Hardball Retrospective – What Might Have Been – The “Original” 1969 Reds

In “Hardball Retrospective: Evaluating Scouting and Development Outcomes for the Modern-Era Franchises”, I placed every ballplayer in the modern era (from 1901-present) on their original team. I calculated revised standings for every season based entirely on the performance of each team’s “original” players. I discuss every team’s “original” players and seasons at length along with organizational performance with respect to the Amateur Draft (or First-Year Player Draft), amateur free agent signings and other methods of player acquisition.  Season standings, WAR and Win Shares totals for the “original” teams are compared against the “actual” team results to assess each franchise’s scouting, development and general management skills.

Expanding on my research for the book, the following series of articles will reveal the teams with the biggest single-season difference in the WAR and Win Shares for the “Original” vs. “Actual” rosters for every Major League organization. “Hardball Retrospective” is available in digital format on Amazon, Barnes and Noble, GooglePlay, iTunes and KoboBooks. The paperback edition is available on Amazon, Barnes and Noble and CreateSpace. Supplemental Statistics, Charts and Graphs along with a discussion forum are offered at TuataraSoftware.com.

Don Daglow (Intellivision World Series Major League Baseball, Earl Weaver Baseball, Tony LaRussa Baseball) contributed the foreword for Hardball Retrospective. The foreword and preview of my book are accessible here.


OWAR – Wins Above Replacement for players on “original” teams

OWS – Win Shares for players on “original” teams

OPW% – Pythagorean Won-Loss record for the “original” teams

AWAR – Wins Above Replacement for players on “actual” teams

AWS – Win Shares for players on “actual” teams

APW% – Pythagorean Won-Loss record for the “actual” teams


The 1969 Cincinnati Reds 

OWAR: 59.0     OWS: 355     OPW%: .619     (100-62)

AWAR: 37.4      AWS: 267     APW%: .549     (89-73)

WARdiff: 21.6                        WSdiff: 88  

The “Original” 1969 Reds outdistanced the Giants by a fourteen-game margin to secure the National League pennant. Pete Rose (.348/16/82) aka “Charlie Hustle” led the NL with 120 runs scored and registered personal-bests in home runs, RBI, batting average, OBP (.428) and SLG (.512). “The Toy Cannon”, center fielder Jim Wynn swatted 33 big-flies, nabbed 23 bags and tallied 113 runs. Completing the outfield trio with 30+ Win Shares, Frank “The Judge” Robinson crushed 32 long balls and knocked in 100 baserunners while posting a .308 BA.

The Cincinnati infield, with the exception of second-sacker Tommy Helms, produced 23+ Win Shares each. Tony “Big Dog” Perez (.294/37/122) manned the hot corner while the “Big Bopper”, Lee May (.278/38/110) earned his first All-Star assignment over at first base. Leo “Mr. Automatic” Cardenas (.280/10/70) provided a steady bat at shortstop. “Little General” Johnny Bench (.293/26/90) delivered an encore to his 1968 NL Rookie of the Year campaign. The Reds’ reserves featured the fleet-footed Cesar Tovar (.288, 45 SB) and Tommy Harper (73 SB) along with seven-time Gold Glove Award-winning center fielder Curt Flood.

Bench ranked second behind Yogi Berra at catcher in the “The New Bill James Historical Baseball Abstract” top 100 player rankings. “Original” Reds teammates enumerated in the “NBJHBA” top 100 rankings include Frank Robinson (3rd-RF), Pete Rose (5th-RF), Jim Wynn (10th-CF), Tony Perez (13th-1B), Vada Pinson (18th-CF), Curt Flood (36th-CF), Lee May (47th-1B), Leo Cardenas (50th-SS), Johnny Edwards (53rd-C), Tommy Harper (56th-LF), Cookie Rojas (69th-2B), Cesar Tovar (79th-CF), Tony Gonzalez (82nd-CF) and Tommy Helms (99th-2B).

  Original 1969 Reds                                                                     Actual 1969 Reds

Frank Robinson LF/RF 5.31 31.84 Alex Johnson LF 2.86 18.84
Jim Wynn CF 7.36 36.09 Bobby Tolan CF 4.43 26.52
Pete Rose RF 4.83 36.77 Pete Rose RF 4.83 36.77
Lee May 1B 3.31 25.11 Lee May 1B 3.31 25.11
Tommy Helms 2B -0.93 5.57 Tommy Helms 2B -0.93 5.57
Leo Cardenas SS 2.81 23.74 Woody Woodward SS 0.45 5.83
Tony Perez 3B 5.77 30.41 Tony Perez 3B 5.77 30.41
Johnny Bench C 5.69 29.93 Johnny Bench C 5.69 29.93
Cesar Tovar CF 3.37 20.31 Jimmy Stewart LF -0.1 4.89
Curt Flood CF 2.14 19.71 Ted Savage LF 0.29 3.27
Tony Gonzalez CF 1.89 17.19 Pat Corrales C 0.28 2.82
Tommy Harper 3B 1.78 16.64 Chico Ruiz 2B 0.03 2.68
Art Shamsky RF 2.61 16.22 Darrel Chaney SS -1.23 1.8
Johnny Edwards C 1.94 14.95 Jim Beauchamp LF -0.06 0.99
Vada Pinson RF 0.11 10.97 Fred Whitfield 1B -0.24 0.36
Brant Alyea LF 0.62 6.52 Danny Breeden C -0.1 0.08
Joe Azcue C 0.61 6.49 Bernie Carbo -0.04 0
Don Pavletich C 0.5 4.96 Mike de la Hoz -0.01 0
Chico Ruiz 2B 0.03 2.68 Clyde Mashore -0.01 0
Cookie Rojas 2B -0.66 2.56
Vic Davalillo RF -0.21 2.26
Gus Gil 3B -0.64 1.8
Darrel Chaney SS -1.23 1.8
Len Boehmer 1B -0.91 0.58
Fred Kendall C -0.26 0.31
Bernie Carbo -0.04 0
Clyde Mashore -0.01 0

Claude Osteen (20-15, 2.66) established career-highs with 321 innings pitched, 41 starts, 16 complete games, 7 shutouts and 183 strikeouts. Mike Cuellar (23-8, 2.38) claimed the Cy Young Award and fashioned a personal-best 1.005 WHIP. Jim Maloney contributed a 12-5 mark with a 2.77 ERA as a member of the “Original” and “Actual” Cincinnati rotations. Diego Segui tallied 12 wins and 12 saves to anchor the bullpen. Wayne Granger saved 27 contests in his sophomore season for the “Actuals” and topped the Senior Circuit with 90 appearances.

  Original 1969 Reds                                                                   Actual 1969 Reds

Claude Osteen SP 5.09 24.65 Jim Maloney SP 3.93 14.63
Mike Cuellar SP 4.91 24.57 Jim Merritt SP 0.72 10.63
Jim Maloney SP 3.93 14.63 Gary Nolan SP 1.71 7.02
Casey Cox SP 2.14 12.03 George Culver SP -0.37 3.64
Gary Nolan SP 1.71 7.02 Gerry Arrigo SP -0.29 2.99
Diego Segui RP 1.38 11.3 Wayne Granger RP 1.32 14.75
Dan McGinn RP -0.04 6.86 Clay Carroll RP 1.04 10.09
Jack Baldschun RP -0.3 3.57 Pedro Ramos RP -0.6 1.6
Billy McCool RP -0.04 2.88 John Noriega RP -0.19 0
John Noriega RP -0.19 0 Camilo Pascual SW -0.31 0
Mel Queen SP 0.37 1.17 Tony Cloninger SP -2.26 2.86
Sammy Ellis SP -0.33 0 Mel Queen SP 0.37 1.17
Jose Pena RP -0.68 0 Jack Fisher SP -1.91 0.72
Al Jackson RP -0.23 0.54
Dennis Ribant RP -0.05 0.49
Jose Pena RP -0.68 0
Bill Short RP -0.26 0


Notable Transactions

Frank Robinson

December 9, 1965: Traded by the Cincinnati Reds to the Baltimore Orioles for Jack Baldschun, Milt Pappas and Dick Simpson.

Jim Wynn

November 26, 1962: Drafted by the Houston Colt .45’s from the Cincinnati Reds in the 1962 first-year draft.

Leo Cardenas

November 21, 1968: Traded by the Cincinnati Reds to the Minnesota Twins for Jim Merritt.

Cesar Tovar

December 4, 1964: Traded by the Cincinnati Reds to the Minnesota Twins for Gerry Arrigo.

