Athletics fans – all three of you reading this – if you have not already read my previous post, then you’re behind the ball. Quickly – Now! – go read my piece on Carl Crawford to catch up.
It’s been widely reported that Billy Beane, GM of the oppressed Oakland Athletics, is willing to offer Mr. Adrian Beltre a sum of $64 million dollars for his services over a term of five seasons, at the end of which (2015), he will be 36 years old. And according to Scott Boras, he’ll be headed to the Cooperstown; we’ll see about THAT.
Let’s run him through our Tom Tango framework (from the previous post, which I’ve already alluded to); because he’s another Big Fish on the market this offseason.
Assumption 1: He’s worth 4.05 Wins Above Replacement (WAR) in 2011. Why? If you weight his WAR in each season from 2004, i.e. his mega walk year for the Dodgers, to 2010, i.e. his near-mega walk year while not equally fantastic with Boston, with a 50/50 split between Fangraphs’ and Baseball-Reference’s WAR figures*, and then you average those weighted WAR’s, you get 4.05 WAR. That’s simple enough, right? In any case, it’s what I’m using.
*We have the same phenomenon as with Carl Crawford, by the way, where Fangraphs loves Beltre and Baseball-Reference just likes him. I think it’s because Fangraphs is giving him too much credit for defense, or Baseball-Reference too little, but that’s a discussion for another time.
Assumption 2: He will regress at a rate of 0.7 wins per season, which seems at least somewhat likely if not being just slightly pessimistic.
Assumption 3: The dollars ($) per win above replacement will start at $4.5 million in 2010 and increase at a rate of 5% per season.
Plug it in and: VOILA! Using these assumptions he’s worth almost exactly $64 million over the next five seasons ($64.2 M), which is the exact offer believed to be on the table for Beltre at this moment. I love it when a plan comes together. To answer the question of my title: no.
Insightful baseball analysis for the casual crowd, Moneyball masses and you, the Third and King crew
Showing posts with label Tom Tango. Show all posts
Showing posts with label Tom Tango. Show all posts
Thursday, December 2, 2010
Tuesday, November 16, 2010
Giants' 'pen the difference in '10
One of my best readers – who just happens to be my brother – asked my to take a look at why the 2010 Giants were better than the 2009 team, when they had a somewhat similar jigsaw puzzle lineup to go with a stellar pitching staff. He submitted that the bullpen had a lot to do with it and I was pleasantly surprised to discover he was absolutely right.
So how exactly did the Giants go from 88-74 in 2009 to 92-70 in 2010? A lot has been written about Bochy’s 2010 “Band of Misfits,” but few words have been written explaining what the main catalyst was, helping to boost the Giants to a few more wins and propelling them to their first San Francisco World Series Championship. Here goes my crack at it.
On a macro level, the 2010 Giants not only scored more runs than the 2009 team, but gave up fewer too. They scored 657 and relented 611 for a run differential of +46 in 2009, good for sixth in the league and resulting in them finishing 7 back in the division and 4 back in the Wild Card. In 2010, they scored 40 more runs (697) – by no means a tremendous sum – and gave up 28 fewer (583) for a run differential of +114. Obviously, this netted them a few more wins and was good for second best in the league along with a NL West division crown. Up next: the micro level explanation.
First off, for the purpose of explaining the Wins Above Replacement (WAR) stat I’ll cite below – and which I use pretty often – here is a simplistic definition of “replacement” from Tom Tango: “Replacement is the talent level for which you would pay the minimum salary on the open market [~400K], or which you can obtain at minimal cost in a trade.” NOTE: Replacement does NOT equal league average. A replacement player is actually about negative (-) 2.0 wins from a league average player.
The main reason the offense scored more runs in 2010 is because they simply hit more balls over the fence. They did improve on reaching base, too (.309 to .321 on-base percentage), as well as on hitting for more power in general (.389 to .408 slugging percentage). But mostly they hit 162 home runs in 2010 – that’s 1 per game on average – when they only hit 122 in 2009, an increase of 33%. You’d probably have to believe they would score more than 40 extra runs with the higher on-base percentage and power boost, but the increase in double plays from 115 all the way to 158 (37% increase), as well as a general drop in speed from slow to slower, does a lot to explain the moderate climb in runs. It is also interesting they were able to improve despite the struggles of really their only solid contributor in 2009, Pablo Sandoval, but then bargain Huff and former 4-A outfielder Andres Torres had career years, and from the time Burrell and Posey arrived they had a much more formidable lineup. The final tally was that their offensive (position players, including defense) WAR went from 11.6 to 19.6, a sizable increase.
