or 3B but the reality is that the two players you're comparing aren't being compared to the same baseline.

Free Agent 1B by WAR

GiantFan — Fri, 06 Nov 2009 16:38:01 -0000

Free Agent 1B by WAR

A nice graph from BtB showing the free agent 1B and their WAR from last year (wOBA and UZR are also displayed separately).

There's only four that make sense to look at: Nick Johnson, Adam LaRoche, Russell Branyan and Carlos Delgado. Obviously the relative cost will determine which ones would be worth going after.

The quest for a league average offense - Why has our 1B production been so bad?

GiantFan — Thu, 22 Oct 2009 13:37:13 -0000

Firstly, I'd like to apologize for the amount of fanposts I've made recently, a quiet couple of weeks with work and the end of the season have meant I've come up with a few ideas. This post nicely links to the one below which discusses the topic of our favourite Giants 1B.

If i were to ask the question "Which position have we got the worst offensive production from (relative to position) over the last 10 years?" I'm sure the majority would say 1B. So this topic is nothing groundbreaking, I'm merely going to investigate whether that's indeed true and the reasons behind it.

Has our 1B production been that bad?

The easiest way to answer this question is to look at the sOPS+ of each position per year. It basically takes the park adjusted OPS and adjusts for the position (so an sOPS+ of 100 is the league average for the position you're looking at in that year). I've taken the numbers from baseball reference and worked out the average for each position over the last ten years. The results are as follows:

LF: sOPS+ = 150
2B: sOPS+ = 111
C: sOPS+ = 101
SS: sOPS+ = 98
RF: sOPS+ = 94
CF: sOPS+ = 93
3B: sOPS+ = 90
1B: sOPS+ = 81

Overall that works out to be an OPS+ of 102 so we've been slightly above average offensively over the last 10 years (obviously due to an above average first half and below average second half of the decade). On the whole there's not a huge amount of surprise in those numbers. Our LF production was always going to be top due to the fact we had Bonds for a number of years, although it does illustrate just how good Bonds was. The fact that 2B was our next best position again isn't very surprising considering it includes some great years from Kent and some solid ones from Durham.

Clearly by far the worst production came from 1B which is in-line with my expectations at least. To illustrate that point further we've only had one season (2004) in the last 10 years in which our 1B production has been average or above. After the jump I'll illustrate the pretty obvious reason why our production has been so poor from 1B.

Why has our 1B production been so bad?

Because we haven't used any resources on the position. I'm afraid that's an incredibly obvious answer and nothing that you didn't know already and so sorry if you were expecting a deep and meaningful conclusion. The reality is though, that that's the main reason and perhaps more interesting is the illustration of how little resources we've used on the position. What do we mean by resources though? Well, there are three main avenues for bringing talent into an organization. One of those is the trade market, although usually that is a swap of talent rather than an influx, and trades haven't really affected our 1B production a lot over the last ten years. I'm going to study the two other forms of bringing talent into an organization, by drafting and by spending money (often free agency but not limited to it).

The Draft

In order to look how we've used our draft resources I've looked back over the last 10 years and seen how we used our draft picks per position.

OF: 17 picks in top 10 rounds, 10 picks in top 4 rounds, 2 picks in first round (including supplemental). Obviously OF covers three positions and so we've had far more OF picks.
SS: 7 picks in top 10 rounds, 3 picks in top 4 rounds, 2 picks in first round
C: 6 picks in top 10 rounds, 4 picks in top 4 rounds, 2 picks in first round
3B: 5 picks in top 10 rounds, 4 picks in top 4 rounds, 1 pick in first round
1B: 4 picks in top 10 rounds, 1 pick in top 4 rounds, 0 picks in first round
2B: 4 picks in top 10 rounds, 0 picks in top 4 rounds, 0 picks in first round

Now this is by no means an exhaustive study, there's a lot of difference in the value of picks and often picks end up moving position but it does help illustrate that we haven't used many draft resources on 1B (or big hitters on the whole) We've only used 1 pick in the top 4 rounds over the last 10 years and that was Lance Niekro (a 2nd round pick). The only position where we've used as little draft resources on has been 2B, although that's somewhat understandable given our great production from 2B in the majors and the fact that a lot of 2B start life as SS.

Money

I've mentioned one way of bringing talent into the organization, now I'm going to look at the other way, using money to gain and keep talent. I've had a look over the last ten years and calculated how much we've spent on average per position. Now these calculations are only rough, I've ignored the cost of league minimum players, the figures may not be perfect (due to bonuses, deferred payments, etc) and I've tried to distribute money fairly when players have played multiple postions.