Claude Osteen

September 16, 1961: Traded by the Cincinnati Reds to the Washington Senators for a player to be named later and cash. The Washington Senators sent Dave Sisler (November 28, 1961) to the Cincinnati Reds to complete the trade.

December 4, 1964: Traded by the Washington Senators with John Kennedy and $100,000 to the Los Angeles Dodgers for a player to be named later, Frank Howard, Ken McMullen, Phil Ortega and Pete Richert. The Los Angeles Dodgers sent Dick Nen (December 15, 1964) to the Washington Senators to complete the trade.

Mike Cuellar 

Before 1963 Season: Sent from the Cincinnati Reds to the Cleveland Indians in an unknown transaction.

Before 1964 Season: Obtained by Jacksonville (International) from the Cleveland Indians as part of a minor league working agreement.

Before 1964 Season: Returned to the St. Louis Cardinals by Jacksonville (International) after expiration of minor league working agreement.

June 15, 1965: Traded by the St. Louis Cardinals with Ron Taylor to the Houston Astros for Chuck Taylor and Hal Woodeshick.

December 4, 1968: Traded by the Houston Astros with Tom Johnson (minors) and Enzo Hernandez to the Baltimore Orioles for John Mason (minors) and Curt Blefary.

Honorable Mention

The 1907 Cincinnati Reds 

OWAR: 39.9     OWS: 275     OPW%: .527     (81-73)

AWAR: 30.3       AWS: 198      APW%: .431    (66-87)

WARdiff: 9.6                        WSdiff: 77

Cincinnati ended the 1907 season in a fourth-place tie with Philadelphia but finished only six games behind the front-running Cubbies. “Wahoo” Sam Crawford (.323/4/81) laced 34 doubles, 17 triples and led the circuit with 102 runs scored. Orval Overall (23-7, 1.68) flummoxed opposing batsmen, posting a 1.006 WHIP with a League-high 8 shutouts. “Long” Bob Ewing compiled 17 victories with a 1.73 ERA and a WHIP of 1.094 while completing 32 of 37 starts. Patsy Dougherty swiped 33 bags while Mike Mitchell rapped 12 three-base hits in his rookie campaign. Harry Steinfeldt drilled 25 two-baggers and Socks Seybold drove in 92 baserunners.

On Deck

What Might Have Been – The “Original” 1997 Red Sox

References and Resources

Baseball America – Executive Database


James, Bill. The New Bill James Historical Baseball Abstract. New York, NY.: The Free Press, 2001. Print.

James, Bill, with Jim Henzler. Win Shares. Morton Grove, Ill.: STATS, 2002. Print.

Retrosheet – Transactions Database

The information used here was obtained free of charge from and is copyrighted by Retrosheet. Interested parties may contact Retrosheet at “www.retrosheet.org”.

Seamheads – Baseball Gauge

Sean Lahman Baseball Archive

Is Pitcher BABIP All Luck?

This article was originally published on Check Down Sports.

For those of you who have been reading baseball content at Check Down Sports semi-regularly, you’ve probably seen one of us talking about players and teams we think are performing at a level far from expected.

A lot of times when attempting to explain the reasoning behind abnormal pitching performance, we cite a few reasons, and then attribute the rest to good or bad luck. Luck we usually associate with a batter’s batting average on balls in play (BABIP), which is agreed upon by most as beyond the control of the pitcher.

The influx of ball-tracking systems in MLB has allowed for a boatload of new measurements that, until a few years ago, were only dreams in the minds of analysts and evaluators. One of those — the velocity of ball exiting the bat (exit velocity) — is a popular, yet informative piece of data.

Intuitively, it makes sense that the softer the ball leaves the bat, the less likely the ball should result in a hit. A pitcher who suppresses exit velocity should allow fewer batted balls to become base hits than a pitcher who gives up a high exit velocity. Yes, bloops and seeing-eye ground balls will find open space, but on average, I think this assumption makes sense.

But thanks to Statcast and baseballsavant.com, this assumption doesn’t have to be an assumption at all. We can test it out.

Baseball Savant has exit-velocity data since the beginning of 2015, so that’s where I started. I gathered average exit velocity against for pitchers with at least 190 batted-ball events in 2015 and 2016 (298 total). I then got the BABIP for those pitchers in those seasons from FanGraphs. Next, using STATA, I ran a simple linear regression with the two variables. Results are shown below.

Screen Shot 2016-07-14 at 12.33.28 PM

Screen Shot 2016-07-13 at 11.05.37 PM

The scary math-stuff explained:

  • A pitcher’s BABIP isn’t entirely caused by luck
  • Exit velocity has a minor, yet significant, effect on BABIP
  • 6% of a pitcher’s BABIP can be explained by exit velocity
  • If a pitcher decreases his average exit velocity by 1 mph his BABIP will decrease by 0.005 points, on average (i.e. a pitcher decreases his average exit velocity from 90 to 89 mph — his .300 BABIP would fall to .295. In turn, this would lower his ERA)
  • The bottom-left quadrant is ideal. Though, because of exit velocity’s small effect on BABIP, probably not sustainable. We’ve seen Arrieta and and Chris Young come back to earth a bit in 2016
  • The top-left quadrant includes candidates for improvement in the second half of 2016 or 2017. Pitchers here have been unlucky in terms of BABIP. Their exit velocities suggest they should have a lower BABIP, and, therefore, ERA


Playing Probability: Drew Pomeranz and Anderson Espinoza

The Drew Pomeranz-Anderson Espinoza trade has plenty of unknowns. With Pomeranz in the midst of a breakout season and blowing by his previous innings totals and Espinoza at the age when you normally would be graduating high school it is hard to know what to expect. Will Pomeranz continue to dominate or will he return to his previous self where he was either injured or mediocre? Will Espinoza blossom into the pitcher everyone sees when they watch his stuff, or will he fail to get his command under control and turn into the pitcher that his 4+ ERA in Single-A might seem to predict? To evaluate a trade like this you would have to do a lot of guesswork on the futures of these players. So instead of digging into the details it can be best to zoom out, and play the probabilities.

Predicting the futures of prospects is one of the most difficult tasks. Not only do you have to deal with the small samples of short minor-league seasons, but you have to project how those statistics will translate one, two or even three levels above their current competition. Additionally, you have to predict how the player will grow and mature as he enters his prime years. While some try to discover the answers to this question on an individual scale, it can be more effective to embrace the randomness present in each individual human being and create an average performance level for similar groups of players. To do this I tried to think of ways to quantify what makes Espinoza such a noteworthy prospect.

First, I thought that maybe the combination of age and strikeout rate that Espinoza has produced at Single-A Greenville this year might stand out and place him in a small group of notable players. A quick look at even his same team however seemed to prove this hypothesis wrong, as his teammate Roniel Raudes has almost identical statistics to Espinoza this year and is the same age. Raudes however is ranked 24th in Baseball America’s Red Sox preseason rankings, demonstrating that Espinoza’s stats this year are not anything incredibly special by themselves, even when accounting for age.

Next I examined how starters who have debuted for at least 50 innings in a year by age 21 have fared in the major leagues, since Espinoza is predicted to arrive in either his age 20 or 21 season. There are 44 pitchers that meet these qualifications and debuted between 1990 and 2005, some of which are big names (such as Hernandez, Sabathia, Greinke and Kerry Wood) while others not so much. In the time before they were scheduled to reach free agency, these players accumulated on average a total of roughly 8.2 wins above replacement of production over that span. There are some problems with this calculation, however. For one, almost a third of starters debut by age 21, so it is not terribly extraordinary. For every Felix Hernandez in this group there in a Rich Hunter who had one good minor-league season, which prompted a promotion to the big leagues. In his case, his career lasted only that one season. There are also pitchers such as Bud Smith in this group who were once top prospects but faltered in the big leagues. In his case he was once ranked first in the Cardinals’ system, a spot ahead of Albert Pujols, which can work as a friendly reminder that not all big name prospects pan out.

Another way of looking at Espinoza’s value is just to take his prospect ranking for what it is. Kevin Creagh and Steve DiMiceli have done research to try to put a trade value on prospects using Baseball America’s prospect rankings. The following table outlines their findings.

Tier Number of Players Avg. WAR Surplus Value
Pitchers #1-10 22 14.6 $69.9M
Pitchers #11-25 43 8.3 $39.0M
Pitchers #26-50 85 6.4 $29.8M
Pitchers #51-75 104 3.7 $16.5M
Pitchers #76-100 113 3.5 $15.6M


Analyzing data on Baseball America lists from 1994 to 2005, the two men created this table to calculate the average surplus value of players from each tier of the rankings. (Their process is fairly complicated, so it is worth it to take a look at their process here in an earlier version of the study). This can be a very simple yet effective way to evaluate prospects based on both their stats and scouting report (since both are used to create the rankings) while also eliminating as many individual biases as possible (of course the prospect rankings are subject to those same problems).