The starting pitching was again brilliant in 2010, and they performed remarkably similar in both years.
In 2009, the starters had a 3.58 ERA. In 2010, they had a 3.54. They struck out 8.2/ 9 innings (K/BB ratio of 2.38) in 2009 and 8.0 (K/BB ratio or 2.36) in 2010. They had a 1.251 WHIP in 2009 and 1.253 in 2010. And finally, they gave up .93 HR/9 in 2009 and .95 in 2010. The starting staff’s numbers were eerily similar.
But one of the most significant differences between the 2009 and 2010 team was without a doubt the bullpen. I suppose the beards must have helped. They improved in almost every category that matters most. In 2009, their ERA was 3.49 and just 2.99 in 2010. They struck out more batters going from 8.0/9 to 8.6 K/9 – that’s almost a batter per inning. They also improved on their strikeout to walk ratio, going from 1.96 to 2.20. They reduced their walks plus hits per 9 innings (WHIP) from 1.345 to 1.310. And finally, and perhaps most notably, they did a fantastic job of keeping the ball in the park. They gave up .74 HR/9 innings in 2009 but just .57/9 in 2010. That pushed the pitching WAR from an already fantastic 20.1 to 21.4, and I’ll surmise the impact was greater than the +1.3 WAR would suggest. In my mind I have an image of Santiago Casilla coming into the game in the 6th and 7th innings throwing 97 MPH sinkers in 2010 – as opposed to Bob Howry coming in and giving up gopher balls – and I can literally SEE the difference. Let us not soon (or ever) forget how the ‘pen performed in the game six clincher of the NLCS in Philly.
Looking to 2011, they should have mostly every bullpen arm returning – or at least the ones that matter most. Brian Wilson will be closing. Sergio Romo will be setting up. Santiago Casilla will be getting late and important innings, and Javier Lopez will facing some lefties in those situations, too. They’ll still have Affeldt to give them another lefty out of the pen – and Dan Runzler if he isn’t converted to a starter*. And then they’ll likely have Guillermo Mota or someone that can likely repeat his contribution from 2010. Relievers are highly volatile, partly because of the few innings and thus small sample. But the peripheral stats for a lot of the Giants relievers suggest they can still dominate hitters, and if not, at least perform adequately. The bullpen is more or less set for another marathon in ’11.
*Why ARE the Giants attempting to convert Runzler into a starter, anyway? This is an excellent question. Without all of the information that the Giants have, I can’t definitively answer it. What’s my take? I guess maybe the Giants think he could be a starter, and that he’d offer more value if he was groomed as one. This is true. If the transformation is successful, the Giants will have 6 starters. That’s interesting. It seems likely that if Runzler can handle starting duties: 1) The Giants perhaps want him as insurance in case one of their Fab Five goes down or 2) The Giants are going to actively shop a starter – Runzler or other – to get a position player in return and thus will have a need for a sixth starter. It’s entirely plausible (read likely) that opposing GM’s have already made calls about Runzler, and proving Runzler is a viable starter can only improve the haul the Giants get in return.
To reiterate what I’ve just alluded to, the Giants’ starting pitching, much like the bullpen, is set for 2011. Lincecum should be one of the best number one starters in the league. Cain should be one of the best number two starters in the league. Sanchez should be one of the best number three starters in the league. Bumgarner should be one of the best number four starters in the league. Zito will almost certainly be THE best fifth starter in the league – and it’s not too tall a task, considering most teams pick theirs from a hat. With Runzler, they have options. You can never have too much pitching, and the Giants are the envy of baseball in this category.
Finally, Sabean has a lot of work to do with the lineup, but I’ve already covered that in a previous writing. If Uribe bolts, I’ve no idea how the Giants will have a decent everyday shortstop on opening day. Let’s see what tricks Sabean has up his sleeve this offseason.
Wins above Replacement (WAR) statistics grabbed from Baseball-Reference.com
So how exactly did the Giants go from 88-74 in 2009 to 92-70 in 2010? A lot has been written about Bochy’s 2010 “Band of Misfits,” but few words have been written explaining what the main catalyst was, helping to boost the Giants to a few more wins and propelling them to their first San Francisco World Series Championship. Here goes my crack at it.
On a macro level, the 2010 Giants not only scored more runs than the 2009 team, but gave up fewer too. They scored 657 and relented 611 for a run differential of +46 in 2009, good for sixth in the league and resulting in them finishing 7 back in the division and 4 back in the Wild Card. In 2010, they scored 40 more runs (697) – by no means a tremendous sum – and gave up 28 fewer (583) for a run differential of +114. Obviously, this netted them a few more wins and was good for second best in the league along with a NL West division crown. Up next: the micro level explanation.