LF: $14.15M per season
2B: $6.16M per season
RF: $5.59M per season
CF: $4.87M per season
SS: $4.71M per season
3B: $3.97M per season
1B: $3.44M per season
C: $3.01M per season

Now again, there's no great surprises with some of those numbers, we've spent a lot of money on LF due to Bonds, etc. I was surprised that 1B didn't come bottom as i was expecting it too, but then i forgot just how much we spent on JT Snow (works out to $1.85M per year on 1B over the last 6 years). It's clear though that we've spent very little money on 1B.

Conclusions

Comparing the production and cost lists you get the expected result that you get what you pay for (or conversely you pay for what you get) as the ordering is very similar (aside from the positioning of C). If you try and work out value for money then it depends on what replacement level you choose to set, but either way, the numbers are pretty consistent with the best value for money from C and the worst value for money from 1B.

The draft list is different in that it produces a delayed result but it doesn't bode particularly well for our production of home-grown 1B in the next few year, especially given the Villalona situation. It's no surprise that we've got such poor offensive production out of 1B given the lack of resources we've devoted to the position and the only way we'll end up breaking the cycle is if we commit resources to that position, either via the draft, via free agency or by trade.

The quest for a league average offense - How big changes can have no effect

GiantFan — Mon, 19 Oct 2009 14:11:46 -0000

I thought that it'd be useful to contrast my last post with a post about how it's easy to spend a lot of money and make big changes only to find out that it makes no improvement to the team. Again, like last time, this is not a projection of what will happen, merely a look at the performance of the last couple of years.

Corner Infield - Are you really improving the offense?

The following analysis is purely looking at the offensive side of the equation. We all criticize our poor offense and one of the main aims this off-season should be trying to upgrade it. In order to do that though we will have to focus resources on that ares. The following is not meant to suggest that the additions are worthless or that we shouldn't make them, just that the may not actually improve the offense at all.

Ryan Garko. You can argue as to whether he's classed as an incumbent or not but he's the baseline with which I'm going to make comparisons. Over the last two seasons he's been worth 3.7 runs above average (all runs are taken per 500 AB). Over his career he has a wOBA of .347

Hank Blalock. I've not heard his name so much on here but it is one name that's come up in the past with regards to the Giants. Over the last 2 seasons he's been worth -1.3 runs above average. For his career he has a wOBA of .340. Difference = - 5 runs (- 0.07 career wOBA)
Adrian Beltre. His name often comes up and he's been a player I've been very interested in us acquiring previously. Obviously his value comes from his superb defense. Signing him would move Sandoval to 1B which would cause a drop in value. Over the last 2 seasons he's been worth -0.6 runs above average. For his career he has a wOBA of .335. Difference = - 4.3 runs (-0.012 career wOBA)
Chone Figgins. He's one of the bigger names on the free agency market this year and I've seen a few people who want us to bring him in. He's difficult to peg offensively as he's been very streaky over his career and again would involve moving Sandoval to 1B. Over the last two years he's been worth 6.8 runs above average. For his career he has a wOBA of .339. Difference = 3.1 runs (-0.08 career wOBA)

Now obviously the second two have defensive advantages and are better overall players but there's a very good chance that adding any of the above players wouldn't actually improve our offense at all. That's not to say that we should ignore them, but be aware that the money we'd be spending would be on run prevention and not towards improving our offense.

Outfield - Are you actually improving the team?

The following looks at the opposite side of the spectrum, the possibility of signing offensive players who give back the runs on defense. Again it's not to suggest that you don't want to spend resources on offensive players, but you have to be careful of wasting money on players who aren't going to improve the team with regards to winning games.

Fred Lewis. The incumbent example I'm going to use is everyones favourite. Over the last two seasons he's been worth 2.0 WAR (all WAR are taken per 500 AB due to variable opportunities).

Vladimir Guerrero. An old favourite. Had a down year offensively this season, but poor defense has hampered him in recent years. Over the last two seasons he's been worth 1.73 WAR (per 500 AB). Difference = - 0.27 WAR
Bobby Abreu. Recently turned down an extension with the Angels and a possible free agent. An welcome upgrade offensively but poor defense takes away a good portion of his value. Been worth 1.62 WAR over the last two seasons. Difference = - 0.38 WAR
Jermaine Dye. Mentioned often on here (as a player to avoid) as a free agent. Disappointing offense in two of the last three years and terrible defense make him not very valuable. Over the last two seasons he's been worth 0.73 WAR. Difference = - 1.27 WAR

Obviously improving the offense is important but defense needs considering otherwise the money you're spending may not actually be improving the team at all.

