The Hardball Times -- Scott Spratt

The Hardball Times -- Scott Spratt http://www.hardballtimes.com/main Baseball. Insight. Daily. en studes@hardballtimes.com Copyright 2013 2013-05-17T08:57:15+00:00 The Roto Grotto: rates versus opportunities http://www.hardballtimes.com/main/fantasy/article/the-roto-grotto-rates-versus-opportunities/ http://www.hardballtimes.com/main/fantasy/article/the-roto-grotto-rates-versus-opportunities/#When:07:33:15 Felix Hernandez went one for three last season, but his .333 batting average was not more valuable in fantasy than the .319 batting average Ryan Braun produced over 598 at bats.

There are several reasons there is not an easy answer. First, the issue applies to counting stats as well as rate stats. Last season, Jose Reyes and Ben Revere each stole 40 bases, but Reyes needed 716 plate appearances while Revere needed only 553 plate appearances. If you were only concerned about winning steals, then Revere was clearly more valuable. You could replace him with a waiver-wire player for the 36 fewer games he played than Reyes and pick up a handful of extra steals.

The Reyes-Revere example illustrates the second complication. There is an additional underlying opportunity that owners have to consider, which is chances to start a player. Hypothetically, two players could have the same number of plate appearances and the same number of stolen bases but play in a different number of games. Over the course of the season, the difference in number of plate appearances for players batting in different spots in the order or batting in the same spot but on teams with different offensive levels can be significant, as Tristan Cockcroft of ESPN recently showed. In addition, players with clear platoon splits tend to pinch hit or be pulled in the late innings of games because of pitcher match-ups.

So, for every statistic, you need to account for opportunities to start a player, and with rate statistics, you need to account for differences in opportunities within each opportunity to start. Your league type will determine the number of opportunities you have to start a player, but I’ll consider leagues with daily lineups for this so I can use games played to approximate it.

First, I calculated each counting stat per games played. Returning to an earlier example, Jose Reyes stole 40 bases in 160 games in 2012, which is 0.25 steals per game. Ben Revere stole 40 bases in 124 games, which is 0.32 steals per game.

Next, I calculated the Z-Scores of counting stats per game. I followed the same method that I used to calculate the Z-Scores for season totals, but substituted the means and standard deviations of those per game statistics. Reyes and Revere had a zSB of 4.00 and 2.90, respectively.

For rate stats, the per game averages are the same as the season averages, but I can then scale those Z-Scores based on opportunities. Last season, Derek Jeter led baseball with 683 at bats. I can use that as my denominator. For example, Felix Hernandez had a raw Z-Score of about 2.45 because of his .333 batting average. However, if I scale that with his three at bats divided by the 683 maximum possible at bats, his scaled zAvg is only 0.01. In contrast, Ryan Braun has a scaled zAvg of 1.63 despite a lesser .319 average because of his 598 at bats.

Now, I finally have counting and rate stats apples to apples. A simple addition of the Z-Scores in each category provides an overall value, similar to a player rate. Here is the top-10 from 2012:

Player	Season	zHR	zSB	zRBI	zRun	zAvg	zTotal
Mike Trout	2012	1.88	4.45	1.33	3.71	1.66	13.02
Ryan Braun	2012	2.64	2.06	2.11	2.13	1.63	10.57
Miguel Cabrera	2012	2.75	-0.51	2.92	1.96	1.95	9.07
Josh Hamilton	2012	3.01	-0.17	2.93	2.09	0.74	8.61
Andrew McCutchen	2012	1.60	1.04	1.41	1.99	1.80	7.85
Edwin Encarnacion	2012	2.82	0.42	2.12	1.54	0.63	7.52
Giancarlo Stanton	2012	3.16	-0.15	1.94	1.50	0.68	7.13
Jose Bautista	2012	3.05	-0.07	1.98	2.09	-0.14	6.92
Matt Kemp	2012	1.89	0.40	1.65	2.11	0.83	6.88

As you can see, different players reach the top in different ways. Four of the top-10 players are actually negative contributors in a category (or two).

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2013-05-09T07:33:15+00:00 The Roto Grotto: average averages and comparing rate stats http://www.hardballtimes.com/main/fantasy/article/the-roto-grotto-average-averages-and-comparing-rate-stats/ http://www.hardballtimes.com/main/fantasy/article/the-roto-grotto-average-averages-and-comparing-rate-stats/#When:07:54:15
Rate statistics are more difficult to handle because they are really two stats in one, the first a standard counting stat and the second the opportunities for that counting stat. For example, batting average is a rate statistic comprised of hits, a counting statistic, and the opportunity for hits, which is at bats.

As with other counting stats, hits can be more or less valuable for your team depending on their context. If you are one hit away from tying another team’s total on the last day of the season, then one hit is tremendously valuable. If you are far away from both the closest leader and trailer of you, then one hit will be less valuable. However, every hit is a positive event.

In contrast, the opportunity event is always a negative event, a fact that requires a bit of framing to understand. Yes, a .300 average is more valuable over 600 at bats than over 300 at bats, assuming a .300 average will increase your team average. However, the reason that is the case is because of the additional hits, not because of the additional at bats.

I could calculate both hits and at bats as a percentage of league totals, as I did with the counting stats. The problem is that a hit is a positive event that does not equal the negative event of one at bat. A batter that produces one hit per three at bats is among the best in baseball.

I can, however, still calculate the league average, and then use it as a benchmark for comparison. Here is the batting average of all non-pitchers over the last three seasons:

Season	Average	StdDev
2010	.261	.026
2011	.259	.028
2012	.258	.031

In recent years, league average has declined slightly. In 2012, it was .258. I also included the standard deviation of the batting averages of players with at least 300 at bats in those seasons, which has been close to 30 points in each season.

With the league average and standard deviations, I can calculate the Z-score of a specific player’s batting average. A Z-score is a simple expression of how much better or worse a sample statistic is compared to the mean on a scale of its standard deviation. A Z-score of 1 is one standard deviation above the mean while a Z-score of -1 is one standard deviation below the mean.