First off, for the purpose of explaining the Wins Above Replacement (WAR) stat I’ll cite below – and which I use pretty often – here is a simplistic definition of “replacement” from Tom Tango: “Replacement is the talent level for which you would pay the minimum salary on the open market [~400K], or which you can obtain at minimal cost in a trade.” NOTE: Replacement does NOT equal league average. A replacement player is actually about negative (-) 2.0 wins from a league average player.
The main reason the offense scored more runs in 2010 is because they simply hit more balls over the fence. They did improve on reaching base, too (.309 to .321 on-base percentage), as well as on hitting for more power in general (.389 to .408 slugging percentage). But mostly they hit 162 home runs in 2010 – that’s 1 per game on average – when they only hit 122 in 2009, an increase of 33%. You’d probably have to believe they would score more than 40 extra runs with the higher on-base percentage and power boost, but the increase in double plays from 115 all the way to 158 (37% increase), as well as a general drop in speed from slow to slower, does a lot to explain the moderate climb in runs. It is also interesting they were able to improve despite the struggles of really their only solid contributor in 2009, Pablo Sandoval, but then bargain Huff and former 4-A outfielder Andres Torres had career years, and from the time Burrell and Posey arrived they had a much more formidable lineup. The final tally was that their offensive (position players, including defense) WAR went from 11.6 to 19.6, a sizable increase.
The starting pitching was again brilliant in 2010, and they performed remarkably similar in both years.
In 2009, the starters had a 3.58 ERA. In 2010, they had a 3.54. They struck out 8.2/ 9 innings (K/BB ratio of 2.38) in 2009 and 8.0 (K/BB ratio or 2.36) in 2010. They had a 1.251 WHIP in 2009 and 1.253 in 2010. And finally, they gave up .93 HR/9 in 2009 and .95 in 2010. The starting staff’s numbers were eerily similar.
But one of the most significant differences between the 2009 and 2010 team was without a doubt the bullpen. I suppose the beards must have helped. They improved in almost every category that matters most. In 2009, their ERA was 3.49 and just 2.99 in 2010. They struck out more batters going from 8.0/9 to 8.6 K/9 – that’s almost a batter per inning. They also improved on their strikeout to walk ratio, going from 1.96 to 2.20. They reduced their walks plus hits per 9 innings (WHIP) from 1.345 to 1.310. And finally, and perhaps most notably, they did a fantastic job of keeping the ball in the park. They gave up .74 HR/9 innings in 2009 but just .57/9 in 2010. That pushed the pitching WAR from an already fantastic 20.1 to 21.4, and I’ll surmise the impact was greater than the +1.3 WAR would suggest. In my mind I have an image of Santiago Casilla coming into the game in the 6th and 7th innings throwing 97 MPH sinkers in 2010 – as opposed to Bob Howry coming in and giving up gopher balls – and I can literally SEE the difference. Let us not soon (or ever) forget how the ‘pen performed in the game six clincher of the NLCS in Philly.
Looking to 2011, they should have mostly every bullpen arm returning – or at least the ones that matter most. Brian Wilson will be closing. Sergio Romo will be setting up. Santiago Casilla will be getting late and important innings, and Javier Lopez will facing some lefties in those situations, too. They’ll still have Affeldt to give them another lefty out of the pen – and Dan Runzler if he isn’t converted to a starter*. And then they’ll likely have Guillermo Mota or someone that can likely repeat his contribution from 2010. Relievers are highly volatile, partly because of the few innings and thus small sample. But the peripheral stats for a lot of the Giants relievers suggest they can still dominate hitters, and if not, at least perform adequately. The bullpen is more or less set for another marathon in ’11.
*Why ARE the Giants attempting to convert Runzler into a starter, anyway? This is an excellent question. Without all of the information that the Giants have, I can’t definitively answer it. What’s my take? I guess maybe the Giants think he could be a starter, and that he’d offer more value if he was groomed as one. This is true. If the transformation is successful, the Giants will have 6 starters. That’s interesting. It seems likely that if Runzler can handle starting duties: 1) The Giants perhaps want him as insurance in case one of their Fab Five goes down or 2) The Giants are going to actively shop a starter – Runzler or other – to get a position player in return and thus will have a need for a sixth starter. It’s entirely plausible (read likely) that opposing GM’s have already made calls about Runzler, and proving Runzler is a viable starter can only improve the haul the Giants get in return.
