Each year when it comes time for Hall of Fame voting we hear a lot about the problems with the voting process.  The Baseball Writers’ Association of America (BBWAA) has recently made some changes to address issues related to the qualifications of the voters.¹  However, other problems with the voting process persist.  From a psychometric perspective, a primary concern is that the ballot is norm-referenced, meaning other players on the ballot matter.  The issue of whether a player is a Hall of Famer should be based on their performance on the field and not based on whether they happen to hit the ballot with 10 other players who may also have potential Hall of Fame credentials.  As it currently stands, the Hall of Fame ballot is more about whether a player is a Hall of Fame-caliber player compared with the other players on the ballot and given that voters can only vote for 10 players.  That voters have a limited number of votes also imposes a ceiling effect and in years when there might be more than 10 Hall of Fame caliber players on the ballot, some might not get votes they would otherwise get.

The Rasch model² provides a criterion-referenced, sample-free method for analysis.  This means that it would be possible for voters to vote for as many players as they want regardless of who else is on the ballot without compromising the selection quality.  Furthermore, a player would never need to be removed from the ballot because they received too few votes and new voters could be added without changing the threshold for election.

In order to demonstrate this method I enlisted the help of 16 friends to cast votes on whether they thought each player was a Hall of Famer or not.  They simply answered Yes or No for each of the 32 players on the ballot.  If they weren’t sure they were allowed to leave their response blank, and since the Rasch model is robust to missing data, blank responses do not impact the player measures.  Tables 1 and 2 provide a summary of the responses for players and voters, respectively.

It is natural that the conceptualization of what constitutes a Hall of Fame player will vary by voter, with some being more lenient and some being severe.  Based on the severity of the voter and the ability of the player, the list of players will form a hierarchy.  This hierarchy is graphically represented in Figure 1.

Figure 1. Voter-Player map

Griffey received 16 Yes votes and one can see that he is at the top of Figure 1.  There were 9 players who did not receive any Yes votes and they can be seen at the bottom of Figure 1.  Hoffman is ranked higher than Clemens even though Clemens had more Yes votes (12 to 8).  However, Clemens received 15 total votes and Hoffman only 12, so Hoffman’s 8 votes were effectively worth more than Clemens’ 12 votes based on the severity of the 12 voters who actually provided a vote for Hoffman. Figure 1 also shows the severity of the voters with Voter05 and Voter14 being the most severe and Voter06, Voter07, and Voter15 being the most lenient.  Because these three voters only cast Yes votes and left the rest blank, they were shown to be very lenient since voters are only calibrated on the responses they provide.

In order to actually determine election to the Hall of Fame, a passing standard would need to be established.  This could be done by a variety of methods³ and could be carried forward each year so that the standard for election would remain the same for everyone.  Since the voting block from BBWAA is relatively stable, anchoring the voters’ Rasch calibration produces a stable scale in which voters can be added and removed easily without changing the passing standard.

I mentioned earlier that a player would not need to be removed due to an insufficient vote tally, but 9 players here did not receive any Yes votes.  It would seem natural that these players would be removed from the ballot to make room for others coming on so that the ballot did not become so large as to put an undue burden on voters.  However, statistically speaking, it doesn’t matter.  Once voter calibrations are anchored the number of players on the ballot becomes irrelevant.  The score scale and passing standard would be the same if the ballot was one player or 100 players.

Needless to say, this is a simple demonstration using a non-representative sample.  It would, however, alleviate some of the issues that plague the voting process.  The discussion would then hinge on a player’s record and not on the intricacies of the ballot.  Borderline players would not be dismissed simply because they were the 11^th best player on the ballot that year and voters would be free to vote for any number of players they felt fulfilled the criteria of being a Hall of Famer.

References

1. http://baseballhall.org/news/hall-of-fame-announces-change-to-bbwaa-voting-electorate
2. Rasch, G. (1960). Probabilistic models for some intelligence and attainment tests. Copenhagen, Denmark.: Danish Institute for Educational Research.
3. Cizek, G. (2012). Setting Performance Standards. New York: Routledge.

2015 was the year of Bryce Harper. He led qualified hitters with a 197 wRC+, the highest since the turn of the century among players not named Barry Bonds. This was a vast improvement on his already-impressive 2014 season, in which he totaled a 115 wRC+.

Depending on how you look at things, you could say Bryce Harper was the most improved batter in 2015. I choose not to for two reasons: 1) it’s too easy, and 2) it makes this article more fun. There’s also another more objective reason: with only 395 plate appearances in 2014, Harper didn’t qualify for the batting title.

This poses a question: what minimum do we set to determine who improved the most between 2014 and 2015? If we say that the player needed to qualify for the batting title each year, we get Chris Davis as the most improved batter, who increased his wRC+ from 94 in 2014 to 147 in 2015. If we set no minimum, our wonder-boy is none other than notorious slugger Carlos Torres, the Mets pitcher who upped his wRC+ from -100 to 491.

Clearly, there needs to be some minimum. For the purpose of the article, I’ve decided to set it at 100 PA. This seems a reasonably small enough number to include a wide array of players, but large enough to get rid of anomalies (I’m looking at you Carlos). When we set this minimum, we discover that the batter whose wRC+ increased the most between 2014 and 2015 is… Ryan Raburn. However, since Jeff Sullivan already talked about Raburn, I decided to go with the next name on the list: J.B. Shuck.

If you don’t know who that is, I don’t blame you. I didn’t until I started this research. If you do know him, I’m going to guess that you’re either a White Sox, Indians, or Angels fan. Either that, or you have more time to watch baseball than a college student taking a full course-load of credits. Who’s to say?