Here are the Z-scores of the batters that were closest to each whole deviation in 2012:

Player	Average	Z-score
Ryan Braun	.319	2
Ruben Tejada	.289	1
Mark Ellis	.258	0
Mike Napoli	.227	-1
Carlos Pena	.197	-2

A player with a high Z-score will have a correspondingly high average. The reason Z-score is a useful statistic is that it allows you to compare different statistics on different scales. Jeffrey Gross explains it well in his article from a few years ago on his auction-pricing model. I’ll hit on a lot of those same points in the coming weeks, and I will try to apply some of those principles of draft preparation to in-season strategy.

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2013-05-02T07:54:15+00:00 The Roto Grotto: dispersal of hitter counting stats http://www.hardballtimes.com/main/fantasy/article/the-roto-grotto-dispersal-of-hitter-counting-stats/ http://www.hardballtimes.com/main/fantasy/article/the-roto-grotto-dispersal-of-hitter-counting-stats/#When:07:39:15
Fortunately, I can approximate those values with data that has been published. In their draft kit prior to 2012, ESPN published the minimum, maximum, and average benchmarks for all of their standard, 10-team roto leagues from 2009-2011. They may not help for deeper leagues or those with exotic categories, but these tables provide a solid foundation for the most common categories and league sizes.

Looking at those averages, you can see that a team that projects to have 300 home runs, 1,100 runs, 1,050 RBI, and 175 steals should expect to net 10, eight, seven, and seven roto points in those respective categories on average. You can also see that it would require about 24 runs, 42 RBI, and 26 steals to barely beat out an average nine-point team in each of those categories.

Since that example team has already reached 300 home runs, enough to earn the maximum 10 roto points, the owner can go ahead and trade away players he expects to hit home runs for players that produce in those needed categories. Since he needs just 24 runs compared to 42 RBI and 26 steals, he might be tempted to try to trade for players that score a lot of runs. However, he already expects eight roto points in runs compared to only seven points in both RBI and steals.

Really, even if the team needed the same number of runs, RBI, and steals for the same additional roto points in each category, the owner should not be indifferent to the category he trades; statistics in those categories do not occur with the same frequency. Fewer steals happen in a season than there are runs and RBI, and far fewer wins are earned by pitchers than any other traditional counting statistic.

Since I do not know which players were owned by what percentage of fantasy teams in a given season, I opted to calculate the roto point benchmarks as a percentage of the total of each statistic in the entire league. For now, I am assuming that a similar percentage of each available statistic is captured by the owned players in a fantasy league, which seems fair since each category has the same available roto points.

Here are the totals of each hitter counting statistic from 2010-2012, as well as the three-year average of each:

Season	HR	Runs	RBI	SB
2012	4934	21017	19999	3229
2011	4552	20808	19804	3279
2010	4613	21308	20288	2957
Avg	4700	21044	20030	3155

And based on the ESPN averages, here are the roto point benchmarks for each category as a percentage of the league totals of each statistic:

	% of Annual League Total, 2010-2012
Roto Points	HR	Runs	RBI	SB
10	6.59%	5.52%	5.64%	7.02%
9	6.22%	5.34%	5.45%	6.35%
8	5.97%	5.22%	5.31%	5.93%
7	5.77%	5.12%	5.20%	5.58%
6	5.58%	5.02%	5.09%	5.27%
5	5.39%	4.92%	4.98%	4.95%
4	5.19%	4.81%	4.86%	4.64%
3	4.96%	4.68%	4.71%	4.29%
2	4.68%	4.51%	4.53%	3.85%
1	4.23%	4.21%	4.21%	3.21%

This table presents the relative cost of each additional roto point with every category on the same scale. You can eyeball the columns to get a sense of which categories are more dispersed than others. Standard deviation summarizes those differences in a single number for each category:

	HR	Runs	RBI	SB
StdDev	0.72%	0.40%	0.44%	1.16%

The hitter counting stats fall into three distinct tiers. On average, runs and RBI are the easiest categories in which to gain and lose ground, stolen bases are the hardest, and home runs are in-between. That may not be enough information to determine the optimal target category for the example team, but it is enough to demonstrate that the differences in volume of needed runs, RBI, and steals does not eliminate any category from consideration in a potential trade.

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2013-04-25T07:39:15+00:00 The Roto Grotto: context is king http://www.hardballtimes.com/main/fantasy/article/the-roto-grotto-context-is-king/ http://www.hardballtimes.com/main/fantasy/article/the-roto-grotto-context-is-king/#When:07:04:15
Because I learned to play in points leagues, I’ve always found rotisserie scoring to be somewhat alien. Baseball already introduced derivative statistics such as batting average and ERA which have components for both successes and opportunities. Those rate stats complicate matters because not all .270 averages are created equal. A .270 average over 600 at-bats can be more or less valuable than a .270 average over 300 at-bats, depending on the collective average of the rest of your starters and the available alternatives in your league.

Rotisserie scoring creates a derivative statistic out of every relevant statistic. A stolen base is not worth a clean five points. In fact, a stolen base is not worth a clean any amount of points because the importance of the next stolen base you receive depends entirely on the context of your placement in your league.

Context is a dirty word to sabermetrics. So much of the important research from projection systems to player value assessment is built upon the removal of context from individual production. Runs batted in remains a universal fantasy statistic, but most fantasy players would cringe if I asserted one player’s superiority in real baseball because of his advantage in RBI totals.

Fantasy players have learned to accept the dissonance that value in real baseball is not the same as value in fantasy baseball, but I continue to see that mentality falter in extreme situations that call for more radical departures in the valuation of players from the value of their brand names.

Here is a simple scenario to illustrate my point. In a hypothetical rotisserie league with one week left in the season, here are the standings of three teams:

	Totals					Roto Points
	R	HR	RBI	Avg.	SB	R	HR	RBI	Avg.	SB	Total
Team A	989	245	966	0.272	179	3	3	2	2	3	13
Team B	923	207	1017	0.29	176	2	2	3	3	2	12
Team C	888	190	931	0.268	154	1	1	1	1	1	5

Team A has a one-point lead over Team B, and only the average and stolen base categories are undecided. Team B is secure in first in average and needs only four additional steals to pass Team A. Similarly, Team C can potentially catch Team A in average but is too far behind in the other categories to make up ground in the final week.