To reiterate what I’ve just alluded to, the Giants’ starting pitching, much like the bullpen, is set for 2011. Lincecum should be one of the best number one starters in the league. Cain should be one of the best number two starters in the league. Sanchez should be one of the best number three starters in the league. Bumgarner should be one of the best number four starters in the league. Zito will almost certainly be THE best fifth starter in the league – and it’s not too tall a task, considering most teams pick theirs from a hat. With Runzler, they have options. You can never have too much pitching, and the Giants are the envy of baseball in this category.
Finally, Sabean has a lot of work to do with the lineup, but I’ve already covered that in a previous writing. If Uribe bolts, I’ve no idea how the Giants will have a decent everyday shortstop on opening day. Let’s see what tricks Sabean has up his sleeve this offseason.
Wins above Replacement (WAR) statistics grabbed from Baseball-Reference.com
Saturday, January 30, 2010
Ryan Theriot > Bengie Molina? Thank wOBA!
One of the stats that has actually made quite a dent in traditional baseball statistics is OPS (on-base % plus slugging %). OPS is quite useful but, as it turns out, it also leaves a lot to be desired. The main reason for that is the statistic gives equal weights to on-base percentage and slugging percentage when it shouldn’t, and furthermore, both of these statistics themselves are not without their flaws. While OBP does tell you the rate at which a player reaches base, it fails to describe how far the player reached. SLG tells you how many bases a player gained, but unfortunately, it fails to see a walk, hit by pitch, etc as a positive outcome. Also, OBP is actually more valuable than SLG. But the buck doesn’t just stop there. Slugging assumes that each base is worth 1.000, so a single is 1.000, a double 2.000 and so forth which is highly inaccurate. Well, a bunch of smart people (Tom Tango and friends) actually analyzed the true “worth” of each of these outcomes and as it turns out it’s just not that simple. One thing we know for sure is that the worst thing a hitter can do is make an out. After running all of the data, and assuming that an out is worth 0, the actual run values are as follows:
HR =1.70, 3B = 1.37, 2B = 1.08, 1B = 0.77, NIBB = 0.62
So the creator of the formula used these values and applied them to each player. When they did it with a league average player and divided by plate appearances, they got around .300. Pretty nice. But, they realized if they simply multiplied by 15% they got the league average OBP…even better. The statistic is now scaled to OBP so league average is somewhere between .330 and .340 (like OBP). The new values and statistic is now:
(0.72xNIBB + 0.75xHBP + 0.90x1B + 0.92xRBOE + 1.24x2B + 1.56x3B + 1.95xHR) / PA
*NIBB = Non-Intentional Walk, HBP = Hit by Pitch, 1B = Single, RBOE = Reached base on error, 2B = double, 3B = triple, HR = Homerun and PA = Plate Appearance
Some might be wondering, why is someone rewarded for reached base on error? Well, because it’s actually possible to create errors, kind of. For example, if Bengie Molina hits the ball on the ground anywhere it’s an automatic out if the fielder can get to it. On the other hand, if Chone Figgins hits the ball directly at the short stop and he bobbles it, Figgins may reach on an error. His batting average isn’t rewarded for this in anyway but he has helped his team and in a way Molina never could hope to. And as it turns out, this outcome is actually worth more than a single. Go figure.
Continuing my bashing of Bengie Molina, allow me to show you how his terrible OBP can be quite detrimental. Bengie Molina posted a .727 OPS in 2009, which isn’t very good. Ryan Theriot managed to post an even lower OPS of .712 in 2009. He must be the inferior offensive player. Wrong. Molina’s wOBA is actually .308 to Theriot’s .318. Though Theriot slugged 73 points less than Molina, his OBP was 58 points higher, and, wOBA shows us that his 58 OBP points to Molina’s 73 slugging points were actually worth an additional 10 points in wOBA. This is just a quick example and a good way to illustrate just how much Molina’s extraordinary out making skills truly do hurt his team, offensively of course.
Why am I telling you this? Well, if you like to read my posts here at PaapFly, you better get used to wOBA because I’ll probably start implementing it into them more often. My goal is to not only expand my knowledge of sabermetrics and baseball statistical analysis, but also each of yours. Lastly, I don’t dislike Molina nearly as much as it may seem. He’s actually a somewhat useful player in that he’s neither a terrible hitting catcher nor a terrible defensive catcher. He’s likeable, he’s durable and he plays hard. What’s not to like? Well, Molina simply represents the wrong direction that Brian Sabean has been pointing the Giants’ franchise and, because of this, Molina gets a large portion of my criticism. Sorry, Bengie.
Cheers,
Rory
If you want to read more on wOBA, check this and this out.
Subscribe to:
Posts (Atom)