The reason the casual fan might not know Shuck is because, well, he’s not exactly a star player. Here are the players with the lowest wRC+ in 2014 of those with at least 100 PAs:

That’s right, he was literally the worst batter that year. Almost as bad as if I were to join the majors. It should be no surprise, then, that he was able to improve so much — he had the lowest starting point. Even so, he still had needed to improve quite drastically in order to surpass Harper’s wRC+ improvement. And that’s exactly what he did:

In 2015, Shuck improved so much that he almost managed to be an average player. But how did he manage to do it? Was it a matter of luck, or did he actually get better?

The number that stands out the most in Shuck’s 2014 season is his .146 BABIP (batting average on balls in play). For those of you that don’t know, that number is quite bad. Like, less than half of what it should be. His BABIP in other seasons is right around league average, so something must have gone amiss last year. Looking at the underlying numbers, some things showed up:

So. His FB% and Pull% numbers were way up as compared to other years. For some context, the league-average FB% has been approximately 34% the past two years, while Pull% has been approximately 40%. These numbers suggest that Shuck spent too much time trying to pull the ball over the fence two years ago, and the video suggests the same thing. Here’s an example of him trying to do just this to a pitch on the outside corner, but instead weakly grounding to first. You can see how he opens his hips before he even starts his swing, forcing him to simply slap at the ball if he wants to make any contact:

And here he is in 2015, driving a similar pitch into left field:

The cause of his change in approach is hard to say. He did get a new hitting coach to start off the year, switching from Jim Eppard to Don Baylor. From 2013 to 2014, the Angels as a team increased their FB% from 33% to 34% and their Pull% from 37% to 42%, so that argument does have some merit. Regardless of the reason, it’s clear that it had an effect. Here’s Shuck’s ISO by zone:

As can be seen on the left, Shuck had trouble hitting anything not on the inside edge of the plate in 2014. This past year, he learned to control more of the strike zone, and even though there’s less red than there was in 2014, there’s also a lot less dark blue. Shuck drove the ball from all parts of the zone to all parts of the field, and his numbers improved because of it.

While Shuck may not be an All-Star anytime soon, his year-to-year improvement is truly remarkable. If he can go from being the worst hitter in baseball to an average one, anyone can. And if that doesn’t inspire the Brendan Ryans of the world, I don’t know what will.

A few days ago, Joe Posnanski wrote about the possibility of Bryce Harper getting the first $500m contract ever. I agree with him on how both amazing and ridiculous it would have sounded 2 or 3 years ago. I also agree it is possible, almost likely, to happen. I might not be a Bryce Harper fan but he is so young that is he is on track to accomplish big things. He is not Mike ‘King’ Trout but he is very good.

Harper’s current contract runs through the end of 2018, which is when I assume he would get the big fat check. The Nationals will try to extend his contract before he is a free agent, just like the Marlins and the Angels did with Giancarlo Stanton and Mike Trout. However, in this post we will assume Harper will not pursue that path, making him a highly-coveted free agent in 2018. I will also exclude the possibility of 9- or 10-year contracts, which would make the mark easily achievable. Let’s run the numbers for Harper’s future:

We have here Harper’s projected value profile. As usual, I am using FanGraphs’ model, which has a player’s aging curve that follows +0.25 WAR/year (Age 18-27), 0 WAR/year (Age 28-30),-0.5 WAR/year (Age 31-37),-0.75 WAR/year (38 and older). It also assumes that open-market WAR sits at $8.4m in 2016 and grows at 5% per year. The starting point is Steamer’s 2016 projection: 6.8 WAR.

Three years from now, in the winter of 2018, he will be negotiating his new contract that includes his theoretical peak 27-30 years at ~7.8 WAR/year. The truth is that a 7-year / $500m+ contract would only be likely if by 2018 he can position himself as a player who consistently accounts for almost 8 wins per year. That is the only reason a team would be eager to invest half a billion dollars in a single player, marketing-related reasons aside.

Now, the question comes down to what he needs to do by 2018 in order to cement that positioning. The model needs him to be a 21.2-win player during the next 3 seasons. While Harper might have taken a significant step up performance-wise, we need to remind ourselves that before 2015 he was “just” a ~4-5 WAR guy. In order to meet the model’s expectations he needs to double those numbers, and remain at that level  for 3 years in a row (i.e.: Between 8-9 WAR for that 3-year period). If he meets those marks, Harper would have accrued 40 WAR during his career by 2018. While that is entirely possible, it is not easy. This is the list of highest cumulative WAR by age-25:

So, two conclusions can be quickly drawn. First, Mike Trout is not human. He is only 24 years old and is already on this list with guys like Cobb, Mantle, Foxx and company. Second, no, it is not an easy task for Harper. I know that you are thinking that he just put up a 9.5 WAR season, why can’t he do it again? Another season like that and he should get to his target easily but, truth be told, those Trout-esque seasons are unlikely to happen. I say this for three main reasons. First, Harper is not an elite defender and has gotten worse every year. For the last 3 seasons (2013-2015), he ranks 37^th in UZR/150 out of 60 qualified OF. In 2015, he compiled -8.5 on Defense (Def) metric, per FanGraphs, which is position-adjusted, in his case for RF. Out of the 69 individual seasons with 8 or higher WAR from players 25 or younger, only 5 players (Hank Aaron, Ted Williams, Stan Musial, Mike Trout (!) and Bryce Harper) had -8 or worse Defense. No, it is not impossible but it is hard.