In Espinoza’s case, he was just recently ranked 15th on the Baseball America midseason top-100, a five-spot jump from his preseason rank (though six players ahead of him have now graduated to the big leagues). This ranking puts him in the second tier of pitchers on the BA rankings and in line to have an average surplus value of 39 million dollars with a projection of 8.3 wins. This is almost exactly the 8.2 wins calculated earlier, though found through a very different method. While these rankings are in no way perfect, it is about as close as you can get to putting a concrete value on Espinoza’s skills (especially since both methods seem to agree), so we will use the $39 million value as a benchmark to compare with Pomeranz.

Moving on to Pomeranz, it is important to find a way to factor in all the different scenarios. You might have multiple ideas about how to take into consideration both this year’s statistics and those of the past, to try to come up with some sort of middle ground. While this is the right idea, this method is going to rely on assumptions that are unlikely to be made completely accurately. Instead we can use projection systems are much better at doing these calculations for us. For Pomeranz, this is the best way to include all the information about the many aspects of his performance and boil it down to one number.

Based on the depth chart projection on FanGraphs (the average between ZiPS and Steamer weighted for projected playing time), Pomeranz is projected to be worth about 1.3 wins the rest of the way. This is considerably worse than he has been so far this year, though much better on a per-inning basis than previous years, and still a valuable player. Using this projection, you can also project out the final two years of his contract assuming that he will continue pitching up to the same standard, by just doing a little math.

With roughly 46% of the season remaining at the All-Star break, you can use his rest-of-season projection to estimate his value over a full season, which ends up being 2.8 wins. First, though, you must use the same process as in the prospect ranking analysis to discount future performance since production today is considered more valuable than years down the line. Multiplying the value of each subsequent season by 0.92 you can account for the future discount rate of 8% (used in the prospect evaluation). Performing this adjustment results in 2017 and 2018 being valued at 2.6 and 2.4 wins respectively for a total of 6.3 wins over the three years with the Red Sox. At eight million dollars per win this totals to be around $50.4 million worth of production.

Finally, we must account for surplus value by subtracting how much Pomeranz is projected to make in arbitration. This year he is only making $1.35 million, but that should see a substantial increase after an All-Star season. I have little experience projecting arbitration but it seems reasonable that he would see his contract jump up to around $6 million in 2017 and see a more modest improvement up to around $10 million after a decent but somewhat less valuable season.  These estimates would total to around $17 million going to Pomeranz from the Red Sox in the three years, and subtracting this from the $50.4 million, you end up with around $33 million in surplus value.

In the end these surplus values are very similar. Espinoza’s value comes in a little higher at $39 million compared to Pomeranz at $33 million, but the $6 million gap is nothing the Red Sox would have to lose sleep over. In reality though that gap is just the gap in average outcomes for both players and it is more likely to be much more lopsided toward one side or the other. While this seems to show that the Padres are getting a slightly better deal, you can easily rationalize this trade for the Red Sox by saying that wins today matter more now for the Red Sox than for the average team since they are in the midst of a tight wild card and division race where they are favored over the division leader in the playoff odds on both FanGraphs and Baseball Prospectus. That way of looking at it does make it much more appealing for Boston. It was a trade that they had to make given their situation. Not a great trade, not a bad trade, but an adequate trade that could turn out either way.

The real takeaway from all this however is on San Diego’s side of the deal. For them it wasn’t just an adequate deal, and it wasn’t even just a good deal. It was a great deal! For them, the pushing of wins down the road is a net gain for them as opposed to a net loss as it is for the Red Sox. They pushed Pomeranz’s average outcome of 6.3 wins (which was being wasted on an noncompetitive team) down the road to a time when they may be competitive and wins will matter much more. Not only that, but they also increased the projected output to 8.3 wins. While it is possible that Espinoza could flop and be a major bust, that is all part of the math that works in the Padres’ favor. For every underwhelming Espinoza, there is a great Espinoza; one that was acquired in exchange for a player that had little value to the club at this point in time and someone the team will get to watch for years to come.

Why Dylan Bundy Will Succeed as a Starter

(Originally written before last Sunday)

It was announced that Orioles pitcher Dylan Bundy will start this Sunday on the road against the Rays. The move makes sense — the Orioles need good starting pitching and Bundy could become a good starter. I think Bundy will do very well as a starter, and in this article I’ll talk about why.

Dylan Bundy’s career started with incredible promise. Drafted fourth overall, his first eight starts in the minors were punctuated by a 0.00 ERA and a 20/1 K/BB ratio. By the end of the year, he was considered the top prospect in all of baseball. The next few years were rife with injuries — first Tommy John surgery in 2013, followed by complications in his shoulder which caused him to miss almost the entire year in 2014. Bundy hasn’t looked like the same pitcher since. His fastball velocity this season started at 92 MPH, much lower than the high 90s we saw in the minors. But since the beginning of June, Bundy has made a remarkable turnaround. Since June 9, the numbers are beyond outstanding, with 14.1 IP, 19 SO, only 4 BBs, and 0 earned runs. But the peripheral stats are even better.

I am currently in the process of writing an article about how I think the most important skill of a starting pitcher is getting to two strikes quickly. Since June 9, Bundy has done this better than any pitcher in baseball. In the top 10: Clayton Kershaw, Max Scherzer, and Stephen Strasburg, arguably three of the best pitchers in baseball. This obviously is not to say that Bundy is one of the best pitchers in baseball; his track record is far, far too short to proclaim that. But it bodes well for Bundy that over the past month he is controlling the ball as well as baseball’s top pitchers.

Bundy’s fastball velocity is also encouraging. Bundy throws a rising four-seam fastball, which bodes well for his ability to miss bats. But at the low 90s, he wasn’t able to generate a lot of swings and misses, and as a fly-ball pitcher was susceptible to home runs. Last appearance, Bundy threw his fastball harder than he’s ever thrown it.


The chart may not look like much, but there’s a clear trend here: Up. His fastball velocity has increased over 4 MPH since the beginning of the season, which is a gigantic leap.

The Orioles desperately need starting pitching, and Bundy could be that answer. The Orioles do not have the worst starting pitching in the league. In terms of WAR, that is currently the Reds. But the Orioles’ staff is really bad, even if they look worse pitching in a hitter’s park. Chris Tillman is their only competent starter, while the rest of their rotation contain some of the worst pitchers in the league. So stretching Bundy into a starter seems appealing.

There is a risk that Bundy will be much worse as a starter. Pitchers are notorious for throwing harder in the bullpen than they would as a starter, and given that the majority of Bundy’s success has come at a higher velocity, it would be reasonable to assume Bundy will not be nearly as effective as a starter as he is as a long reliever. I think this is correct thinking; we should not expect Bundy to start and still average 11 K/9. But his numbers as a reliever have been elite, so there is a lot of room for Bundy to come down and still be a quality starting pitcher. Starting pitcher is where Bundy has the most upside, and the sooner he gains experience, the sooner we can expect him to improve.

Bundy will probably do well this Sunday, especially against a Rays team that strikes out the second-most in the league. Don’t think this is a mirage. Bundy has the stuff and command to succeed, and I think we will see that as a starter.

The WIS Corollary

Interestingly enough, one of the major postwar genres of Anglo-American literature was the academic comedy. Popularized in large part by Philip Larkin and the “Movement,” authors strove to poke fun at academic institutions and the conventions followed by the terrifically aloof professors. The most famous novel to fall into this genre is Lucky Jim by Kingsley Amis. The book features Jim Dixon, a poverty-stricken pseudo-pedant with a probationary position in the history department of a provincial university. A veritable alcoholic, Dixon attempts to solidify his position by penning a hopelessly yawn-inducing piece entitled “The Economic Influence of the Developments in Shipbuilding Techniques, 1450 to 1485.” Short novel made shorter, it doesn’t help him retain his position, but it does succeed in illustrating the banal formalities that academic writing necessitates.