With such a simple example, it is clear to see that Team B should trade its players who hit for average to Team C in exchange for players who steal bases. A Billy Butler for Michael Bourn swap would make sense. Butler and Bourn are similarly valued players—they went about two rounds apart in ESPN standard ADP in 2013—who are about as safe as possible in the relevant categories.

But what if Team C does not have Michael Bourn? What if its stolen bases came from players like Alcides Escobar and Cameron Maybin? Given the context, it does not matter at all. Neither Escobar nor Maybin provides the run production that makes Bourn a sixth-rounder, but because runs are already decided, they have no value in context. The only cause for preference of Bourn over either Escobar or Maybin is in the expectation for additional steals.

More than anything else, pride is what holds us back from making similar trades. Draft results are a reflection of expected value of players as of draft day and should be discarded as a rubric as soon as the season starts, but it is difficult to do so. If I drafted Billy Butler in the fourth round, it will feel like a loss to trade him for a player drafted more than a couple of rounds lower, but context can turn a trade of Billy Butler for Alcides Escobar into a fantasy title.

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2013-04-18T07:04:15+00:00 Creating your own platoons http://www.hardballtimes.com/main/fantasy/article/creating-your-own-platoons/ http://www.hardballtimes.com/main/fantasy/article/creating-your-own-platoons/#When:06:45:15 here, here, and here, and Brad Johnson outlined some predictable real-baseball platoons that you can replicate in fantasy here. Teams like the Rays and the Athletics make it easy. Half of their lineups consist of platooned players, and so if you are looking for that advantage in leagues with daily lineups and sufficient bench spots, you can just mirror their starting lineups in your own. However, just because many other teams do not employ platoons does not mean that you cannot create fantasy platoons of their players.

Last season, right-handed pitchers started 68 percent of all games. As an example, let’s say that I have two batters from two different teams that I wish to platoon against right-handed pitchers. To make the estimate easy, let’s assume that each hitter plays on the same days as the other but in different games, Game A and Game B. If we assume that the handedness of pitchers in each game on a given day are independent events, then the probability that either or both hitters will face a right-handed starter equals one minus the probability that neither hitter faces a right-handed starter:

P(A(RHP) or B(RHP)) = 1 – P(A(LHP) and B(LHP) = 1 – (0.32 * 0.32) = 0.90

With two batters, I have a 90 percent chance of having at least one of the two enjoying a favorable matchup against a right-handed pitcher. If my bench is really deep, I could even add a third platoon option that plays in Game C:

P(A(RHP) or B(RHP) or C(RHP) = 1 – P(A(LHP) and B(LHP) and C(LHP)) = 1 – (0.32 * 0.32 * 0.32) = 0.98

With three batters, I have a 98 percent chance of having at least one hitter facing a right-handed starter.

The reason such a platoon can be so effective in fantasy is that owners frequently overspend relative to expected production on hitters versus pitchers. There are several reasons that makes some sense. First, pitchers are more likely to suffer an injury. Second, pitchers are easy to stream to take advantage of match-ups against poor offensive opponents and pitcher’s parks. Still, if everyone in your league shares that mentality, elite hitters will have severe inflation, and you will need to play backward, so to speak, to capitalize.

This exact scenario played out for me before the 2012 season in the inaugural draft of my Ottoneu linear weights league. For those who are unfamiliar with the format, linear weights attempt to assign points for events that match their value in terms of run expectancy in real baseball. But, really, the specifics of the scoring don’t matter. All you need to platoon hitters is a league with daily lineups.

In last year’s draft, elite hitters were selling for 10 percent and more above my price sheet. Some owners believe that you should allow each auction to set tier prices and then look for relative bargains in the tiers based on your own preferences. I do not. To me, my price sheet represents the fair value of every player I deem is above the replacement level based on the league format. Every dollar spent over my prices or spent on a player not on my sheet represents a dollar of discount I can capture in another player. For that reason, I am particularly susceptible to a pitcher-heavy team because I always let the preferences of other owners dictate my budget allocation.

In this case, I ended up with a staff of Felix Hernandez, Cliff Lee, CC Sabathia, and David Price and a bullpen of Craig Kimbrel, Aroldis Chapman, Sean Marshall, Kenley Jansen, Glen Perkins, Chris Perez, and Jim Johnson for basically half of my $400 budget. I kept all of those pitchers except for Johnson, who I traded in-season. However, I lost my one impact outfielder, Adam Jones, to arbitration—Ottoneu allows owners to vote off a player from each team, and he was best value selection in 2012. I was left with a handful of inexpensive outfield keepers in Shane Victorino, Garrett Jones, and Tyler Colvin, a skeleton crew of players with clear splits. With most of my money tied up in keepers at other positions and with plenty of bench slots, an outfield platoon was clearly my best bet.

For a platoon to be worthwhile, it needs to replicate the production of starter-caliber players, trading bench spots for the cost savings. However, those costs savings only exist if a player’s platoon value is hidden. Sure, you can platoon a couple of players with small splits, but those players tend to be more expensive since they can be used every day. The value picks tend to either platoon in real life or to have more dramatic splits that depress their overall numbers.

I wanted to find some outfield candidates to target, and so I calculated the points per plate appearance for batters versus both left and right-handed pitchers from 2010-2012. All of the at-bat events were simple to handle, and I distributed stolen bases and caught stealings based on the ratio of plate appearances against each pitcher hand. I also included a minimum of 200 plate appearances versus right-handed pitchers over that time frame.

I cannot simply sort by points per plate appearance versus right-handed pitchers and pick out the best ones because, even though I plan to use these platoon players only in games started by right-handed pitchers, they will still face some number of left-handed relievers in those games. Therefore, I first calculated how often left-handed and switch-hitters faced left and right-handed pitchers in games started by a right-hander in 2012.