Second, he is an above-average baserunner, but not an awesome one. Lastly, Harper has not exhibited good health over his career. He has had injuries in 2 out of 4 seasons, which may not seem many but in 2013 and 2014 he only played 67% of Nationals games. Predicting health is tough, especially because there are unforeseeable events. You cannot control a hit by pitch at your wrists or a concussion sliding in second base but your health track record is your best bet on your future injury report. Those three things are vital to get to 21.2 WAR during the next 3 years. Harper needs those factors to come in play in order to get to the 7yr/$500m contract. Harper’s advantage is his age – just like Jason Heyward this offseason.

We have implicitly talked about Mike Trout. He is arguably the best player in baseball right now and was on track to smash the contract record, until he negotiated a 6yr/$144.5m contract extension. That will keep him locked up from ages 24 to 29 at LAA. Now, the question is what type of contract will he command in 2020? Mind you, it is hard enough to try to predict what a Free Agent might get in 2016, but still we took a stab a it.

Here is Trout’s projection. Again, 2016 WAR is courtesy of Steamer. We might think the aging curve slightly benefits Trout because it forecasts a ~10% increase in WAR, and he has not posted those 10 WAR seasons since 2 years ago. Then again, let’s toy with the idea. The $500m contract here seems more feasible for three reasons. First, in MLB, you get paid for what you did and not for what you will do.  By 2020, Trout could have ~85 WAR under his belt –he would be 28 years old. That is just ridiculous and will not happen, right? No one, ever, has done that by age 28 (Ty Cobb is the leader with 78.6 WAR). But what if he does? What if Trout is around the 70 WAR mark with 8 or 9 great seasons on his resume? Second, he needs to do what he has already done e.g. Trout has posted two +10 WAR already. The other two seasons were 8 and 9. This guy runs well and plays above-average defense. Trout does it all and will not stop. Third, unlike Harper, Trout has been very much healthy. During the 2013-2015 period, he played 157, 157 and 159 games, respectively. Again, injuries are hard to predict but we will take what he has shown so far as a given, which is good health. Fourth, fair to say, time value of money. A dollar today is not worth the same as a dollar tomorrow. Therefore, getting a $500m contract in 2020 should be easier than in 2018.

In summary, I think Harper can do it but I would not bet on it. From my perspective this is a long shot. If you ask me today on who is more likely to become baseball’s first 500-million-dollar man, I would put my money on Mike Trout to beat Bryce Harper on this as well.

Note: This analysis is also featured in our emerging blog www.theimperfectgame.com

Only a few of the major free-agent names remain available as we approach the new year. One of the most intriguing is Alex Gordon. He’s not only been an excellent fielder over the course of his career, but he’s also been an above-average hitter. His age-25 and 26 seasons were cut short by injury and I think we can give some leeway to a 23-year-old rookie for not having an above-average bat, but otherwise he’s had an excellent career. Here’s are his stats throughout his career:

There’s no doubt that he’s a great baseball player. He’s also accumulated 3 seasons with 6+ WAR since his rookie season. But there’s always the question as to whether a player has peaked or not, especially when their age starts creeping into the 30s. To try and answer this I look at the OPS values he’s put up over the years and extrapolated those numbers into his age-40 season. Below, in black, are the seasons that he’s already played. I’ve also included a line-of-best-fit through the data with the black portion representing past seasons and the red portion representing his future offensive output. Based on the seasons he’s put together, the model predicts that he will peak at about 34 years of age. Most players peak in their late 20s, but it’s not unheard of for players to peak later. Projections should always be taken with a grain of salt, but whichever team decides to take a shot on Gordon could expect his offensive production to remain relatively constant over the next few years.

So what does this graph tell us? Well basically nothing! It’s not very good practice to extrapolate past the range of your data, but it is interesting nonetheless. Also, considering Gordon has been so good for so long it’s tough to assume that he hasn’t peaked yet. That’s not to say he can’t continue to improve or even perform at a high level, but since it’s getting later in the offseason and so much money has been thrown at pitchers let’s assume he signs for 4 years. Below are his projected OPS values and as you can see from the graph above that Gordon may not even be in his offensive prime.

So far I’ve shown you data for Gordon’s career and also used that data to project his performance over the next 4 years. Assuming he signs a 4-year contract this off-season I wanted to find his closest comparables from his career so far and see how those players performed through their age-35 season. In order to compare players I used the Mahalanobis distance for all players that fell into the following criteria; (1) played in every season from their age 29 to 31 seasons, (2) at least 1200 ABs over that time and (3) played every season in their age 32-35 seasons. The Mahalanobis distance was calculated using common offensive statistics standardized by the number of at-bats. Here is a table with the lowest Mahalanobis Distance’s to Alex Gordon through his career thus far as well as their cumulative WAR for their age 32-35 seasons.

Last year Dave Cameron broke down the cost for WAR here and found the number to be somewhere around $7 million. Tim Dierkes projected a 5-year, $105-million contract or roughly $21 million per year. In order to live up to that annual salary, he would have to produce about 3 WAR per season which is 12 WAR for a 4-year contract and 15 WAR for a 5-year contract. Melvin Mora, Jim Edmonds, Ken Caminiti and Lou Whitaker each exceeded that 3-WAR threshold.

As this offseason progresses, offers will undoubtedly be presented to his agent so now it’s only a matter of when he signs. Based on the players that he was compared to, Alex Gordon definitely has the potential, not to mention the ability to exceed the standards of the contract he inevitably signs.

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. Consequently, Frankie Frisch is listed on the Giants roster for the duration of his career while the Indians declare Rocky Colavito and the Mariners claim David Ortiz. 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 finest single-season rosters for every Major League organization based on overall rankings in OWAR and OWS along with the general managers and scouting directors that constructed the teams. “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.