In sabermetrics, there is a heavy reliance on sometimes inscrutable jargon, acronyms that sound like baby words (“FIP!”), and Mike Trout’s historical comps (Chappie Snodgrass is not a very good one in case anyone is wondering) that quite understandably renders the average fan mildly frustrated and the average fan over sixty wondering how we will ever make baseball great again. Typically, I enjoy those articles very much because they communicate news efficiently and analytically. Occasionally, however, articles stray into the Jim Dixon range of absolute obscurity, examining the baseball equivalent of “Shipbuilding Techniques,” whatever that may be. Such writings form the cornerstone of sabermetrics as they mesh history, theory, and sometimes economics.

Fortunately or unfortunately, my article today isn’t quite Dixon-esque, but it retains some of that style’s more tedious elements. It falls more closely into the category of two-minute ESPN quick sabermetric theory update. I don’t think that’s a thing. Seemingly pointless introduction aside, please consider what you know about DIPS theory. I won’t insult your intelligence, but it was developed by Voros McCracken at the turn of the millennium and has served as one of the principal tenets of the pitching side of sabermetrics ever since then. The theory, in its most atomic form, essentially posits that pitchers should be evaluated independently of defense because it’s something they cannot control. Hence “defense-independent pitching statistics.”

Certainly, it was a revolutionary concept and one that has even gained quite a bit of traction in the mainstream sports media. Announcers talk about how a certain pitcher would look a lot better pitching in front of, say, the Giants instead of the Twins. Metrics like xFIP only serve to quantify that idea.

But every grand theory or doctrine (DIPS is essentially sabermetric doctrine at this point) requires a corollary to frame it. And so I propose something I like to call the “WIS Corollary to DIPS,” where WIS stands for Weather Independent Statistics. The natural extension of evaluating pitcher performance independently of defense is to evaluate players independently of weather because it also exists outside of player control.

The basic idea of this is that weather plays enough of a role in enough games to superficially alter the statistics of players such that they cannot be accurately and precisely compared with the other players in the league because all of them face different environmental conditions. Taking that into consideration, all efforts must be made to strip out the effects of weather when making serious player comparisons. Coors Field is why Colorado performances are regarded with such skepticism, while the nature of San Francisco weather and AT&T Park is supposedly why that location serves as an apt environment for the development of pitchers.

Think about it — it’s something we already do. We look at home/road splits, we evaluate park factors, we try and put players on +/- scales. We talk about this constantly even at youth games. I have heard parents say many times, “If only the wind hadn’t been blowing in so hard he might have hit the fence.” It’s honestly a commonly held, yet generally unquantified, notion that the general public has.

Player X hits a blooper at Stadium C that falls in front of the left fielder for a hit. Player Y hits a blooper at Stadium D with the exact same exit velocity and launch angle as Player X’s ball, but it carries into the glove of an expectant left fielder. Should Player X really get credit for a hit and Player Y for an out? Basically all statistics, striving to communicate objective information, would say yes. If this kind of thing happens enough times over the course of a season, it can make a significant difference. A couple of fly balls that leave the park instead of being caught at the fence would put a dent in a pitcher’s ERA, while changing a player’s wRC+ by no small sum.

For that reason, players should be measured as if they play in a vacuum. One of the biggest goals of sabermetrics is to isolate player performance in order to evaluate him independently of variables he cannot necessarily control. Certainly, this has some far-reaching consequences if the idea gets carried out to its natural conclusion. Someone would likely end up developing a model that standardized stadium size, defensive alignment for varied player types, and other things of that nature. I’m not necessarily advocating for that, just for stripping out the effects of weather.

WIS by itself isn’t radical, but the extent to which it’s applied could be considered as such. As of now, it’s something consciously applied a relatively small portion of the time, but I think that it’s something that should be considered as much as possible. Obviously, there are issues with this. You can’t very well modify “raw” statistics like batting average or ERA so that they reflect play in a vacuum. What you could conceivably do is create a rather complicated model that requires a complicated explanation in order to describe how the players should have performed. And that’s something which brings us to an important point; the metrics that would employ this information would not be for the average fan; rather, they would be aimed at the serious analyst.

This is something I’ve already tried to employ with a metric I created called xHR, which uses the launch angle and exit velocity of batted balls to retroactively predict the number of home runs a player should have hit. The metric is still in development, but I think it’s something that works relatively well and can be applied to other types of metrics. For instance, an incredibly complex and comprehensive expected batting average could utilize Statcast information to determine whether a given fly ball would have been a hit in a vacuum based on fielder routes and the physics of the hit. By no means am I trying to assert that I have all, if any, of the answers. The only thing I’m trying to do here is to bring debate to a small corner of the internet regarding the proper way to evaluate baseball players.

Probably the most crucial thing to understand here is that the point of sabermetrics is to accurately and precisely evaluate players in the best possible way. Sabermetricians already do an incredible job of doing just that, but perhaps it’s time to take things a step further in the evaluation process by developing metrics that put performances in a vacuum. I know that baseball doesn’t happen in a void, but the best possible way to compare players is to measure them* as if they do.

WIS Corollary — One must strip out the effects of weather on players in order to have the most accurate and precise comparison between them.

*Oftentimes it’s necessary to compare players while including uncontrollable factors, like sequencing, especially when doing historical comparisons. It’s important to note that the WIS Corollary is applicable only in very specialized situations, and would generally go unused.

The Yankees’ Bad Decisions and How They Can Reverse Them

Before this season, everybody knew that the Yankees wouldn’t exactly be in contention this year.  But nobody could have predicted the extent to which their performance would dip — especially in hitting.  They have gotten an amazing performance from Carlos Beltran (wRC+ of 132), but that’s about it.  Oh, and Beltran has been a complete flop at fielding, managing to accumulate a -10 DRS only halfway into the season.  To offset his defensive issues, the Yankees can’t move him to first base, because he’s blocked there by Mark Teixeira, who’s earning $22.5 million a year.  And it’s not as if Mark Teixeira is earning his fat paycheck, either.  As of of the All-Star break, he has a -1.1 WAR.  So with Teixeira’s $22.5 million paycheck this year and less-than-desirable performance, trading him to make room for Beltran at first base is not an option.

So what about moving Beltran to DH?  Or moving Teixeira to DH and having Beltran play first?  A bit of a problem there.  See, Alex Rodriguez is right now occupying the DH spot.  And he’s earning $20 million this year while hitting .220 with a -0.7 WAR.  And while, theoretically, the Yanks could move Alex over to the hot corner to make room for Beltran, there’s the small problem of Chase Headley, who’s earning $13 million a year.  And while, yes, the Yankees could trade Chase Headley, who holds enough value to be desired by some clubs, nobody in the Yankees front office wants to even think, much less see, this scenario:  Almost 41-year-old Alex Rodriguez bumbling around the hot corner, feebly trying (and failing) to convert routine ground balls into outs.

So Beltran will be staying in right field until the inevitable happens:  one of the many 30-year-olds on the Yankees gets injured.  Some of those those 30-year-olds — Beltran, Texeira, and Rodriguez — combine to have a -0.3 WAR.  That is well below league-average.  Their earnings on the other hand…$57.5 million combined for 2016 alone.  Paying $57.5 million for -0.3 WAR.  However way you look at it, that’s a bad deal.  A really bad deal.  And that’s only three of the 25 people on the Yankees roster.  And you can be sure that the other 23 aren’t a general manager’s dream.  Quite the contrary.  Let’s go position by position and see exactly how horrible the Yankees’ hitters are when compared to their salaries.


Position: Players: Combined Salary: Combined WAR:
Catcher Brian McCann, Austin Romine 17.5 million 1.5
First Base Chris Parmelee, Rob Refsnyder, Ike Davis, Dustin Ackley, Mark Teixeira 27.9 million -1.1
Second Base Starlin Castro 7 million -0.4
Third Base Chase Headley, Ronald Torreyes 13.5 million 1.1
Shortstop Didi Gregorious 2.4 million 1.5
Right Field Carlos Beltran, Benjamin Gamel, Aaron Hicks 16.1 million 0.6
Left Field Brett Gardner 13 million 1.0
Center Field Jacoby Ellsbury 21.1 million 1.4
DH Gary Sanchez, Alex Rodriguez 20.5 million -0.8

So the Yankees’ payroll for hitters alone is $139 million for 2016.  Although that is a big sum — a gigantic sum — it wouldn’t have been noteworthy if the big names had performed and driven the Yanks to a playoff run.  Instead, though, those big names have performed terribly (except for Beltran) and the Yankees have almost no chance of making the postseason.