Fortunately, the ratios were fairly consistent for all players. On average, a left-handed hitter faces a right-handed pitcher in 88.7 percent of his plate appearances and a switch-hitter does the same in 90.5 percent of his plate appearances in a game started by a right-handed pitcher.

Using those two averages, I calculated the combined points per plate appearances for left-handed and switch-hitters in games started by a right-handed pitcher based on their production facing each handed pitcher from 2010-2012. I also included a column of the discrepancy between points per plate appearances versus right-handed and left-handed pitchers. Here are some interesting names, bookended by the overall points per game of select players.

Tier 1 Points Per Plate Appearance, 2010-2012:

Player	Hand	LHP PA	RHP PA	PointsPerPA	LHP PointsPerPA - RHP PointsPerPA
Mike Trout				1.68
Justin Morneau	L	347	571	1.66	0.69
Shin-Soo Choo	L	458	874	1.62	0.87
Brandon Moss	L	63	260	1.60	0.51
Joe Mauer	L	453	772	1.57	0.56
Andre Ethier	L	417	786	1.57	0.91
Carl Crawford	L	266	522	1.53	0.52
Buster Posey				1.51

As you can tell based on the bookends of Mike Trout and Buster Posey, these guys are the elite platoon options. As such, all of them play every day in real baseball and will fetch the price of their non-platoon peers in an auction. That said, both Morneau and Moss are fairly cheap in most leagues—though keep in mind Moss has the smallest sample—and Choo and Ethier may be cheap enough to deploy in a platoon, where they go from solid to elite.

Tier 2 Points Per Plate Appearance, 2010-2012:

Player	Hand	LHP PA	RHP PA	PointsPerPA	LHP PointsPerPA - RHP PointsPerPA
Andrew McCutchen				1.46
Luke Scott	L	209	652	1.45	0.66
Anthony Rizzo	L	107	261	1.43	0.80
Wilson Betemit	B	184	507	1.39	0.56
Pedro Alvarez	L	281	691	1.34	0.54
Matthew Joyce	L	132	591	1.33	0.80
Lance Berkman	B	112	466	1.31	0.78
John Jaso	L	111	654	1.30	0.87
Garrett Jones	L	311	858	1.30	0.65
Ryan Doumit	B	309	675	1.29	0.57
David DeJesus	L	222	754	1.28	0.65
Pablo Sandoval				1.27

For the most part, the next set of players are fairly inexpensive. I was able to buy Joyce and keep Jones at a combined $7, and I would have done more if more of the names on this list were outfielders. For owners in two-catcher leagues, take note that Jaso, Doumit, and Jarrod Saltalamacchia of the third tier could be combined into an above-average platoon for very little cost.

Tier 3 Points Per Plate Appearance, 2010-2012:

Player	Hand	LHP PA	RHP PA	PointsPerPA	LHP PointsPerPA - RHP PointsPerPA
Chase Headley				1.26
Seth Smith	L	135	704	1.24	0.98
Will Venable	L	135	780	1.23	0.51
Eric Chavez	L	49	386	1.22	1.22
Jason Kipnis	L	241	431	1.21	0.58
Adam Lind	L	241	725	1.20	1.08
Jarrod Saltalamacchia	B	76	402	1.15	0.65
Dewayne Wise	L	83	274	1.14	0.51
Derek Jeter				1.14

Tier 3 is a bit less useful from a practical standpoint, but not because of their lesser point totals. These guys have still outproduced players like Derek Jeter of the past three seasons per plate appearance in games started by a right-hander. However, Smith and Venable are in real-baseball platoons, which make them obvious to other owners that want fantasy platoons. For players in deeper leagues or in AL- or NL-only leagues, DeWayne Wise and Eric Chavez may be worth a buck or two at the end of an auction. If they find the playing time, they can be useful players versus right-handers.

In the end, I netted Garrett Jones, Tyler Colvin, Matt Joyce, and Will Venable for a total of $11 to cover my fourth and fifth outfield spots. As a basis of comparison, Nelson Cruz sold for $11. I had to forfeit some bench slots I could have used on prospects or pitchers with upside, but with an elite staff and a team ready to win now, I believe it will be worthwhile.

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2013-03-28T06:45:15+00:00 What MLB can learn from Kyle Lohse and the NFL http://www.hardballtimes.com/main/article/what-mlb-can-learn-from-kyle-lohse-and-the-nfl/ http://www.hardballtimes.com/main/article/what-mlb-can-learn-from-kyle-lohse-and-the-nfl/#When:07:01:15 Kyle Lohse remains unsigned, a pioneer of the unintended consequences of the new collective bargaining agreement. Prior to the new agreement, the better free agent players were classified as Type A or Type B based on their previous production levels. If another team signed a Type B player from your team, you were awarded a compensatory pick. If another team signed a Type A player from your team, you were awarded both a compensatory pick and the signing team’s best remaining draft pick.

Once teams realized just how much value over cost successful draft picks could provide on their rookie contracts, they changed their behavior. Based on the recent history of some of the savvier teams, it was unclear whether the true prizes of each offseason were the star players or the compensation for the teams that lost them.

Of course, the players were not so thrilled by the disincentives to free agent signings that were pushing down star player contracts. In fact, the new free agent compensation model seems to target preventing situations like the one facing Lohse. It simply missed the mark.

With the old model, players were designated as Type A and Type B free agents based on their statistics, some of which were less reflective of their actual value than others.

The easy example is with closers. A reliever’s ability to earn saves is at the discretion of his manager, and so it would not be unreasonable for a team to ignore saves when deciding among the free agent relievers, some of whom had closed and some of whom had not. However, since saves were an input of the type classification, previous closers were far likelier to be designated Type A and, therefore, be more expensive because of the loss of a draft pick.

If Lohse had been where he is now a couple of years ago, he would likely face a similar situation. Traditional statistics show that Lohse has been an elite pitcher. Last year, he went 16-3 with a 2.86 ERA, the eighth-lowest of qualified starters. He would have been a Type A free agent.

However, Lohse has dramatically outperformed his peripherals. Over the previous two seasons, Lohse has had an ERA well below his FIP—by nearly a full run in 2012—and, even last year, his strikeout rate was a pedestrian 16.6 percent. If a team expected Lohse to see his results match his peripherals going forward, he may not be worth the loss of the draft pick required to sign him, especially if he demands a salary comparable to that of other pitchers of his skill level.