Terminology

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

Assessment

The 1969 Cincinnati Reds          OWAR: 58.1     OWS: 362     OPW%: .619

Based on the revised standings the “Original” 1969 Reds recorded 100 victories and claimed the National League Western Division by 14 games over the Giants. Cincinnati topped the circuit in OWS and OWAR. GM Gabe Paul acquired 27 of the 40 ballplayers (68%) on the 1969 Reds roster.

Pete Rose (.348/16/82) notched his second straight batting title and paced the League with 120 runs scored. “Charlie Hustle” rapped 218 base knocks including 33 doubles and 11 triples while establishing personal-bests in OBP (.428) and SLG (.512). Jim Wynn aka the “Toy Cannon” unleashed 33 bombs, nabbed 23 bags, tallied 113 runs and topped the circuit with 148 bases on balls. Frank “The Judge” Robinson (.308/32/100) registered 111 aces and finished third in the MVP balloting. Third-sacker Tony “Big Dog” Perez belted 37 round-trippers, knocked in 122 runs and merited his third consecutive All-Star invite. “The Little General” Johnny Bench swatted 26 big-flies and drove in 90 runs during his sophomore season. Lee “Big Bopper” May crushed 38 moon-shots and plated 110 baserunners to earn his first appearance in the Mid-Summer Classic.

Johnny Bench places runner-up to Yogi Berra in the All-Time Catcher rankings according to Bill James in “The New Bill James Historical Baseball Abstract.” Teammates listed in the “NBJHBA” top 100 rankings include Robinson (3^rd-RF), Rose (5^th-RF), Wynn (10^th-CF), Perez (13^th-1B), Vada Pinson (18^th-CF), Curt Flood (36^th-CF), May (47^th-1B), Leo Cardenas (50^th-SS), Johnny Edwards (53^rd-C), Tommy Harper (56^th-LF), Cookie Rojas (69^th-2B), Cesar Tovar (79^th-CF), Tony Gonzalez (82^nd-CF) and Tommy Helms (99^th-2B).

Mike Cuellar (23-11, 2.38) earned the Cy Young Award while fashioning the lowest WHIP (1.005) of his career. Claude Osteen (20-15, 2.66) delivered career-bests in victories, innings pitched (321), strikeouts (183) and WHIP (1.143). Jim Maloney contributed a 12-5 record with a 2.77 ERA and Casey Cox (12-7, 2.78) furnished strikingly similar statistics. Diego Segui anchored the bullpen with 12 wins, 12 saves and a 3.35 ERA.

The “Original” 1969 Cincinnati Reds roster

Honorable Mention

The “Original” 1974 Reds                 OWAR: 52.6     OWS: 336     OPW%: .557

Cincinnati scrapped with Atlanta in the final weeks of the season. The Braves emerged with the division crown by two games while the Reds paced the National League in OWAR and OWS. Johnny Bench (.280/33/129) scored a career-high 108 runs and topped the RBI charts. Jim Wynn walloped 32 circuit clouts, drove in 108 baserunners and amassed 104 tallies. Pete Rose’s batting average dipped below .300 for the first time in ten years. All the same, “Charlie Hustle” paced the circuit with 45 doubles and 110 runs scored. Dave Concepcion earned his first of five Gold Glove Awards and contributed a .281 BA with 14 wallops and 41 steals. Hal McRae (.310/15/88) responded with 36 doubles after earning a full-time role. Ross “Scuz” Grimsley furnished an 18-13 record with a 3.07 ERA.

On Deck

The “Original” 1939 Yankees

References and Resources

Baseball America – Executive Database

Baseball-Reference

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

Seamheads – Baseball Gauge

Sean Lahman Baseball Archive

This week we will be following up on our previous piece regarding Least Valuable Players. There we identified Hanley Ramirez and Pablo Sandoval as the two worst performers of the 2015 season. We are not fans of beating the proverbial dead horse, but given that this was just the first year of their respective contracts we were interested in figuring out just how bad these deals are shaping up for the Red Sox.  The short answer? It is bad. Like, Lucas Duda throwing to home in high-pressure situations bad.

As you might recall, during the 2015 season Sandoval accumulated a -2 WAR whilst Hanley finished up with a -1.8 WAR.  Even after that woeful start, the Red Sox are still on the hook for 5 more years and $89.4 million for Sandoval, and 4 years and a total $90.25 million for Hanley. Just let that sink in for a minute: 9 seasons of potentially below-replacement-level performance for $180 million. That makes the Barry Zito deal sound like a real steal.

For the sake of argument, they do not have to be that bad for the rest of the contract, do they? I mean, these guys were 3-win players just two seasons ago; maybe this was just a hiccup. Well, Steamer seems to partially agree with this logic and projects them to improve substantially. More precisely, it projects Sandoval to have 1.8 WAR and Hanley 2.2 WAR during 2016. Returning to these levels of performance is something, but is it enough to salvage these deals?

We replicate the player assessment analyses we used in our piece comparing offseason splurges in pitchers, just to figure out the net value of these deals.  We use Steamer’s 2016 projections as a starting point for WAR and then apply a player aging curve that goes as follows:  WAR increases annually by +0.25 for ages 18-27, stays flat for ages 28-30, decreases annually by -0.5 for ages 31-37 and lastly decreases annually by -0.75 for ages 38 and above. With regards to the market value of wins we start off at $8 million per win and we apply a 5% yearly inflation rate.

Even after considering the improvements suggested by Steamer, none of the 9 seasons controlled by the Red Sox would produce a net positive value, and overall the net loss of these deals comes at $79.4 million.

