The Modern Eras Committee Just Elected Bartolo Colon to the Hall of Fame

Jack Morris pitched 18 seasons while Bartolo Colon has now pitched 20. They both have a career winning percentage of .577. Morris has 2478 career strikeouts while Colon has 2454. Morris had 254 wins while Colon has 240 in an era where they are harder to obtain. Colon won a Cy Young while Morris’s highest finish was third. Morris has an ERA+ of 105, compared to Colon’s career ERA+ of 107. In fact, if you only looked at their first 15 years, Colon’s ERA+ of 114 outperforms Morris’s ERA+ of 109 even more!

Perhaps you strongly believe that, despite their statistical similarities, Jack Morris was significantly better than Bartolo Colon. Still, the fact that an argument could be made that Colon is as good a pitcher as Morris shows just how big a mistake the Modern Eras Committee made in electing Jack Morris to the Baseball Hall of Fame this weekend.

The list of players who were once viewed as “obviously not Hall of Famers” does not stop at Colon, either. In his article on ESPN.com, David Schoenfield said that it would be foolish to treat Morris as a benchmark for Hall of Fame induction. This argument is in defense of the Hall of Fame’s level of “rigor” — many think that without maintaining a certain level, the Hall of Fame may lose it’s significance. However, I believe there is another characteristic that the Hall of Fame must preserve even more so than rigor in order to maintain its credibility — and that is justice. If the Hall of Fame exposes itself as being discretionary in its election of members, it will quickly lose its relevance.

By electing Jack Morris to the Baseball Hall of Fame, voters both lowered the level of rigor previously required for election and have left the Hall of Fame in a current state of injustice until the following eligible players are also elected: Curt Schilling, Mike Mussina, Andy Pettitte, Dave Stieb, Rick Reuschel, Orel Hershiser, David Cone, Sam McDowell, Luis Tiant, Kevin Brown, Vida Blue, Bret Saberhagen, and Kevin Appier,

And the following cases are re-opened for election: Dwight Gooden, David Wells, Jim Kaat, Tommy John, Wilbur Wood, Ron Guidry, Jimmy Key, Frank Tanana, Dennis Martinez, Mark Langston, Chuck Finley, Mark Buehrle, Frank Viola and Jose Rijo.

Every single one of these 30 pitchers had a higher career ERA+ than Jack Morris and have either a higher career value, a higher peak value, or both.

Looks like Colon may be able to hang up his cleats a little more confidently this off-season now that Morris is in the Hall.


Another Weird Charlie Blackmon-ism

Charlie Blackmon is an atypical human being.

For one thing, he is a professional baseball player, meaning he is in the extreme upper echelon of athletic ability. But he is atypical even in his personal life, and his recent success has only highlighted his eccentric personality. He still drives a 2004 Jeep Grand Cherokee that he got in high school. He once had to be rescued on the side of the highway by DJ LeMahieu when he ran out of gas. He buys his clothes from Amazon. And of course, he is easily recognized by his impressive beard-and-mullet combo (the latter of which is pronounced “mu-lay” according to Blackmon).

Based on all his quirks, it should be no surprise just how unique his major-league career has been. He didn’t see regular playing time until his age-28 season, an age when some guys are already entering free agency. Despite this late start, he has steadily grown into an MVP candidate. In 2014, his first full season, Blackmon posted 2.0 fWAR, the exact threshold for a starting caliber player. In the three subsequent seasons, he posted an fWAR of 2.3, 4.1, and 6.5. I thought it seemed rather strange to have back-to-back seasons with ~2 WAR improvement, so I went to the leaderboards.

I searched for all batters with a minimum of 400 PAs in each of the past three seasons, producing a sample of 111 players. Then, I calculated the difference between each player’s 2015 WAR and 2016 WAR, and did the same for 2016 to 2017. This gave me two year-to-year improvements for each player, and I threw both values onto the scatter plot below, with Blackmon highlighted in purple.

2015 2017 WAR Improvements

Players generally don’t see improvements like this in back-to-back seasons; Blackmon is about as far to the top-right as you can get in this plot. Of course, value can come from many different places, and a player might make large defensive improvements one year and large offensive improvements the next. While Blackmon did see some improvement in his defensive metrics this season, the bulk of his improvements have come while batting. To get the following plot, I followed the same method as above, this time for wRC+.

2015 2017 w RC Improvements

Again, we see Blackmon floating towards the top right. Baseball is a game of adjustments, and if a batter enjoys a period of success, pitchers will generally approach him differently to gain an advantage. This is why players generally go through cycles, following the push and pull of the game. The past few years, Blackmon seems to be part of a small group of players who have been immune to this tug-of-war effect. He has stayed one step ahead of the pitchers, not only maintaining his gains but improving upon them as time goes on.

How has he found these improvements? Between 2015 and 2016, his walk rate and strikeout rate remained fairly constant, so he must have been getting much better results on balls in play. Sure enough, his batting average increased by 37 points and his ISO increased by 65 points, giving him 49 extra points of wOBA overall. At the time, Jeff Sullivan looked under the hood and found that Blackmon’s GB% was trending downward, and he had been attacking the low strike more so than ever before. Presumably, he realized that his swing path was conducive to driving low pitches into the air, and that balls in the air are more valuable, so he made the adjustment and enjoyed a power spike.