The introduction of qualifying offers was meant to eliminate the problem. Assessing the value of players is no longer as simple as it was 20 years ago, and the gap between public and private understanding will only increase as salaries do and teams invest more heavily in research and scouting. Qualifying offers shift the responsibility of valuing players back to the teams, and so those choices will always mirror the most sophisticated opinions on their value.

Had the new free agent compensation system been the only change to the new collective bargaining agreement, it might have accomplished what it intended. However, many of the other changes in the agreement took away from the ability of teams to build their farm systems, and that had a spillover effect on free agency.

Before the new agreement, teams had several ways to acquire players with a value proposition equal or similar to that of a first-round selection. They could use their own picks. They could use the draft picks of teams that signed their Type A free agents and the between-round compensation picks for Type A and B free agents. They could pay over slot on their later round selections to entice the hard-to-sign players with first-round talent. They could invest heavily in international signings.

Now, many of those options are gone. Teams are subject to severe penalties that prevent them from consistently going over slot and spending heavily on international players. Even for players extended a qualifying offer, the first-round pick that the signing team loses is lost to the ether, leaving the original team with only a compensatory pick. A team’s annual performance-based first-round selection is practically its only option to acquire a first-round talent. Of course, they are reluctant to part with one to acquire a free agent like Lohse who will demand a salary in line with his skill.

As a result, Lohse is in the worst kind of limbo. He is good enough that a team like St. Louis has no qualms about extending him a qualifying offer that would guarantee him around $13.3 million, but only for one year. But he is not good enough that another team will want to lose a first-round pick unless it can sign him for several years at what he would perceive to be a major discount because of his production level.

Lohse is stuck with limited options. He can probably find a multi-year contract at a steep discount compared to his expectations, or he can sign a one-year deal with some kind of clause that prevents his new team from extending him a qualifying offer. Either way, Lohse will be underpaid, and the players are back where they were two years ago, stuck with a collective bargaining agreement with disincentives that reduce free agent contracts.

So what can the players do about it? Owners are happy because the new agreement put in place mechanisms to cap spending in both player development and in free agency. If the players want to push for limitations or escalators on qualifying offers, similar to the franchise tag in the NFL, owners will want concessions. I think their best bet is another approach.

With both the previous and the current free agent compensation models, teams benefit from clumping their signings. The Yankees provided the blueprint in 2008. That offseason, they signed three Type A free agents in CC Sabathia, Mark Teixeira and A.J. Burnett.

If you have a first-round pick and you sign a Type A or qualifying offered free agent, you lose that first-round pick. Then, your best pick is a second-round pick, and so if you sign another Type A or qualifying offered free agent, you lose only that second-round pick. With each additional signing, the draft pick lost becomes less severe. That is why the Indians seem like the best fit for Lohse, since they already expended their top two picks when they signed Nick Swisher and Michael Bourn.

However, it is not a coincidence that the Yankees were the team with the blueprint. Not many organizations can afford to sign several of the most expensive free agents in a single offseason. Even though the Indians can acquire Lohse for a smaller cost than any other team, if he pushed them over budget, that lesser cost still would be prohibitive.

To me, the potential solution that everyone should agree on is the ability to trade draft picks. Then, the Indians could sign Lohse for a contract more in line with his expectations and then trade him to another team for a draft pick in between the first-round pick that other team would have had to forfeit if it had signed him and the lower-round pick the Indians will lose because of their earlier signings of Swisher and Bourn.

Hypothetically, the Indians could sign and trade Lohse now—assuming Lohse would agree to be traded—for a package of prospects that approximate that middle ground. But allowing teams to trade actual draft picks would greatly increase the chances the Indians could find a match in a team that wants Lohse and could provide that fair return.

Players would win with better contracts. Teams—even small market ones—would have an incentive to sign free agents. Front offices would gain a new tool to help them stockpile draft picks. And owners would win because trading picks would create interest in the draft, which the NFL has proven can be a powerful marketing and revenue-generating event.

With the current agreement, the first step to trading draft picks is already in place. One of the penalties for going over slot is a loss of draft picks that then are awarded to small-market teams, and those picks—and only those picks—are tradable. For the five-year period of the current agreement, that could act as a trial run, and the next step could be incremental. To potentially counteract the free agent disincentive, tradable picks in the first three rounds would be sufficient. If that is effective and entertaining, then it can expand from there.

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2013-03-06T07:01:15+00:00 Daisuke Matsuzaka’s wild new contract http://www.hardballtimes.com/main/article/daisuke-matsuzakas-wild-new-contract/ http://www.hardballtimes.com/main/article/daisuke-matsuzakas-wild-new-contract/#When:08:11:15 Daisuke Matsuzaka stateside. Over the weekend, the Indians signed Matsuzaka to a minor league deal with incentives that could reach $4 million, quite a dramatic fall.

The narrative for me with Matsuzaka was his apparent unwillingness to throw strikes. In both the 2006 and 2009 World Baseball Classics, Matsuzaka won the MVP and led Japan to the title. There, he would challenge hitters, and his stuff was such that he dominated them. He had the same stuff with the Red Sox, but whether by confidence or mismanaged routine or some other unaired inhibiter, he seemed to always fall behind hitters.

As a fan, it is much easier to pull for a player who maximizes his natural abilities. Most fans had no chance at the big leagues from the day they were born, and so I believe empathy comes mofre easily for the perceived hard workers.

As with J.D. Drew, there was little empathy for Matsuzaka. Here was the pitcher who was supposed to become the next franchise ace, constantly with full counts, constantly with bases loaded, constantly out before the sixth inning and with no obvious reason for any of it. I can just picture Red Sox nation in a collective outburst of “Just throw strikes!”

Because of that perception, it never seemed appropriate to write Matsuzaka off. The cliché in baseball is that once you display a skill, you own it, and he had certainly displayed the total package in the WBC. Even now, when I first read about his minor league contract, I couldn’t help but wonder at his sleeper potential.