We ran the numbers, and in order for the Red Sox to recoup their investments, even after letting 2015 go down as a sunk cost, Sandoval would have to accrue 10.35 WAR for the rest of the contract (73% more than the projection), whilst Hanley would have to accumulate 10.56 WAR (82% more than the projection). This seems to be rather unlikely, especially when you consider that the Steamer projection already seems bullish, implying a 4 WAR improvement between seasons.

We wanted to test just how bullish this prediction is. We set out to find the past seasons most like the ones Sandoval and Hanley just endured and tried to identify how those players fared off the year after as well as for the rest of their careers.  We searched the last 30 seasons for players between the ages of 28 and 32, that produced a -1.5 WAR or worse in at least 400 PA, after accumulating at least 5 WAR in the previous two seasons. Namely, we were searching for players that had been performing at a high level, still in their prime or early phases of decline, which suddenly plummeted in performance.

Lo and behold the mother of small samples. We found just 5 players that met these requirements, 4 of them are already retired and one of them, well, one of them also plays for the Red Sox.  Out of these five players four of them improved after their decline season, the other one was out of the game. Out of the ones that improved, only one, Joe Carter, was able to meet the 4 WAR improvement inherent to the Steamer prediction, actually he was the only one that was better than replacement level after the decline season. So far Joe Carter has also been the only one able to play more than one season in the majors after the decline, with the jury still out on Allen Craig.

Just how good were Joe Carter’s first five seasons after the decline? Well he won back-to-back World Series with the Blue Jays, starred in one of the most memorable moments of baseball history and amassed a total of 9.4 WAR; a figure similar to what would be required for either Hanley or Sandoval to break even in their contracts. Just how bad is the alternative? Another season of negative WAR (-1 is the average for those not named Joe Carter) and 0 WAR from then on; a scenario like this would produce net value losses for the Red Sox close to $200 million or 150% more than what emanates from the Steamer scenario.

I know that we are dealing with extremely small sample sizes, but entertain this thought for a second. Let’s imagine that the above players represent the universe of possibilities and hence Pablo and Hanley each have a 20% chance of becoming Joe Carter and returning a net value of 0, that means breaking even and getting fair value for investment, and 80% of teetering off and producing a net loss of around $100 million.  Under that scenario the expected value of keeping both players comes somewhere at a net loss of $160 million over the life of the contracts. That is not necessarily crippling as it translates roughly to 3-4 lost wins per year, but these bad decisions can find a way to add up quickly.

Based on this, the Red Sox would certainly welcome another Dodger bailout, however this time around they might have to add additional value for a deal to go through. Moving forward the Red Sox might want to pursue one of three alternatives. First off, they might use that $160 million expected net loss value as an upper bound of how much they would be willing to send (in either money or player value) to another team as compensation for taking these contracts off their hands. Secondly, they could settle on the Steamer projection and set that upper bound on $80 million. Lastly, they could try to make the other team believers of the Joe Carter dream, and try to get away with not sending anything else, and even hoping to get something of value back in return, but this seems rather unlikely.  In theory by sending something (money or players) of less value than those upper bound figures to facilitate the deals they would be effectively cutting their losses.

With regards to the debate between sending some money or a player with value to make the deal work, it should be noted that $160 million in net player value over 5 years is something like Xander Bogaerts and a Top 25-50 prospect. Despite all the good will that recent deals have gained Dombrowski, there is no way Red Sox Nation would look kindly into giving up that kind of talent just to undo a mistake. The Red Sox are looking to become consistently competitive for the years to come and it does not make much sense to mortgage the team’s present and future by giving up so much controllable high-end talent. It may be time for the Red Sox to leverage their financial fortitude, bite the bullet, subsidize part of the contracts if need be, and move on.

A question that tends to pop up around this time of year: “When does fantasy baseball season start?” Of course, we all know that fantasy-baseball season never ends, especially for those of us in keeper and dynasty leagues. To wit, Brad Johnson’s “Keeper Questions” thread posted just the other day is now sitting at 350 comments and growing. As we all collectively count the days ‘til spring training and opening day, one of the most oft-discussed and most subjectively-answered topics is “Who do I keep?” Fantasy baseball players intuitively understand the idea of aging, at least qualitatively. Older players are less valuable, given that their performance is more likely to decrease due to both injury and ineffectiveness. But how much is age worth, really?

Read the rest of this entry »

Say “John Hart” and many baseball fans will immediately think of a two-word phrase: “Cleveland Indians.” Hart made his name with the wider baseball public by skillfully transforming the perennial doormat into a juggernaut. From 1995 through 2001, the Indians finished first six times and second once. They went to the World Series (and lost) twice. The Indians last postseason appearance prior to 1995 was the 1954 World Series, when they got swept by the New York Baseball Giants.

Say “John Hart” and some baseball fans will think of another two-word phrase: “contract extensions.” Hart was a first-mover in employing the tactic of buying out a player’s arbitration and early free-agency years, paying a little more now in exchange for a lot less later. This was part of Hart’s broader strategy, useful anywhere and necessary in Cleveland, of squeezing the maximum value out of every dollar spent.

Say “John Hart” and almost no baseball fans will think of yet another two-word phrase: “senior citizens.” One of Hart’s less-heralded strategies was raiding the top end of the aging curve, signing players well past their born-on dates to patch the numerous holes in Cleveland’s roster that a decent but top-heavy farm system couldn’t fill. In 1995, Dennis Martinez tied for the second-best pitching season in baseball history by a 41-year-old.

In a coincidence proving that our lives are governed by powerful yet unseen forces far beyond our comprehension, Hart now finds himself in charge of the other Native American themed major-league franchise. (I strongly advocate renaming the Cleveland franchise thus, but that’s a topic for a different post.)  Hart’s experience in Cleveland will no doubt shape his approach to remaking the Braves into a contender, but the challenges he faces in Atlanta are in some ways more daunting, and the solutions he employed in Cleveland may be less effective today.