Right now the MLB is averaging six million dollars per 1 WAR.  That may sound like a lot, but compared to the Yankees it is nothing.  Since it is halfway through the season, their 4.8 combined WAR is 9.6 on a full-season scale.  139 million divided by 9.6 is 14.5.  That means that the Yankees are paying $14.5 million per 1 WAR.  That is more than two times league average.  Although they are overpaying for many players, the big blows come from five players only:  Alex Rodriguez, Mark Teixeira, Carlos Beltran, Brian McCann, and Jacoby Ellsbury.  All these players signed their mega deals after one of, if not the best season in their careers.  Except for Beltran, who got a three-year deal, all these players signed deals for five or more seasons.  Here is the rundown on their salaries.

Alex Rodriguez:  Alex’s deal is probably the stupidest of all other Yankees deals in history.  He was signed to a 10-year deal with the Rangers in 2001, and was traded to the Yankees in 2004.  His contract would then expire after the 2010 season, when he would be 35 years old.  But the Yankees, for some reason, decided to renew his contract two years before it expired, in 2008.  If the Yankees had signed him to a new five-year deal, that would not have been too bad.  But instead, the Yankees signed him to another 10-year deal worth 275 million dollars, $25 million more than his former deal.  So now he is signed through the 2017 season, when he will be 43 years old.  If the Yankees would have only agreed to let A-Rod go after the 2010 season, they would have avoided all the bad/OK years of his career, which, incidentally, started in 2011.

Mark Teixeira:  In 2009, the Yankees signed Mark Teixeira, who was coming off of a 6.9 WAR season, to an eight-year, 180-million-dollar deal.  To be fair, it was not a bad signing for the Yanks.  Teixeira was 29 in his first year as a Yankee, and got a 142 WRC+ while accumulating a 5.1 WAR.  Then the next year he dipped to a still-respectable 3.4 WAR.  But he was on a downwards path.  After one final good year in 2011, he slowly declined into what he is now: an expensive waste of a perfectly good roster spot.  But don’t condemn the Yankees for that.  Yes, they probably slightly overpaid for a .250 average/30 HR first baseman, but it wasn’t a horrible signing.  What was bad about it was the deal itself.  Not the money involved or the years.  The reason.  Why did the Yankees need a first baseman?  The year before the deal, 2008, Jason Giambi hit 32 homers and had a 131 wRC+ at first.  Yes, his deal was up after the season, but the Yankees could have easily re-signed Giambi without having to pay him $180 million.  So the Yankees didn’t need Teixeira.  They just wanted him.  And that is the same trap they’ve fallen into ever since the dawn of free agency.

Carlos Beltran:  The Yankees signed Beltran to a three-year deal worth $45 million in 2014.  At the time, he was 36 and coming off a good season with the Cardinals.  In fact, it was a great season — hitting-wise.  At defense, there is no way around it.  He was simply terrible.  He made almost all of the plays he got to, but he didn’t get to many.  He couldn’t run fast if you pointed a gun at him.  And somehow, for some reason, the Yankees expected him to play outfield for three more seasons — until he was 39.  And guess what?  It hasn’t worked out too well.  His hitting has been very good, but that hitting value has been stripped from him by his terrible fielding.

Brian McCann:  In 2014, the Yankees signed Brian McCann to a five-year deal worth $85 million.  At the time, it seemed like a good deal; a catcher who could hit well, signed for only $17 million a year.  In any other circumstance, that would be considered a good deal.  A great deal even.  But there was one problem.  It was a 30-year-old catcher they signed for five years.  A 30-year-old catcher who most likely wouldn’t survive two more years crouching behind the plate every inning for 140 games a year.  So for two years, they Yankees got a good deal.  But this year is the third year of the deal.  And surprise, surprise, your 32 1/2-year-old catcher is not performing too well behind the plate.  -6 DRS there.  And, frankly, his hitting is just not good enough to compensate the bad fielding behind the plate.

Jacoby Ellsbury:  In 2014 the Yankees gave Jacoby Ellsbury a 153-million-dollar, seven-year deal.  Ellsbury, who was 30 years old in 2014 and had a history of getting injured, was coming off of a 5.0 WAR season.  But that was mostly due to his well above-average speed.  He used it to his advantage on the basepaths and in the outfield.  All that is fine and good, but there is one problem:  Speed is the first tool to disappear from a player’s repertoire because of age.  And the Yankees’ deal with Ellsbury started when he was 30.  And after a 39-steal year for the first year of the deal, Ellsbury unsurprisingly swiped only 20 bags in the second year.  He used to consistently have 10 DRS every season; now, with the loss of his speed, that 10 has turned into zero.  And aside from steals and defense, Ellsbury doesn’t hold much value.  He hits about five homers a season, and is good for a .280 average.  And five homers, a .280 average and 20 steals is not worth 21 million dollars a year.

So where do the Yankees go from here?  Beltran, Teixeira, and Rodriguez’s contracts will all end this year or the next.  Then they are stuck with only Ellsbury’s and McCann’s.  McCann’s expires in 2018, and, realistically, the Yankees can deal with $17 million a year for two more years.  And with the way Ellsbury’s been playing this year, the Yankees can easily trade him for a small prospect and pay half of his remaining contract.  So if they trade Ellsbury the Yankees will be left with an (almost) clean slate at the end of this year.  How do they fill it?  Here are some suggestions on what and what not to do.

1.  Stay away from pricey free agents ages 31+.

2. Make sure not to sign any player who will be 37+ at any point during the deal.

3. Pay attention to the draft.  For the next few years, the Yankees won’t be very good, so they should make use of their high draft picks and start developing prospects, rather than just buying overpriced free agents.

4.  Only buy value-high, salary-low free agents, i.e. Ben Zobrist.

5.  Stay away from deals spanning eight years or longer.

Let’s get more in-depth with these five bullet points.  Oh, yeah, there’s a sixth:

6.  Get a new G.M.

So let’s get more in-depth with these six bullet points.  1. Stay away from pricey free agents ages 31+:  This rule should be one the Yankees know well by now.  After breaking this rule many times with no good results, this rule should be a relatively easy one for the Yanks to swallow.  Remember, the rule states “pricey free agents,” so that doesn’t include older players (35-36; as you will see in the next bullet, signing Jamie Moyer is forbidden) who still retain some value and can be signed for cheap.

2.  Make sure not to sign any player who will be 37+ at any point during the deal:  It doesn’t matter if this deal is for three years or for 11.  The message is clear:  older players are at higher risk of either sharply declining or getting injured.

3.  Pay attention to the draft:  For most of their history, the Yankees lived in an era of no free agents, so they were able to rip the poor teams of their great prospects with the promise of good money.  Now, when almost every team has enough money, and those who don’t (Rays, Astros, Marlins, Pirates) are smart enough so that they won’t give away their prospects, this strategy is much harder.  So the Yankees switched their focus to high-priced free agents.  This new “strategy” has had its ups and downs.  Most of the ups came earlier, when teams didn’t try to retain their stars after the six years of cost-control.  Now, with many stars (Stanton, Strasburg) being offered luxurious extensions by their teams, most of the talent never hits the free-agent market until much later, when it is not worth much.  So, with extreme reluctance, the Yankees must turn their attention to the draft, an event they have somewhat ignored over the past years.  Although they do make an effort to sign players and do draft players with good potential, they have not made a real effort to dig deep and find hidden gems.  Remember, Mike Piazza was drafted in the 62nd round.  And furthermore, they must not be tempted to trade away these hidden gems they worked so hard to get in return for a major-league player with not half the talent as the prospect.

4.  Only buy high-value, low-salary free-agents:  In years past, this strategy would have worked wonders for the worst team in the league who has a small budget.  Imagine:  who would sign a .272 hitter with 10 homers to a 56-million-dollar contract 10 years ago?  Almost nobody.  But just this year, Ben Zobrist received that contract.  And according to WAR, he should have received more.  Here is a list of smart free agents for after the 2016 season:

Catcher:  There are three good catchers eligible for free agency after the 2016 season:  Jonathan Lucroy, Wilson Ramos, and Matt Wieters.  Out of the three, Jonathan Lucroy and Matt Wieters will most likely be the most wanted.  So that leaves Wilson Ramos.  He is 29 and a solid backstop with hitting potential.  Smart buy:  Wilson Ramos

First Base:  There are actually no standout smart buys at first base.  Justin Smoak, Carlos Santana, and Sean Rodriguez are all options.  The one who has the most value when compared the the estimated price, though, is Sean Rodriguez.  He is also one of the youngest first baseman of all free agents.  Smart buy:  Sean Rodriguez.