If for no other reason than to help me sleep better at night, I wanted to look at the data to see if my perception of Matsuzaka matched reality. The crux of that perception is my belief that Matsuzaka (1) routinely fell behind hitters but (2) had the command to throw strikes more often had he wanted to.

First, I used a PITCHf/x database to look at first-pitch ball percentage—which I calculated as a percentage of all pitch results, including balls-in-play—of pitchers who have faced at least 1,000 batters since 2007. At 41.3 percent, Matsuzaka came in 157th of 369 total pitchers, which is more toward the wild side but still less egregious than I expected.

Next, I calculated the ball percentage of those pitchers when they were in three-ball counts, excluding 3-0 counts, when pitchers typically attempt to throw a strike at near-complete expense of making a quality pitch. Here, Matsuzaka made it inside the top-30 with 28.7 percent of three-ball count pitches resulting in balls. The top of that list is the who’s who of wild pitchers from the last few years, including Oliver Perez, Carlos Marmol, Andrew Miller, Dontrelle Willis, Daniel Cabrera and Edinson Volquez.

Highest ball percentage for pitchers in 3-1 and 3-2 counts since 2007:

Player	3 Ball Count Ball%
Clay Hensley	34.9%
Robinson Tejeda	33.5%
Rich Hill	33.4%
Oliver Perez	31.9%
Manny Parra	31.4%
Javier Lopez	31.0%
Carlos Marmol	30.9%
Andrew Miller	30.9%
Dontrelle Willis	30.8%
Daniel Cabrera	30.6%
J.P. Howell	30.5%
George Sherrill	30.4%
Chris Perez	30.4%
Jonathan Sanchez	30.4%
Rafael Perez	30.0%
Edinson Volquez	29.8%
John Grabow	29.6%
Ricky Romero	29.6%
Dustin McGowan	29.5%
Ian Snell	29.5%
Jeff Samardzija	29.4%
Kip Wells	29.3%
Daisuke Matsuzaka	28.7%
Joel Hanrahan	28.6%
Greg Smith	28.6%
Chien-Ming Wang	28.4%
Joe Smith	28.4%
A.J. Burnett	28.3%
Garrett Olson	28.2%
Dillon Gee	28.1%

Interestingly, Matsuzaka has the third lowest first pitch ball percentage of those 30 names, behind only Javier Lopez and George Sherrill. In fact, if you sort the full list of 369 pitchers by the difference in ball percentage from 0-0 counts to 3-1 and 3-2 counts from smallest to largest, Daisuke, again, makes the top-30.

Smallest difference between ball percentage in 0-0 counts and 3-1 and 3-2 counts since 2007:

Player	First Pitch Ball%	3 Ball Count Ball%	Difference
Javier Lopez	38.6%	31.0%	7.6%
Robinson Tejeda	42.8%	33.5%	9.3%
George Sherrill	40.0%	30.4%	9.6%
Clay Hensley	44.7%	34.9%	9.8%
Jason Frasor	37.2%	26.8%	10.3%
Erik Bedard	37.9%	27.4%	10.5%
Joe Beimel	36.5%	26.0%	10.5%
Glen Perkins	36.8%	26.2%	10.6%
Johan Santana	35.9%	25.2%	10.7%
Rich Hill	44.4%	33.4%	10.9%
Roy Oswalt	35.0%	24.0%	11.0%
Micah Owings	39.1%	28.0%	11.1%
Chris Perez	41.6%	30.4%	11.2%
Brandon Webb	36.8%	25.4%	11.4%
Colby Lewis	35.2%	23.8%	11.4%
Kris Medlen	35.6%	24.1%	11.5%
Brad Thompson	37.0%	25.4%	11.6%
Clay Buchholz	39.6%	27.9%	11.7%
Chris Young	38.8%	27.1%	11.7%
Nate Robertson	39.7%	27.8%	12.0%
Jeff Karstens	36.9%	24.9%	12.0%
Rodrigo Lopez	35.6%	23.5%	12.1%
Jeff Samardzija	41.5%	29.4%	12.2%
Jair Jurrjens	38.6%	26.4%	12.2%
J.P. Howell	42.7%	30.5%	12.2%
John Grabow	41.9%	29.6%	12.2%
Manny Parra	43.8%	31.4%	12.4%
Chris Capuano	36.6%	24.0%	12.5%
Daisuke Matsuzaka	41.3%	28.7%	12.7%
Jonathan Sanchez	43.1%	30.4%	12.7%

The results suggest that I was wrong about Matsuzaka. There was never more to his tendency to fall behind in counts than his overarching inability to throw strikes. In that context, his new contract with the Indians seems entirely appropriate for a pitcher with a history of poor command and arm problems.

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2013-02-11T08:11:15+00:00 Swings and misses http://www.hardballtimes.com/main/article/swings-and-misses/ http://www.hardballtimes.com/main/article/swings-and-misses/#When:08:56:15 Josh Hamilton signed a five-year, $123 million contract with the Los Angeles Angels about a month ago, I did not know what to make of it. Of the available free agents, Hamilton had the best offensive season in 2012, but given his history of substance abuse and his streakiness on the field, how does one predict what Hamilton will be in 2013, never mind in 2017?

Hamilton defies comparison. On the field, I’ve always thought of him as the second-coming of Vladimir Guerrero, but while the two players share raw tools, plate discipline tendencies, and a presence that I can only think to describe as swashbuckling that always pulls my eyes to the television screen, Guerrero had a dependable greatness that Hamilton seems only to reach a few months at a time.

The seed of that comparison is their penchant to chase pitches. From 2007-2011, Guerrero swung at 44.0 percent or more of pitches thrown outside of the strike zone each season. Hamilton crossed that threshold for the first time in 2012 with a 45.4 percent O-Swing%, but he had not fallen too far short in 2011 at 41.0 percent.

Guerrero probably did not chase balls quite so frequently in his prime—FanGraphs has O-Swing% numbers that date back to 2002, and Guerrero was closer to 30 percent than 40 percent at that point—but his non-existent strike zone will always be in the first sentence of his legacy. And while Guerrero never saw a pitch he did not like, his lack of discipline belied his incredible consistency as a hitter.