The Spiders team that Hart took over in 1991 already had most of the high-impact players that would power the team to its seven years of dominance. Here are the starting 8, the starting DH, and the rotation for the 1995 team, along with the player’s age and bWAR that year. A “+” indicates a player Hart obtained.

C      Tony Pena (38/0.3) +

1B    Paul Sorrento (29/0.4) +

2B    Carlos Baerga (26/2.6)

3B    Jim Thome (24/5.9)

SS    Omar Vizquel (28/1.4) +

LF    Albert Belle (28/6.9)

CF    Kenny Lofton (28/4.1) +

RF    MannyBeingManny (23/2.9)

DH   Eddie Murray (39/2.4) +

P       Dennis Martinez (41/5.7) +

P       Charles Nagy (28/2.4)

P       Orel Hershiser (36/3.7) +

P       Mark Clark (27/0.6) +

P       Chad Ogea (24/3.2)

Baerga, Belle, Ramirez, Thome: those Four Horseman of Lake Erie (ok, fine, you try making a metaphor) were already in the house when Hart took over. He added two critical pieces to the lineup, however. Quickly deciding that Alex Cole wasn’t the answer to any baseball question worth asking, in late 1991 Hart obtained Kenny Lofton from the Astros for … well, go ahead and click to find out. Lofton was traded six times in his career, and in five of those trades the team receiving Lofton committed larceny.

It took Hart longer to give up on shortstop (and former second overall pick) Mark Lewis, but after 800 ineffective plate appearances, Hart had seen enough. Recognizing that this guy has made Seattle’s Omar Vizquel redundant, Hart reeled him in for the low, low price of Felix Fermin and Reggie Jefferson.

That still left numerous vacancies on the major league roster, and Hart set about filling them by purchasing AARP’s mailing list for Northeast Ohio. Pena, Murray, El Presidente, and Hershiser were all old enough to know their way around the bingo parlor, and Hart got value from all of them except Pena.

Hart made two key additional moves, bringing in failed starters Jose Mesa and Eric Plunk and showing them immediately to the bullpen. The two combined for a whopping 6.0 bWAR in 1995.

And then those contract extensions!  Below are the player’s maximum salary with and after playing for the Indians, 2015 dollars (millions), as well as their ages during their last season in Cleveland:

________             Hart           Age                Hartless

Lofton                       5.3               29                    10.9

Ramirez                    6.2               28                   28.1

Baerga                       7.1               27                      7.0

Belle                          8.6               29                    17.6

Thome                      11.6              31                     17.1

Hart struck gold in four of the five cases – except for Baerga, these players’ salaries skyrocketed after they escaped the Cleveland contracts. Baerga was a misfire – he peaked very early and the first year under his new contract (1993) was the last year he would be dominant. In the other four cases, however, Hart got the players’ best years at a relative discount, and then allowed his competitors to overpay for the decline years.

Not that Hart avoided older players entirely – as we’ve seen, he prowled Sunset Acres with almost sinister determination. The years he didn’t want to pay for were the early to mid 30s; those were the years in which he seemed to think that market inefficiencies most significantly favored the players. Before that window he could get maximum performance, and after that window he could get veterans at discounts reflecting the players’ acute awareness of their own career mortality.

Vizquel is the obvious exception, though even here Hart got a bargain. Vizquel’s salary maxed out at $7.5 million  with the Spiders (in 2015 dollars), astonishingly low for a player who, while he probably doesn’t belong in the Hall of Fame, would hardly be an absurd choice. Vizquel played for Cleveland from age 27 through 37, thus encompassing many of the very years Hart avoided with others.

Here Hart was perhaps exploiting yet another market inefficiency, the bat bias. Vizquel never really hit – he had just two years with a wRC+ over 100, and his career number is 83, which isn’t that great even for a middle infielder. But oh, could he field. Only four active shortstops have played more than 2000 innings and have a better UZR/150 than Vizquel’s career 8.7. Vizquel’s glove was solid gold, and his relative weakness at the plate meant that Hart could buy that gold at a discount.

The system Hart inherited in Atlanta had less talent than Cleveland’s in 1991, though Hart has set about remedying the situation. From last year’s regulars, only Freddie Freeman figures to be on Atlanta’s next postseason team. Atlanta’s system has three prospects in the top 10 of their respective positions, according to MLB Pipeline: Dansby Swanson and Ozhaino Albies (both shortstops) and Sean Newcomb (LHP), with Swanson and Newcomb being Hart imports. Indeed, Newcomb came over in exchange for one of those better-than-Vizquel shortstops, Andrelton Simmons, he of the career 21.4 UZR/150; a phenomenal figure but one of arguably less relevance since Omar’s day thanks aggressive defensive shifting. (This isn’t necessarily to say Simmons’ number is inflated by shifting, but rather that less range-y guys might provide relatively more defensive value than previously thanks to the shifts.)

Assuming Hart keeps them both, Swanson will move to third or (less likely) Albies will move to center; his bat is unlikely to carry any other position. Atlanta’s upper levels have little obvious offensive potential, with center fielder Mallex Smith being a conspicuous exception. Though still largely a stranger to top-100 prospect lists, Smith has a career .768 OPS in the minors, unaided by the PCL, and will be just 23 this year. He struggled in AAA last season, but overall looks like he could be a useful speed-oriented center fielder. And he got a big up here.