Second Base:  Most of the second base free agents next year are way above our target age.  The few that are in our age range are Gordon Beckham, Chris Coghlan, Daniel Descalso, and Neil Walker.  We can safely say that Gordon Beckham and Daniel Descalso are off the list, simply because they don’t provide the value to be a smart buy.  Neil Walker’s price will have shot way up after the amazing campaign he is having this year, so that leaves us with Chris Coghlan.  Chris, who is 32, holds loads of value as he can play second base as well as corner outfield positions.  He is also having one of the worst seasons of his career as of now, so he will be really cheap come the season’s end.  Smart buy:  Chris Coghlan.

Third Base:  At third there are only a few free agents in the Yankees’ age range.  They are Luis Valbuena, Justin Turner, and Martin Prado.  Luis Valbuena is eliminated, because he is too big and awkward to stick at third base.  Somewhere in his near future he will be transitioned to first base.  So that leaves us with Justin Turner and Martin Prado.  These are both good value picks, but Justin Turner must be eliminated.  He will be way too expensive, two years removed from the best season of his life (so far) and part of a playoff contending team.  Martin Prado is our smart buy for third base.  He has been amazingly consistent his whole career, and coming from the Marlins, his price tag will be relatively low.  Smart buy:  Martin Prado.

Shortstop:  There are only four shortstops available after the season ends, and three of them fit the basic criteria:  Alcides Escobar, Erick Aybar, and Ruben Tejada.  Ruben Tejada is the first elimination, as he does not have enough experience in the big leagues to validate his performance.  Alcides Escobar also must go, because he is most likely going to be re-signed by KC.  And even if he is not, his price will be driven up by their bids.  That leaves Erick Aybar.  He is consistent, and hardly ever injured.  He is also mired in a huge slump right now, which will significantly drive down his price.  Smart buy:  Erick Aybar.

Right Field:  There is simply no other competition for smart buy.  Josh Reddick has amazing defense in right, can hit very well, and is only 30 years old.  He is also playing for the obscure Athletics right now, which will drive down his price.  Smart buy:  Josh Reddick.

Left Field:  There are so many standout left fielders going into the 2016-2017 free agency that they will all drive down the price of each other.  That will allow the smart buy to be a big player.  The big left field names are Michael Saunders, Matt Holliday, Ian Desmond, and Yoenis Cespedes.  Matt Holliday is too old, so he’s out.  Yoenis Cespedes is too fluky, and can be injury-prone, so he’s also out.  That leaves us with Ian Desmond and Michael Saunders.  Both of these players are having breakout seasons so far.  Ian Desmond offers more flexibility in the field, as he can play shortstop, second base, and all the outfield positions, including center.  Michael Saunders might be a little cheaper, but it is hard to tell.  It was close but Ian Desmond is our smart buy for left field.  Smart buy:  Ian Desmond.

Center Field:  There are three very good possibilities:  Carlos Gomez, Dexter Fowler, and Austin Jackson.  Dexter Fowler would be a very good pick almost any other season, but he is having a breakout year so far for the Cubs, so he’s out.  Austin Jackson, on the other hand, is having one of his worst seasons ever.  The problem with him is that he doesn’t seem capable of ever making a return to the player he used to be.  He is coming off three straight seasons in which he failed to hit higher than .270.  So Carlos Gomez it is.  He has struggled mightily with the Astros, but that is probably just the effect of playing in a huge ballpark rather than in hitter-friendly Milwaukee.  Smart buy:  Carlos Gomez.

DH:  With many players to choose from, the only player that really catches the smart buyer’s eye is Pedro Alvarez.  He hasn’t found much playing time with power-packed Baltimore, so that will bring down his value significantly.  Smart buy:  Pedro Alvarez.

Starting Pitchers:  With so many choose from, there will be five smart buys for starting pitchers.  There are many soon-to-be free agent starting pitchers ages 28-31.  Smart buy #1 (note:  smart buy number does not imply any greater value for pitcher):  Brett Anderson.  Coming off an injury but with many years of experience with him, Brett Anderson is a great pick for any team.  Smart buy #2:  Jaime Garcia.  So far he has been OK this season, but very consistent.  A very good pick for any team looking for a cheap starting pitcher with a high ceiling.  Smart buy #3:  Jeremy Hellickson.  Hellickson has never had a horrible year in his career.  Although he did have one 5.00 E.R.A. year, he still had a positive WAR.  And aside from that season, he has been pretty good, but not good enough to warrant a big contract.  Smart buy #4:  Matt Moore.  Moore is a dependable, extremely young left-hander.  In fact, he is one of the youngest starting pitchers on the market for next year.  Smart Buy #5:  Ivan Nova.  Although he’s had a rough year so far, you have to love the potential!  He is only 30 years old, and best of all, he’s on the Yankees right now, so they have easy access to him.

Those are all the smart buys.  I am not suggesting that the Yankees sign every single one of those players, but three of four of them wouldn’t hurt.  In fact, they would most likely help the Yankees turn their club around quickly — much quicker than anyone projected them to.

5.  Stay away from deals spanning eight years or longer:  This rule will help prevent the Mark Teixeira deals, the Alex Rodriguez deals, and the Jacoby Ellsbury deals.  This way, if a player is signed for five years, and only performs well for three years of the deal, the Yankees only have to deal with two bad years.

6.  Get a new G.M:  Brian Cashman simply hasn’t gotten it done.  He was given a good team with an unlimited budget and has turned it into one of the worst clubs in baseball.

Hopefully, the Yankees will use their bad experiences to their advantage and become one of the smarter teams in baseball, a la the Astros, Rays, and Pirates.  With the help of these rules and suggestions, they can become the most dangerous team in the MLB, with money and smarts.

Park Factors to (Maybe) Monitor

Every baseball stadium is different.  This is an obvious fact, but its obviousness can obscure its importance.  Every baseball stadium is different, so baseball is different in every stadium.  Some of these differences are easy to discern such as HRs in Denver and Cincinnati.  Others though are more easily masked — did you know that the White Sox’ U.S. Cellular Field raises walks by 7%?  Each game is a combination of outcomes affected by each team’s talent and, to a lesser extent, these park factors.  FanGraphs is nice enough to publish its park factors here.

With the league-wide increase in exit velocity and home runs, I was interested to know if any park factors may be changing as well.  With roughly half of the 2016 season in the books, I thought now was as good a time as any to take a look.  Rather than go through the laborious calculations necessary to find park factors like those at FanGraphs, I came up with a quick and not at all exact way to look at just this season.  Essentially, I found each team’s home and away rates of 1B, 2B, 3B, HR, SO and BB per plate appearance.  I then compared each to league average on the same scale as wRC+ (100 is average).  I then calculated a quick park factor on the same scale for each of the above stats as follows (1B factor shown below):

((Team Home 1B Rate – (Team Away 1B Rate – 100)) + 100) / 2 = 1B Park Factor

For example, the Marlins have hit 4% more singles than average at home (104 1B+), and 27% more singles than average on the road (127 1B+), so the Marlins Park 1B park factor would be 88 (depresses singles by 12%).

I am fully aware of the many problems with the methodology (ignores half of the data, small sample, not enough regression included, team road schedules aren’t guaranteed to have average park factors, etc.), but like I said, I wanted something quick, and I am only focused on the extremes anyway.  This should at least show us which parks to consider monitoring or examining further.