Guerrero had an 11-year run from 1998-2008 in which he never fell short of 129 wRC+, and he never struck out more than 14.0 percent of his plate appearances in a season. Guerrero was a prodigy of see ball, hit ball, which manifested in a career contact rate of 67.4 percent on pitches outside the strike zone, which put him in the top-third in baseball most seasons.

In contrast, Hamilton has always struck out more than 14.0 percent of his plate appearances. For his career, he falls just short of 20.0 percent. That has not prevented him from having success, which may be why Hamilton has become more willing to chase pitches out of the strike zone. His O-Swing% has increased every season from 2007 to 2012. In 2012, he chased at nearly double his rate in 2007.

Josh Hamilton’s plate discipline by season:

Season	O-Swing%	O-Contact%
2007	32.5%	61.7%
2008	36.5%	62.1%
2009	37.6%	59.0%
2010	35.9%	63.1%
2011	36.7%	61.0%
2012	42.3%	49.4%
Total	37.8%	58.3%

His contact rate rose in step with his increased aggressiveness prior to 2012, but when that cratered to a career-low 49.4 percent O-Contact% this season, it left Hamilton with a very peculiar distinction.

Hamilton led all qualified hitters in swing percentage on pitches out of the zone, and he also had the third-highest swing-and-miss rate on pitches out of the zone. Since 2002, only one other player has had a season in the top-10 in both categories, and he was also in the top-5 in both of them. That player is Miguel Olivo in 2011.

Hitters in the top-10 in O-Swing% and O-Miss%, 2002-2012:

Season	Name	O-Swing%	O-Miss%	O-Swing% Rank	O-Miss% Rank
2012	Josh Hamilton	45.4%	47.9%	1	3
2011	Miguel Olivo	45.6%	48.8%	3	4

Let that sink in for a moment. Hamilton may be the best hitter in baseball and Olivo may be the worst. In 2011, Olivo had a wRC+ of 75, making him 25 percent worse than a league-average hitter. In 2012, Hamilton had a wRC+ of 140, making him 40 percent better than a league-average hitter. Hamilton’s batting average in 2012, .285, was 32 points higher than Olivo’s on-base percentage in 2011, .253. And yet, their two seasons share in this unusual distinction.

So what does it mean for Hamilton’s success as an Angel? I don’t know. Since both his chase rate and his contact rate were career extremes, perhaps his plate discipline will return to the more-typical numbers of a few years ago. That is probably what the Angels hoped for when they signed him. Either way, Hamilton’s profile mirrors his career and life trajectory in its strangeness. Even if he had never suffered his addictions, he would still be impossible to predict.

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2013-01-22T08:56:15+00:00 The Padres had the wrong team for their old ballpark http://www.hardballtimes.com/main/article/the-padres-had-the-wrong-team-for-their-old-ballpark/ http://www.hardballtimes.com/main/article/the-padres-had-the-wrong-team-for-their-old-ballpark/#When:09:09:15 Michael Bourn, for example—but, for the most part, teams have a pretty good idea of what they’ll look like on Opening Day. This winter, Los Angeles has been the city of loose purse strings, as the Dodgers and Angels have combined to spend $272 million on Zack Greinke and Josh Hamilton, alone. But, 120 miles south, the biggest addition to the San Diego Padres has not been a player, but rather the new fences in right and left-center field.

Petco has earned its reputation as one of the pitcher-friendliest parks in baseball. Its 2012 Park Factor for runs scored was 0.854, the fifth-lowest in baseball. In particular, Petco inhibits the long ball. Its 2012 Park Factor for home runs was 0.626, third-lowest in baseball. Only Safeco Field in Seattle, at 0.583, and AT&T Park in San Francisco, at 0.522, surrender fewer home runs relative to other parks than Petco, and no park limits home runs for left-handed hitters more than Petco does.

To me, a ballpark with unusual dimensions has always seemed like an advantage. Every free agent has his market price set by his strengths and weaknesses as they play under average conditions. Hitter A could be a dead pull hitter with plus power, limited defensive range, a good batting eye, and poor baserunning skills. Hitter B could be a tremendous contact hitter with elite speed and range, a great arm, and no power at all. And still, the two players could be similarly valuable on the open market.

To specific teams, however, those two players could have dramatically different values based solely on their home parks. Matt Cain is the perfect example. Cain has thrown at least 190 innings the last seven seasons. In that time, he has never allowed more than 8.4 percent home runs per fly ball, and he has only once allowed more than 40 percent ground balls. Cain is an extreme flyball pitcher in a park that limits home runs, a park factor that greatly benefits Cain’s particular skill set. He would be valuable on the open market, of course, but Cain has a career ERA of 3.27, nearly a run lower than his career xFIP of 4.19. Chances are, Cain would perform worse in any other ballpark.

There are not many Matt Cains who reach free agency, but one would expect the Padres to sign players and build their farm system with their park in mind. With so spacious an outfield, the Padres could reap more benefit from outfielders with great range, groundball and line drive hitters, and flyball pitchers than other teams. Their park provided them with an opportunity for a competitive advantage, but they have not capitalized.

Last season, the Padres’ starting pitchers had a groundball rate of 49.5 percent, the highest in baseball. Some of that was bad luck. Cory Luebke had a 35.6 percent groundball rate in 2011, the lowest of San Diego’s five starters who reached 100 innings. He also led the staff with a 3.16 FIP. He threw just 31 innings in 2012 before needing Tommy John surgery.

Still, some of it was planned. They traded away Mat Latos and his 43.7 percent groundball rate. That was probably a smart move for a team that was not contending and needed prospects, but they then filled his void with Edinson Volquez, whose groundball rate exceeded 50.0 percent in 2010 and 2011 on the Reds and then in 2012 on the Padres. In the Adrian Gonzalez trade, they acquired Casey Kelly, whose fastball has a natural sink that led to a 55.8 percent groundball rate in limited innings last season. This winter, the Padres re-signed Jason Marquis, who, like Volquez, has seen groundball rates well above 50.0 percent the last three seasons. Rumors have them chasing Rick Porcello, who has never allowed more fly balls than ground balls in a season.