So the outlines of a playoff core are in place: Freeman, Albies, Swanson, Smith (or perhaps Ender Inciarte, another recent Braves acquisition), and Newcomb. Long-term extensions, anyone? Well, let’s see, Freeman already got his: he’ll be pulling in $22 million in 2021 in a backloaded deal that looks somewhat risky, though it ends at age 31. Swanson, Albies, and Smith will have to wait until they demonstrate some ability in the majors, but the chances that Hart can get away with low AAV contracts through the players’ late 20s seem slim.

In 1994 Hart’s contract extensions seemed like a gamble, but today they look like bargains for the team. Few agents would want to be associated with these kinds of contracts unless the player needs to give a character discount (paging Aroldis Chapman). Indeed, Freeman’s contract may be the model here – a great deal for the team in the early years, while the player claws some of it back toward the end.

With the Indians Hart seemed to generally eschew long-term contracts for pitchers – the limited information I’ve found suggests that he never went beyond four years, though often with a salary-boosting club option (see, e.g., CC Sabathia and Bartolo Colon). So perhaps Newcomb can look forward to one relatively team-friendly 4-year deal to be followed by truckloads of cash from another team. One of Julio Teheran, Aaron Blair, and Touki Toussaint will probably fill the two spot.

As for the rest of the rotation, there are a lot of guys competing for probably two spots (the guys just mentioned, plus Manny Banuelos, Mike Minor, Matt Wisler, Mike Foltynewicz, and maybe three or four guys in the minors). Again, some of these guys may get one 4-ish year deal before moving on. On the other hand, good pitchers today will probably seek at least 5 or 6 years unless, again, there are character or injury issues militating in favor of a discount.

You know his methods, Watson – Hart may attempt to fill any remaining rotation holes with old but talented pitchers. Expect the same for the lineup, but using the young players’ sweet contracts to subsidize those of the veteran imports may be more difficult now than it was in the 90s, since the youngsters are going to leave less money on the table than the old (young) Indians did.

Hart enjoys one modest but rapidly deteriorating advantage: the NL East is a tire fire right now. The purported contending teams (Mets, Nats, maybe the Marlins) are more dysfunctional than Springfield’s nuclear plant. The Phillies have done an admirable job of remaking the front office and the farm system, but the team is still a few years from contention. This sorry window is closing, though. All of Atlanta’s competitors (except the Marlins) have more money to spend than Atlanta does; their dysfunction won’t last forever (except the Marlins). And yes, the Mets have, or should have, money.

Say “John Hart” to Atlanta fans in 2019 and maybe they’ll say “World Series!” But the mountain is steep – today he faces better-informed players and more uniformly competent GM competitors, all armed with big data that was only beginning to come into view in the mid 1990s. Perhaps Hart will lead his troops to fight like Scotsmen; to succeed, they’ll probably need to.

As I noted a few weeks ago, batters who were at least semi-regulars in both 2014 and 2015 were less effective in 2015 than in 2014, as measured by wRC+. That seemed directionally unsurprising — after all, players are subject to aging and regression every year — though the magnitude (an average decline of over five wRC+ points, or over four weighted by plate appearances) was a little higher than I’d expected. Was that decline, I wondered, unusual?

To answer, I calculated the change in wRC+ from one season to the next for players with at least 350 plate appearances in each season. I looked at every year from 1969 (four-team expansion, beginning of divisional play) to the present. (Fine print: I didn’t prorate my results for strike-shortened seasons, and I combined both leagues, with their different DH rules for most of the seasons, in the study. We’re looking at over 10,000 player-seasons, so small variations like the 1994 season and the four years in which the AL didn’t have a DH don’t amount to a lot.) Here are the results, with the second year of the pair of the x axis:

This graph should elicit two responses: (1) it looks as if year-on-year performance is declining, and (2) that is one noisy graph.

So I did another graph, taking the rolling three-year average change instead of the single-year change. Again, the second year of the pair is on the x axis, so 1972 refers to the average change for 1969-70, 1970-71, and 1971-72:

That’s less noisy, but it doesn’t change the conclusion: the year-over-year decline in offensive performance is the steepest it’s been in the nearly 50 years since divisional play began. I’ll use rolling average graphs for the remainder of this article.

The obvious question is: Why? What has changed that’s caused players to be nearly four points worse in terms of wRC+ in recent years when the long-term average decline is less than two, and hovered in a range of 0-2 in most years?

The first possibility that came to mind: Is it an age thing? Are players exhibiting different characteristics based on their year of birth? I divided the batters in my sample into four categories: Young (younger than 25 in the first season of the pair), Prime (25-29), Late Prime (30-34), and Old (35 or older). Here’s the decline in wRC+ for Young players. I used five-year moving averages, since limited sample sizes made the three-year moving averages pretty noisy.

Young players have been getting better, not worse, in consecutive years. That makes intuitive sense: we’d expect batters to improve a bit every year up to their peak in their late 20s. So youngsters aren’t the reason batters appear to be falling off more, year over year.

How about Prime years:

That’s the same scale as the last graph. This is a classic “You can go about your business, move along” graph. There’s been no notable change here. Batters entering their prime years have improved by about 1.5 wRC+ points in consecutive years, year-in, year-out.

Late Prime players:

Now we’re seeing declines, along with more noise. Players under 30, on average, improved their wRC+ from one year to the next. On the other side of 30, we see decline start to set in, to the tune of about a 3.8-point wRC+ average. And it’s gotten worse over the last ten years, rising from an average of about 3.1 in 1986-2005 to 4.1 in 2006-2015.