Conclusions

I guess the first important thing to note is not to spend decent money for marginal (or non-existent) upgrades. It's easy to get seduced by a potential offensive upgrade but that upgrade might not be as big as you'd hope (see Dye/Guerrero downturns) and may come with an even bigger cost on defense. The second conclusion to make is to make sure you know what you're paying for. There's nothing wrong with spending money on a Beltre or Figgins in order to help improve the teams run prevention. But also don't be surprised if you end up without upgrading the offense. That's the balancing act that is the tricky part of any of our off-season moves.

Worst Defensive seasons: Giants 02-09

GiantFan — Thu, 15 Oct 2009 14:05:31 -0000

Worst Defensive seasons: Giants 02-09

Was looking at previous UZR numbers and wondered if people could guess which Giants put up the worst 7 defensive seasons over this period? There are two multiple offenders so you're looking for 5 players. These are total UZR so will be dependent on playing time.

Did we underperfom or overperform expectations? - Pitching

GiantFan — Tue, 13 Oct 2009 12:58:43 -0000

In the previous post i set out to investigate whether we under-performed or over-performed our projections both from an individual and team perspective. In this post I'm going to undertake a similar analysis but instead looking at the pitching staff. It's a bit trickier with pitching as there are other factors to disentangle, but to do the analysis, I've looked at FIP.

The projections I've used are an average of the Marcel, CHONE and ZIPS projections again removing Marcel for players with limited major league experience. In order to keep consistent with the previous post, positive run differences correspond to an over-achievement. The pitchers are listed by number of innings.

Outperformed Their Projection

Tim Lincecum FIP = 2.34, projected FIP = 3.02, Run difference = 17.0

Barry Zito FIP = 4.31, projected FIP = 4.59, Run difference = 6.0

Brain Wilson FIP = 2.50, projected FIP = 3.78, Run difference = 10.3

Brandon Medders FIP = 4.07, projected FIP = 4.77, Run difference = 5.3

Jeremy Affeldt FIP = 3.59, projected FIP = 3.74, Run difference = 1.0

Merkin Valdez FIP = 4.58, projected FIP = 4.90, Run difference = 1.8

Sergio Romo FIP = 2.13, projected FIP = 4.00, Run difference = 7.1

Underperformed Their Projection

Matt Cain FIP = 3.89, projected FIP = 3.74, Run difference = -3.6
Jonathon Sanchez: FIP = 4.17, projected FIP = 3.99, Run difference = -3.3
Randy Johnson FIP = 4.91, projected FIP = 3.97, Run difference = -9.8
Bob Howry FIP = 3.85, projected FIP = 3.84, Run difference = -0.1
Justin Miller FIP = 4.91, projected FIP = 3.65, Run difference = -7.9
Brad Penny FIP = 4.35, projected FIP = 4.31, Run difference = -0.2
Joey Martinez FIP = 4.86, projected FIP = 4.64, Run difference = -0.7
Ryan Sadowski FIP = 4.72, projected FIP = 4.44, Run difference = -0.9

A quick look shows that it's more balanced that the offense, with around half under-performing their projections and around half over-performing them. If you total up the difference then you can a over-performance of the projections by around 22 runs. Like in the offensive case, an adjustment is needed when we're talking about the team as a whole. A quick back of the envelope calculation suggests doing so means we outperformed our component statistics by around 9 runs (taking into account defensive numbers from UZR).

That works out to an over-performance of around 30 runs in total. if we were to combine that with the offensive numbers then we performed very similarly (in terms of RA/RS) to what the projections suggested in pre-season. Converting that into an expected win record we ended up winning 2-3 games more than projected.

Did we underperfom or overperform expectations? - Offense

GiantFan — Tue, 13 Oct 2009 11:28:32 -0000

In the latest post on Extra Baggs, the question is raised as to whether we underperformed on offensive? Now, it's difficult to say whether we underperformed because there's uncertainties over the true talent level. One way of looking into this question is using the pre-season projections to see how well we matched our expectations both individually or as a team.

To do this i'm going to list every non-pitcher who had more than 50 PA's and compare his projection with his performance. To calculate the projections I've taken an average of the Marcel, CHONE, ZIPS and Oliver projections, choosing not to use the Bill James one as it breeds controversy. For players with minimal experience prior to this year I've taken the average of the last 3 as Marcel doesn't include minor league numbers. The players are listed by PA's and as well as actual and projected wOBA I've also added the corresponding run difference over the amount of Plate appearances that player received.