2015 FanGraphs vs. 2016 Observed Park Factors
2015 FanGraphs 2016 Observed
Team 1B 2B 3B HR SO BB Team 1B 2B 3B HR SO BB
Angels 100 96 91 93 102 97 Angels 98 87 80 105 101 103
Astros 99 100 108 105 103 101 Astros 93 103 138 101 104 102
Athletics 99 100 105 93 97 101 Athletics 97 97 145 90 98 94
Blue Jays 97 108 105 106 102 99 Blue Jays 107 116 74 90 103 102
Braves 100 99 93 96 103 101 Braves 106 85 125 94 99 102
Brewers 99 100 102 112 101 102 Brewers 95 106 131 113 98 104
Cardinals 100 99 95 94 98 99 Cardinals 101 104 42 88 96 98
Cubs 99 99 102 102 101 102 Cubs 96 84 105 100 98 111
Diamondbacks 99 99 102 100 98 111 Diamondbacks 99 105 120 102 100 99
Dodgers 98 98 78 102 100 96 Dodgers 98 91 69 116 98 101
Giants 99 97 115 84 100 100 Giants 103 97 163 83 100 109
Indians 100 103 81 101 101 99 Indians 109 121 21 105 93 120
Mariners 98 87 85 98 102 97 Mariners 96 96 92 108 97 108
Marlins 101 100 117 88 98 101 Marlins 88 109 42 102 99 102
Mets 96 95 87 101 101 100 Mets 98 86 80 108 98 111
Nationals 104 102 84 97 97 98 Nationals 104 90 70 98 93 109
Orioles 101 99 86 108 99 100 Orioles 103 93 118 105 89 109
Padres 98 95 97 98 102 101 Padres 99 98 100 94 96 102
Phillies 98 99 92 107 103 102 Phillies 93 87 128 94 104 104
Pirates 101 99 89 90 96 96 Pirates 106 88 157 101 92 110
Rangers 103 101 110 105 98 102 Rangers 106 105 153 86 95 113
Rays 99 95 98 96 102 100 Rays 99 99 98 84 105 95
Red Sox 103 114 105 96 100 100 Red Sox 102 123 90 87 94 109
Reds 99 98 92 113 103 101 Reds 97 94 100 121 102 99
Rockies 110 108 128 113 95 102 Rockies 103 134 170 109 86 114
Royals 101 103 114 93 96 99 Royals 104 113 141 100 90 106
Tigers 101 98 126 98 95 99 Tigers 105 97 135 105 96 105
Twins 102 101 106 98 98 99 Twins 105 101 171 86 87 98
White Sox 99 97 91 108 103 107 White Sox 100 99 86 108 97 107
Yankees 100 97 84 110 101 101 Yankees 94 102 86 120 97 116
Data pulled at All-Star Break

I know that is a lot to digest, and I apologize it is not sortable due to my lack of coding skill — but there are some interesting differences buried in that table.

1B Park Factor

Two parks stick out at the extreme ends for singles.  The aforementioned Marlins Park went from slightly single-friendly to the worst park for singles.  I don’t have a good explanation for this, though the fences were moved in prior to this season which we would expect to set off a ripple affect with the park factors.  The Blue Jays’ Rogers Centre went the opposite direction of the Marlins, showing a move from slightly below-average for singles to the second-best park for singles.  The Jays did change to a dirt infield from turf for 2016, but I would expect that to decrease 1Bs rather than increase them.  Maybe dirt slows infielders down giving them less range?  The Jays have recorded more infield and bunt hits at home than on the road as well, which would increase singles.

2B Park Factor

Coors Field has seen a marked increase in doubles (and triples) in 2016 with a small decrease in HRs, which is very interesting considering they raised several areas of the outfield walls.  The Cubs, Braves, Nationals, Phillies and Pirates have all seen at least a 10-point decrease in 2Bs.  Of that group, the Braves, Phillies and Pirates seem to have traded those doubles for triples which I wouldn’t necessarily expect to hold up as a change in the park factor given the limited samples.  The Phillies also made a change to a longer-cut grass, so a decrease in 1Bs and 2Bs makes some sense.  I am not sure what is going on in Chicago (wind patterns?) and Washington as the decrease in doubles does not seem to be offset by an increase in other similar batted balls.

3B Park Factor

As expected with the extremely limited number of triples, there is a ton of variation across the half-season sample.  The two most likely to represent a true change to the park factors in my mind are the decrease in triples in Marlins Park (moved fences in) and the increase in triples at Coors Field (raised fences), though both likely won’t hold up to this magnitude.

HR Park Factor

There have been large and unexpected decreases in home runs in Toronto and Texas, while the Marlins and Dodgers have seen upticks in homers at home.  Probably nothing but small-sample noise here.  It will be worth checking more rigorously to see if these hold up, particularly at Marlins Park given the change to the fences.

Strikeout and Walk Park Factors

Given the way I have calculated each component park factor, I expected all of them to need an adjustment for home-field advantage.  Interestingly, that was not the case for 1Bs, 2Bs and HRs as the average observed park factor for each was 100 across the league.  I wrote off the 108 average observed 3B factor as small-sample noise, but I believe I picked up some measure of home-field advantage in strikeouts and walks.  On average across the league, home parks decreased strikeouts by 3% and increased walks by 5%.  These have been regressed and the samples for each are among the largest of the component park factors (more PAs end in a K than any specific batted-ball outcome, and there are more BBs than anything except 1Bs), so it feels like this reflects something.

The extreme parks for changes in strikeouts are the Twins’ Target Field and Diamondbacks’ Chase Field.  Adjusting for the home-field difference (the unadjusted numbers are shown in the table above), the Twins’ park seems to be decreasing strikeouts by about 8% more than usual, while the Diamondbacks’ stadium is increasing Ks by 8% more than FG expects.  The Twins did make a change to their CF seating that could be affecting the hitters’ ability to pick up pitches (and thus strike out less), but if that is the case an increase in walks would also be expected — and that is not the case, as the Twins have actually walked less than expected when including the home-field adjustment.

For changes in BBs (after adjusting for home field), the parks in Oakland and Cleveland stick out.  The Coliseum has allowed 12% less walks than expected, while the Indians’ Progressive Field has inflated walks by 16%.  These may be worth exploring as both parks have also affected strikeouts, with the A’s park increasing strikeouts and the Indians’ park decreasing Ks.  It is possible hitters are not picking up the ball in Oakland while they are seeing it well in Cleveland.


So there you have it.  Noisy, likely inaccurate 2016 park factors.  It will be very interesting to see if any of the observed changes detailed above turn out to reflect a true change in the park factors.  My best guess is Colorado, Miami and Toronto will need some type of adjustment from the 2015 park factors given the fairly significant changes to each park debuting in 2016.  It would be fascinating to hear thoughts from the players on the extreme differences found above as well.  The fact that each park is so different is part of baseball’s appeal to me.  Every game really is totally unique, all the way down to the field itself.

David Price Is About to Go Off

View post on imgur.com

On June 25, this was David Price’s tweet to family, friends and fans.  It was a clear signal that he knew the patience of the Boston fans and media was wearing thin.

Fast forward to the All-Star break and his “Made for TV” stats (those that casual fans know best) are underwhelming: a 9-6 record with a 4.34 ERA, which is worse than the MLB average of 4.23.  It’s not so much his ERA that’s the problem to fans, but more his inability to be consistent from start to start.  Price has three starts of six-plus innings allowing two or fewer runs, but also has four starts of allowing six or more runs.  With the rest of the rotation producing an atrocious 4.86 ERA, the Sox desperately needed Price to be the one to stop the bleeding, something he hasn’t been able to do.  But that doesn’t mean his underlying skills have deteriorated and all of a sudden he’s become a league-average pitcher.  In fact, the advanced metrics say he’s been extremely unlucky and that he’s due for a big second half. 

View post on imgur.com

* Rank is solely being used to establish a baseline for Price as a top 10 pitcher.

In 2014 and 2015 combined, Price was ranked in the top 10 of all pitchers in four of the skill-based statistics: K%, BB%, xFIP and SIERA (the latter two being ERA estimators with a weighting towards more pitcher-controlled outcomes).  Through the 2016 All-Star break, Price has maintained or improved his top-10 rank in K%, xFIP and SIERA but dropped a few spots in walk rate.  Despite the move from 9th to 10th in K% rank, his K rate is actually up from 26.2% to 27.1%.  The reason for the drop in rank is that 2016 newcomers to the list Jose Fernandez, Noah Syndergaard and Drew Pomeranz did not meet the minimum innings qualifier for the 2014/2015 combined list.  On the flip side, Price’s xFIP and SIERA are higher than they were the past two years, but he has improved his ranking versus his peers.  This is because xFIPs and SIERAs are both up 10% league-wide versus last year (due to all the home runs being hit) while Price’s increases are smaller.

So what is happening?  If his base skills are fine, why is his ERA so high and his performance so inconsistent?

View post on imgur.com

So everyone is familiar with ERA and can easily infer that 4.34 is no bueno for a $217-million pitcher.  But there is a reason these stats are labeled “Non Skill-Based” — that’s because these stats are influenced by factors outside of the pitcher’s direct control (defense, luck, sequencing, variance, etc…) and therefore have wide variability over small samples.  Three of these stats (HR/FB%, BABIP and LOB%) explain why David Price is a great rebound candidate for the second half.


Price’s current HR/FB (home runs per fly ball) rate is 15.2% — which is good for being ranked 76th out of 97 qualified starting pitchers.  The past two years combined he ranked 19th.  To put this in context, Price’s career average is 9.4% while the 2016 league average is 12.9%.  Price has never recorded a full season (>150 IP) HR/FB rate higher than 10.5%.  Also, on balls hit into play against Price this year, 31.3% of them are fly balls, the second-lowest rate of his career.  The only season in which he allowed a lower fly ball rate was in 2012 when he won the AL Cy Young award.  Price is giving up fewer fly balls this year, but of the fly balls he is allowing, they are going over the fence at the highest rate of his career.  Those that remember Price giving up a HR in 10 consecutive starts this year are nodding violently right now.  His HR/FB% will regress towards his career norm (9.4%) and this should be the main reason for a big second half.