But we haven’t explained the problem yet. There’s nothing in the prior three graphs that would explain why the decline in wRC+ from one season to the next for semi-regular players has risen by over two points, because none of the prior three age groups has fallen off sharply. One more group left; let’s look at the Old players, 35 and up:

Whoa. That’s pretty dramatic. Year-over year, old players who are semi-regulars are declining a lot more now than they have been at any time since the mid-1970s, when trotting out the fossilized remains of Henry Aaron, Deron Johnson, and Billy Williams to play DH seemed like a good idea. This is the noisiest graph I’ve showed you so far, due to the limited number of older players in the game each year, but the marked climb since the 1990s is unmistakable.

Why is that? What’s happening to guys 35 and older? Nothing exactly leaps out, so here are some possible explanations:

Steroids. Admit it — that’s the first thing you thought. Same here. Fifteen or so years ago, you had all these guys in their late 30s putting up .300/.400/.500 lines with a couple dozen (or more, a lot more) bombs. Or at least it seemed that way. And sure enough, the five-year moving average decline in wRC+ for players aged 35 years or older was below the long-term average decline of about five wRC+ points for all but two years between 1989 and 2004. I think this points to a possibility of chemically-delayed aging patterns that have returned to normal, or perhaps even gotten worse.

More old guys. It’s not a secret that baseball players are better when they’re young than when they’re older. But, as noted above, the Steroid Era featured a lot of old guys hitting the crap out of the ball. Maybe that changed the thinking regarding roster construction, and teams are still carrying a lot of older hitters, even though they’re no longer as effective. Well, here’s a graph showing the percentage of players with 350+ plate appearances per season who were 35 or older.

No, GMs aren’t nostalgic for baseball in the late 1990s and early 2000s. There are fewer older players with regular or semi-regular roles today now than at any time over the past 20 years.

Worse old guys. Maybe the problem is just one of quality. Maybe older players today just aren’t as good as they were in years past. Maybe there was something about babies born in the 1970s. (Disco? The clothes? Watergate?) Here’s a chart showing players who were at least semi-regulars in consecutive seasons, aged 35 or older in their first season, and their wRC+ in their first and second seasons.

Nope, the older guys who’re good enough to get at least 350 plate appearances are still good players. They’re just getting worse faster, as evidenced by the widening gap between the red and yellow lines above.

Amphetamines. In baseball, the term performance-enhancing drugs is synonymous with steroids (and, to a lesser degree, HGH) in the public mind. But the list of banned substances is long, including all manner of illegal recreational drugs and, of relevance here, stimulants. Amphetamines — greenies, in baseball vernacular — have been associated with the game dating back to at least the 1960s. Baseball, of course, has a long season, with many more games than any other North American sport. Amphetamines help players improve reaction time, focus, and ward off fatigue. Those benefits accrue to everyone, of course, but they seem particularly relevant to older athletes, who face the inevitability of the aging process, mentally and physically. The amphetamine ban, which began in 2006, has likely had a larger impact on older players than younger ones. Of course, we’re talking about ten years of amphetamine testing, while the decline in older hitter year-on-year performance has lasted longer, so this can be only a partial explanation.

Sunk costs. Regular readers of FanGraphs are well acquainted with the concept of sunk costs; Dave Cameron has written about it repeatedly. Basically, a team should look at its total payroll as a cost of doing business, then allocate playing time in a manner that optimizes its chances of winning ballgames. That’s theoretical, of course. What actually happens is that teams are often reluctant to put high-salaried players into supporting roles. Take the 2016 Yankees, for example. They have a projected 2016 payroll of $230 million. They’ll spend about three quarters of that amount on nine players, all but one older than 30. Ideally, they should be willing to put CC Sabathia ($25 million in 2016, his age-35 season) in the bullpen, or make a DH platoon out of Mark Teixeira ($22.5 million, 36) and Alex Rodriguez ($20 million in each of 2016 and 2017, 40), or release Carlos Beltran ($15 million, 39) if any of them start particularly slowly. That’s what they might do with a 25-year-old making the major-league minimum. But the payroll obligation makes that move harder, even though that obligation’s a sunk cost — the team has to pay it regardless of how much the player plays. Here are the eight players aged 35 or older who, over the past two years, have suffered a wRC+ decline of 25 or more while retaining at least a semi-regular role, along with their contract status beyond the decline season:

All but Beltre and Byrd were below-average hitters in the second year, arguably not deserving of the plate appearances they received. But all but Suzuki, Utley, and Byrd were due at least eight figures after the year of their large decline. By contrast, a decade earlier, in 2004-2005, there were eleven semi-regular batters who, aged 35 or older, who had a wRC+ decline of 25 or more. Of them, only three — Luis Gonzalez and Jim Edmonds in 2005 and Bret Boone in 2004 — were in the midst of unexpired long-term multi-million-dollar contracts. Small sample size warnings and all, but there was a lot more future money committed to declining old batters in 2014-15 than 2004-05. Maybe those players wouldn’t be getting the plate appearances to meet the 350 threshold if it weren’t for the money that’s owed them.

Fastballs. One of the notable changes in baseball in recent years has been that pitchers throw harder. From 2007 to 2015, per PITCHf/x, the average fastball velocity increased from 91.1 mph to 92.4 mph. The increase was 1.3 mph, to 91.9 mph, for starters and 1.5 mph, to 93.2 mph, for relievers. Older batters can take advantage of their knowledge of the strike zone and pitch sequencing, but maybe they just can’t catch up to some pitches.

Granted, I’m guessing here. I’m leaning towards PEDs, both strength-enhancing and amphetamines, faster fastballs, and a tendency to put high-paid players in the lineup regardless of performance as the key drivers. But I’m not sure. This is an interesting trend, and sufficiently well-established that I don’t think we can write it off as a recent fluke. Something’s going on with players in the second half of their fourth decade that hasn’t happened in a long time.