Outperformed Their Projection

Pablo Sandoval: wOBA = .396, projected wOBA = .335, Run difference = 33.6
Juan Uribe: wOBA = .351, projected wOBA = .302, Run difference = 18.4
Andres Torres: wOBA = .379, projected wOBA = .314, Run difference = 9.6

Underperformed Their Projection

Randy Winn: wOBA = .302, projected wOBA = .335, Run difference = -16.9
Aaron Rowand: wOBA = .323, projected wOBA = .337, Run difference = -15.4
Bengie Molina: wOBA = .308, projected wOBA = .321, Run difference = -5.9
Edgar Renteria: wOBA = .283, projected wOBA = .332, Run difference = -21.5
Travis Ishikawa: wOBA = .313, projected wOBA = .329, Run difference = -5.1
Fred Lewis: wOBA = .327, projected wOBA = .334, Run difference = -2.0
Eugenio Velez: wOBA = .309, projected wOBA = .315, Run difference = -1.5
Nate Schierholtz: wOBA = .301, projected wOBA = .346, Run difference = -12.1
Emmanuel Burriss: wOBA = .258, projected wOBA = .294, Run difference = -7.9
Eli Whiteside: wOBA = .266, projected wOBA = .277, Run difference = -1.3
Rich Aurilia: wOBA = .236, projected wOBA = .318, Run difference = -9.5
Ryan Garko: wOBA = .289, projected wOBA = .342, Run difference = -5.9
Freddy Sanchez: wOBA = .274, projected wOBA = .321, Run difference = -4.4
John Bowker: wOBA = .271, projected wOBA = .320, Run difference = -3.1
Matt Downs: wOBA = .231, projected wOBA = .322, Run difference = -4.7
Kevin Frandsen: wOBA = .182, projected wOBA = .323, Run difference = -6.6

There are a number of interesting things to note one of which is it's clear than we had more players under-performing (16/19) their projections than over-performing (3/19). If you total up the differences you end up with -62.0 for the team as a whole. If you're dealing with the team as a whole, you have to include the variance in team totals and individual components. According to calculations we outperformed our components by around 30 runs, so that means we under-performed our offensive expectations by around 30 runs.

On the whole a difference of 30 runs over a whole team isn't a huge amount and shows that the projections on a team wide scale were pretty accurate. Of course that tells us how we under-performed our pre-season expectations this year but it doesn't tell us anything about how we'll perform next season. The projections for these players next year will, on the whole be reduced due to their poor performances this year. There will be a lot of changes in terms of personnel and playing time next year and it'll be interesting to see how the projections are when they come out. I'll done a similar analysis for the pitching which I'll post in a day or two.

The quest for a league average offense - How small changes can have a large effect

GiantFan — Thu, 08 Oct 2009 09:42:06 -0000

As this site has seemed a little downbeat recently i thought I'd introduce a fanpost to remind people that whilst our offense is awful, there is hope for change. Firstly, it's easy to forget at a specific moment in time but things change very quickly in baseball. For example, there's a good chance that only one of the starters on opening day last year (Rowand) will be back on our team next year. So, whilst our offense is poor now there's no guarantee it will be that way forever (cue snarky Sabean comment).

Before i get on to the main part of this post, a disclaimer. This is NOT in any way a projection of what will happen, it is merely a demonstration of how changing around a poor offense isn't a lost cause. As a point of reference the Giants scored 657 runs last season, with the average in the NL being 718. That means, we needed to have scored another 61 runs in order for us to have been league average on offense. The calculations below use the offensive runs above average from Fangraphs and are only very rough & approximate. Below is a number of hypothetical changes to the roster and the effect they would have had on the offense last season.

The "Big Bats"

Prince Fielder / Adrian Gonzalez. I've seen reference a few times to the fact that we're not one bat away from a competent offense and I'd like to point out that that's not necessarily true. Fielder and Gonzalez were last season around 75 runs better on offense than our Ishikawa/Aurilia/Garko combination. So adding one "big bat" could very easily have the effect of bringing our offense up to standard. Of course players of that caliber are almost impossible to acquire, but we're talking in hypotheticals here. Difference: 75 runs.
Matt Holliday / Jason Bay. These are obviously possible targets as they're available in free agency this year. Holliday / Bay were pretty close offensively last season and would have been approximately 50 runs better offensively than Winn was last year. That wouldn't have been enough to bring us up to league average but it would have gone a long way. Difference: 50 runs.

See what effect younger players (Bowker/Posey) could have had as well as the effect our trade targets (Sanchez/Garko) could have had after the jump.