Price is also suffering from an unsustainable BABIP (batting average on balls in play).  His current mark of .321 is well above his career rate (.289) and even above his highest full-season rate (.306).  Once a ball is put into play it is out of the pitcher’s control what happens from there.  This is why defense and luck influence this stat more than skill.  And with that said, statistical outliers here tend to regress towards career norms.  Even though Price is allowing ground balls at a higher rate than the past two years, his 2016 GB% is still lower than his career average.  BABIP can be influenced by the number of ground balls a pitcher allows, but he’s not allowing vastly more than his career average.  His BABIP should have some positive regression in it, which is another predictor of improved second-half performance.


Price’s Left-On-Base% (percentage of runners a pitcher strands over the course of a season) is currently 70.9%, which is also below his career rate (74.7%) and would be his second worst full-season rate (70.0%) if the season ended today.  Similar to HR/FB%, he is ranked 73rd out of 97 qualified starting pitchers.  The past two years he ranked 22nd.  A pitcher with a higher than average strikeout rate should be able to sustain a slightly higher than average LOB%, but it’s playing out the exact opposite way for Price.  This is partly due to his inflated BABIP and HR/FB%; as these statistics continue to regress towards his career norms, the LOB% will creep up to expected levels.

Much has been made of Price’s velocity being down this year compared to any point in his career.  At the start of the season, his velocity was over 2.0 MPH lower than his career average (94.1).  He has since closed this gap almost entirely.  Here is his average fastball velocity by month (with number of starts):

April: 92.0 (5)

May: 92.5 (6)

June: 92.9 (6)

July: 94.0 (2)

If this upward trend in velocity stabilizes somewhere at or above 93.5, then nearly all the performance metrics within his control — velocity, K%, BB%, xFIP and SIERA — will be at or near his career norms.

Let’s dive a little deeper into that early-season velocity issue.  Below are two charts.  The first shows combined performance of 2014 and 2015 for ERA-qualifying starters while the second chart is the same data for the 2016 season through the All-Star break.  The orange circle is David Price.  The red circle (if shown) represents Price’s career average.  The blue circles are a hand selected peer group of the top 10 pitchers in the game (Kershaw, Sale, Arrieta, Scherzer, Bumgarner, Greinke, Strasburg, Syndergaard, Salazar and Fernandez).  Remember those rankings where Price was right around the top 10 — these are the guys usually outperforming him.  The gray circles represent everyone else.  Note: For these first two charts the top-right quadrant is Good, and the bottom-left quadrant is Bad (unless you’re a knuckleballer).

2014-2015 K/9 vs FBv

View post on imgur.com


2016 K/9 vs FBv

View post on imgur.com

The first graph shows David Price clustered where you would expect him — right at the middle-to-bottom of his top-10 peer group, with a healthy average fastball velocity and K/9.  The second graph (2016) shows Price in a similar relationship to his peers, but with slightly lower velocity and a higher K/9.  Note the gap between the orange (Price’s 2016) and red (Price’s career average) dots depicting his improved strikeout numbers this year despite the slightly lower velocity.  This graph also shows what freaks Noah Syndergaard, Jose Fernandez and (to a lesser degree) Jered Weaver are.

The final two graphs show the relationship between ERA and xFIP where xFIP is the more predictive estimator of a pitcher’s skill.  The bottom-left quadrant is Good (think Kershaw) and the upper-right quadrant is Bad (think Buchholz).  Anyone in the upper-left quadrant (Price in 2016) is a candidate for positive regression.

2014-2015 ERA vs xFIP

View post on imgur.com


2016 ERA vs xFIP

View post on imgur.com

The first graph again shows Price in his usual place — at the tail end of the top 10.  In 2014 and 2015 combined he had a very similar ERA (2.88) and xFIP (2.98).  The second graph (2016) shows the disparity between his ERA (4.34) and xFIP (3.16).  Pitchers with this large of a gap between ERA and xFIP are great candidates for regression.  The important takeaway is that his xFIP, relative to his peers, has stayed in that top-10 range.  This supports the point that some bad luck is the main element depressing his ERA.

David Price can easily be the best pitcher in the American League over the next two and a half months.  He already owns the lowest xFIP in the AL at 3.16 — the next-closest is Corey Kluber, at 3.34.  The skills above show he can sustain the xFIP level, but with some change in luck and maintaining his improved velocity, he doesn’t need to “pitch better”; he just needs to keep pitching — and the results will follow.

Can First-Half (x)FIP Predict Second-Half ERA?

This article was originally published on Check Down Sports

Predictions are hard. Getting them right is harder. But everyone loves them, so I’m going to attempt to predict which starting pitchers will improve in the second half of the season, and which are poised to put up worse numbers. This information may be especially helpful for a GM thinking about acquiring a pitcher before the trade deadline, or, maybe more applicably, a fantasy owner trying to surge his team into playoff position.

How do you exactly predict starting-pitcher performance in MLB? Well, it’s pretty commonly known among baseball-thinkers that FIP is more accurate at predicting a subsequent year’s ERA than ERA itself. FIP is a statistic on an ERA-scale that only accounts for what the pitcher can control (strikeouts, walks, and home runs). There’s been a lot of research that looks at differences between ERA and FIP, but to my knowledge, there’s nothing out there to see if it can predict second-half performance. So that’s what I’m going to do here.

I compiled all the starting pitchers who were qualified in both the first and second halves of 2015 (57 total), and ran a basic scatter plot of their first-half ERA, FIP, and xFIP against second-half ERA, to see which of the former was best at predicting the latter.

First-Half ERA and Second-Half ERA


First up is first-half ERA and second-half ERA. A fairly weak correlation — 7% of a pitcher’s second-half ERA is explained by his first-half ERA — albeit significant (p-value < 0.10).

First-Half FIP and Second-Half ERA


Next is first-half FIP and second-half ERA. It’s hard to tell but the dots are, on average, a bit closer to the fit line — 11% of second-half ERA is explained by first-half FIP (p-value < 0.05).

First-Half xFIP and Second-Half ERA


Lastly, we have first-half xFIP and second-half ERA. While FIP uses a pitcher’s actual home-run totals, xFIP uses league-average totals because home run rates fluctuate year-to-year. You can clearly see the dots are much closer to the fit line than in the previous two graphs — 15% of second-half ERA is predicted by first-half xFIP (p-value < 0.01).

Is 15% good? Using the same method as above, I looked at the correlation between 2014 xFIP and 2015 ERA — and found an r² of 27%. So while half-season predictions don’t seem to be as accurate as season-to-season predictions, if MLB teams are making real moves based on a 27% correlation, I’m going to take a leap and say my fantasy team can makes moves based on a 15% correlation.

Now the part you (and I) have been waiting for: Here are the top 10 pitchers poised for second-half improvement followed by the top 10 pitchers who may get worse (sorted by the difference between ERA and xFIP, as of 7/9).

Screen Shot 2016-07-11 at 5.04.32 PM

Screen Shot 2016-07-11 at 5.06.34 PM

Some interesting things to note on the first list:

  • Smyly is owned in 48% of Yahoo Fantasy leagues, Nola in 47%, Ray in 11%, and Bettis in 4%. Pick them up.
  • The rest could be solid buy-low trade options (minus Eovaldi, unless your league values middle relievers).
  • A common theme among the members are high BABIPs and home-run rates (>.300, >15%) — which suggests they have been victims of bad luck.

And the second list, where the opposites are mostly true:

  • While Teheran’s name has come up in trade talks, his numbers suggest he may regress in the second half.
  • Sell-high trade options in fantasy leagues.
  • Low BABIPs and home-run rates (<.275, <10%).

Remembering Black Holes

Do you ever look at a daily lineup and find yourself disappointed with one of the names in it? Do you ever ask why the manager continues to bat a clearly inferior player when there are clearly better options on the bench or in the minors or in your softball league? Do you ever celebrate when a player gets designated for assignment and you never have to see them bat second in front of like six clearly better hitters? Well then I am very sorry, but it’s time to relive some bad memories, team by team, from the past ten years.

Yes, it’s time to talk about the black holes of the recent past.

Now what makes a player a black hole?

Read the rest of this entry »