The square-peg-round-hole strategy has not been limited to their rotation. According to Baseball Info Solutions, Carlos Quentin has cost his teams 20 runs because of his poor range in close to 2,500 innings in left field in his career, and his 46.7 percent flyball rate was ninth-highest among batters with at least 300 at-bats in 2012. San Diego had a payroll of $55.6 million in 2012. With the three-year, $27 million extension Quentin signed, he will make close to one sixth of their entire payroll in 2013, if the team does little else this offseason.

Bringing in the fences for 2013 will probably benefit the Padres as they are currently constructed. I’m sure Chase Headley, who hit 18 of his 31 home runs away from Petco, will enjoy the changes. I just wonder if the Padres would have been better off to embrace the uniqueness of their park and build a team that complements it. Over the last three seasons, the Padres are 122-121 at home and 115-128 on the road. That seven-win disparity is the fifth-lowest in baseball. Extreme dimensions should help a team win at home, especially a team with limited financial resources. It is a small thing, but small-market teams need every advantage they can find.

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2013-01-04T09:09:15+00:00 The price of positional flexibility http://www.hardballtimes.com/main/article/the-price-of-positional-flexibility/ http://www.hardballtimes.com/main/article/the-price-of-positional-flexibility/#When:08:29:15
On paper, the team seemed poised for success. Boston found the worst possible time to sink into a slump in 2011, and that made the team seem so much worse than the final standings showed them to be. Andrew Bailey, Cody Ross, and Ryan Sweeney were not splashy acquisitions, but it truly appeared the Red Sox were not far away. Before the 2012 season, ZiPS liked them for second in the AL East with a better than six in 10 chance of reaching the playoffs. They finished last, 24 games out of the Wild Card.

Tampa Bay fell short of the playoffs, as well, but the 21-game discrepancy between the Rays and the Red Sox did not seem to fully capture the differences between their cultures. While the Boston players and manager traded insults through the media, the Rays played dress up on their road trips. Neither team made it to October, but for Tampa, that seems wholly temporary. It was misfortune, not dysfunction.

On the field, the problems the Rays and Red Sox had in 2012 were remarkably similar. In 2011, Jacoby Ellsbury had one the best seasons any center fielder ever had, leading baseball with 9.1 WAR, but Evan Longoria had an excellent year himself with 6.1 WAR. Each player was hurt in 2012 and held to 74 games played.

Their top 2011 stars were just the start, as the Red Sox and Rays at times lost Dustin Pedroia, David Ortiz, Carl Crawford, Ross, Kevin Youkilis and later Will Middlebrooks, B.J. Upton, Desmond Jennings, Matt Joyce, Luke Scott, Jeff Keppinger, and Sean Rodriguez. And that’s just the hitters. Both teams were near the top of the list of most days lost to the DL for their hitters in 2012. Where it buried the Red Sox, it only inconvenienced the Rays, and I do not believe culture is the entire explanation.

Jacoby Ellsbury produced 1.5 WAR in a little less than half a season in 2012. In the same time, Evan Longoria produced 2.4 WAR. Those totals are reasonable estimates of the wins their teams lost in their time without them, but that does not account for the context of their actual team compositions. The Red Sox pieced together their center field without Ellsbury with the likes of Marlon Byrd, Scott Podsednik, and Ryan Kalish. Those three players combined for -1.3 WAR for the season. The Red Sox continued to throw new center fielders at the problem, but none of them stuck.

In contrast, the Rays did not scour the minors or the waiver wire for a replacement. They employed a bit of creativity. Keppinger split time at second base with Ben Zobrist in 2011, but he had played a bit of third base in stops with the Astros, Reds, and Royals, and Zobrist could man second. The Rays could not have predicted that Keppinger would have his best offensive season since 2007, but they did have a roster full of players that could play multiple positions and so were built to take advantage when he did.

The trade market was thin at third base, but Tampa did not need a third baseman. The Rays needed a third baseman or a shortstop or a second baseman or a right fielder, and so they did not have to overpay or overextend themselves. They traded a non-prospect for Ryan Roberts in the end, but he was one of many options the Rays had, which was many more than the Red Sox had in center field.

Catchers are usually mentioned as the great shortcoming of WAR. We have just begun the attempt to measure all of the little things a catcher can do to make his teammates better. I believe Zobrist deserves the same consideration.

From 2009 to 2012, Zobrist has the third-most WAR of all position players, but I am not sure those 25.1 Wins Above Replacement describe his full contributions. Zobrist allows the Rays to play the platoon advantage and to overcome injuries to players at multiple positions. The Red Sox had depth in 2012, but they did not have the right depth to address their specific problems. With Zobrist, Keppinger, and their other players with positional flexibility, the Rays were prepared for all kinds of problems.

The Red Sox have signed Mike Napoli and Shane Victorino to matching three-year, $39-million contracts. On the surface, those contracts do not make a lot of sense. Napoli has played catcher and DH in his career, two of the more crowded positions on the Red Sox’s roster. Now he will try first base. Victorino may have lost his peak bat, and with Ellsbury in center, his potential value is diminished in right field. Many have speculated that Victorino and Napoli portend imminent trades of Ellsbury and either Jarrod Saltalamacchia or Ryan Lavarnway. I wonder.

The Red Sox wanted a culture change. If the new manager and the lack of long-term contracts are evidence of their new course, I would speculate that Victorino and Napoli are evidence, as well. Neither player is a likely All-Star, but each is versatile. Might these signings allow Boston the position flexibility it will need to cover for potential injuries? If 2012 was an indication, such flexibility would serve the team well.

The Red Sox have a checkered history with their plans for success, but success in baseball has always been more about how you deal with failure. Last season, the Red Sox relied on Ellsbury and Ortiz, and when they lost them, all was lost. This season, they will hope to have them, but now they may not require them to find success.

Click here to learn about THT's download subscriptions.]]> Scott Spratt 2012-12-10T08:29:15+00:00