The Young Guns

John Bowker. It's difficult to determine the effect Bowker would have had on the lineup as he spent most of the year in the minors, but I'm going to use his Fresno MLE (which corresponds to an OPS of 842). There's no guarantee he would have performed like that in the majors but it's an interesting test. Doing a very rough calculation (i'm not sure where you can find MLE wOBA's) Bowker comes out at around 30 runs better offensively than Winn if he'd had the same amount of at-bats. Difference: 30 runs.
Buster Posey. This one is even harder as he played at multiple levels, but i'm going to use his Fresno MLE (which corresponds to an OPS of 761). Again there's no guarantees that he's have performed that way, especially due to SSS but it's an interesting question nonetheless. A rough calculation says that corresponds to around 15 runs better than Molina over the same amount of at-bats. Difference: 15 runs.

The New Acquisitions

Freddy Sanchez. I've seen mention a few times that Sanchez wasn't that big an upgrade (and he obviously wasn't due to injury, etc) but if we'd had Sanchez all year (including, injuries, poor performance and all) we'd have been much better. Taking Sanchez's performance over the whole year (including any missed time) he would have been an upgrade of approaching 20 runs offensively over the Burriss, Downs, Sanchez, Uribe combination. Obviously Sanchez cost a lot to obtain and this isn't a discussion about the trade and it's validity merely a demonstration of relative performances. Difference: 20 runs.
Ryan Garko. Again, Garko's often been said to be of negligible value as he's a platoon player but he would have had a decent effect on our offense last season. I'm going to ignore at-bats against RHP and deal solely on this potential upgrade against LHP. If Garko had hit his career average against LHP (and OPS of 887) then he'd have been worth roughly (depending on the number of at-bats) 15 runs better than the Garko/Aurilia combination we had last season. Difference: 15 runs.

Defense

This discussion was mainly just meant as an illustration of the possibilities of offensive changes, but it's probably worth mentioning defense as an aside.

Prince Fielder / Adrian Gonzalez. Adding in defensive contributions decreases their total value by around 10 runs to around 65 runs total.
Matt Holliday / Jason Bay. Adding in defensive contributions brings Holliday's value down around 10 runs to around 40 runs total. Adding in defensive contributions brings Bay's value down around 30 runs to around 20 runs total.
John Bowker. Using a single season of UZR is too small in of itself, but Bowker's UZR is too small a sample size. Taking him to be roughly league average brings his value down 15-20 runs to 12 runs total.
Buster Posey. Measuring catchers defense is a tricky subject so I'll make no adjustment.
Freddy Sanchez. Adding in defense his value goes up by around 5 runs to 25 runs total.
Ryan Garko. Adding in defense his value goes down a couple of runs so his total value stays around the same.

Conclusions

One interesting result from these calculations is that if you ignore any of the possible big bats and just add up the effect of the smaller changes you get a offensive increase of around 80 runs which would have made us a better than average offense. Now I'm not claiming that we would have scored that many more runs if we'd made those changes at the start of the year. And I'm not claiming that making those changes next season will make us a better than average offense. It's merely a demonstration that it's amazing how such seemingly small changes can have such a major effect on the offense.

And that's the general point that i'm making with this post. It's not meant to be rose-tinted projection of what will happen (or would have happened) there's lot of other things to consider (such as variation in individual performance, variations in the total runs the team scores, defensive implications, etc) and a negative set of conditions could be outlined in the same way. It's just a positive post to demonstrate that we look towards the ridge of offensive competence, that the size of the task isn't what it seems from down here in the gaping chasm of our offensive futility.

AFL Rosters Announced

GiantFan — Tue, 25 Aug 2009 19:46:29 -0000

AFL Rosters Announced

The AFL rosters have been announced and the Giants have 6 members included in the Scottsdale team. Buster Posey, Brandon Crawford, Thomas Neal, Joe Martinez, Daniel Runzler and Steve Edlefson.

Not many huge suprises there but it's nice to see some good prospects going there & it'll be interesting to see how they do.

Top 100 Prospects

GiantFan — Fri, 13 Feb 2009 18:32:37 -0000

Top 100 Prospects

Kevin Goldstein from Baseball prospectus has issued his top 100 prosects. Bumgarner was 3rd, Posey was 9th, Villalona was 47th & Alderson was 60th.

Initial thoughts are that it's great that Bumgarner & Posey are rated so highly. Villalona & Alderson seem very low though, and whilst they may not be as highly rated as our top two i find it strange that there's so much difference between the two sets.