That all makes sense, but it begs the question: how did he improve even more in 2017? If he doubled down on the fly-ball revolution, he risked becoming Ryan Schimpf or Trevor Story.

Much to my surprise, the opposite happened – his GB% actually returned back to his career average. He increased his rate of ground balls, but he still managed raised his ISO by another 42 points. Before you cry BABIP or Coors Field, I’ll briefly note that in both years, his wOBA and xwOBA increased by approximately the same amount, so something real is going on here. In this case, I think he was finding more success on batted balls based on the pitches he didn’t put in play. Stay with me here.

In 2017, Blackmon’s strikeout rate rose by about 2.5%. This is what people in the industry call “not good,” but hold on, his walk rate also rose by…about 2.5%. This isn’t a player who suddenly developed a swing-and-miss problem to sell out for power, this is a player who is intentionally going deeper into counts. When a batter is more selective about the pitches he goes after, he is putting fewer balls in play in early counts, which leads to an increase in both walks and strikeouts simultaneously.

Let’s look at it a different way: Z-swing% measures the percentage of pitches inside the zone that a player swings at, and O-swing% measures the percentage of pitches outside the zone that a player swings at. Generally speaking, you want to swing at strikes and take balls, so you want your Z-swing% to be higher than your O-swing%; the larger the difference, the better your plate discipline.

In 2016, the difference between Blackmon’s Z-swing% and O-swing% ranked in the 9th percentile – he’s always been a bit of a free-swinging leadoff hitter. But in 2017, that difference increased by 4.7%, pushing him into the 26th percentile. While he’s still more aggressive than average, he has become decidedly less so, being more selective about the pitches he attacks and remaining comfortable in deep counts. By swinging at the right pitches, he’s able to avoid the easy outs that result from poor contact on pitches outside of the zone.

We have every reason to believe that Charlie Blackmon just had a career year, and he will never sniff an MVP race again in his career. But then again, we had every reason to believe the same thing last year. When it comes to Charlie, I have some advice: if you expect him to do something, he’s probably getting ready to do the exact opposite. It’s about time we stop trying to figure him out.


Improving WPS

“All happy families are alike; each unhappy family is unhappy in its own way.”  — L. Tolstoy

You can say something similar about baseball games. All boring games are alike; but exciting games are interesting in their own ways. Every boring game has one team building up a big early lead, which is never threatened. But there are many ways to have an exciting game: the pitcher’s duel, the slugfest, the late-inning comeback, extra innings, all in various combinations. And in between them are the bulk of games that are simply ordinary.

All of which makes ranking exciting games a tricky process, at least compared ranking to how boring they are. How does one compare Game 7 of the 1991 WS (1-0 in 10 innings) to Game 4 of the 1993 WS (15-14 in 9 innings) on the same scale? They’re great in different ways.

Back in 2005 I created a system to do just that, a rating system based simply on the runs scored in line score. I may have been the Christopher Columbus of that new world. And ranking the games allows you to rate post-season series-es.

The line-score system did work in the sense that it could tell the difference between a great game and a good one, and between a good one and an ordinary one. But while the line score gives you the basic outline of the game, it was blind to the details of what happens DURING each inning. Zero runs scored in the top of the 1st rates exactly the same; whether there were three pop-ups, or if three singles were followed by a triple play.

Eventually I realized that Baseball-Reference.com (ALL HAIL BBREF) has the play-by-play data for all playoff games, which includes a probability of victory after each play (anything that changes the outs, baserunners or score). Plotted, you can easily see if a game was good; It looks like and earthquake. If it was bad, it looks like the EKG of a corpse. Using those probabilities, we can create a much more accurate game rating. I fiddled with many rating schemes over the last 10 years before settling on one that seems both conceptually simple and that yields reasonable results.

Of course, by then I had been beaten to the basic concept by Dave Studeman (WPA) and Shane Tourtellotte (WPS). Twelve years is too long for laurels resting.

WPA = Sum(change in probability between plays)

Modified WPS = Sum(change in probability between plays) + top three plays + Final play

What I have developed is similar to their work, but I think it has some small advantages. Generally, my ratings will be quite close to Shane’s (R-squared > 99.5%). He correctly realized that simply summing the probabilities doesn’t quite work, which is why he modified it. An example…

There are seven post-season games with a WPA of exactly 4.52. Among them are:

1995 NLCS Game 2

Reds beat the Braves 6-2 in ten innings.

95 Plays, 13 plays changed the odds by at least 10%

top Play a Mark Portugal bases-loaded wild pitch +18%

70 plays with the odds in the 30% to 70% range

compared to

1960 WS Game 7

Pirates 10 Yankees 9 in nine innings

77 plays, 15 plays changed the odds by at least 10%

Of those 4 changed the odds by at least 20%

Of those 3 changed the odds by at least 30%

Of those 1 changed the odds by more than 50%

25 plays with the odds in the 30% to 70% range

 

There is simply no way those games are equal. The 1960 game has five different plays better than any play in the 1995 game. The 1995 game makes up the ground by (1) having 18 more plays (2) having fewer plays where nothing happened because the game was usually within one run.

1960 is still better because a +40% play isn’t twice as exciting as two +20% plays. Bill Mazeroski’s game-ending homer rates as +37%. Bobby Richardson’s game-starting line-out rates at +2%. Making a walk-off homer the equal of about 3 ½ innings with zero hits. NOPE. WRONG.

Shane accounted for this with his modified method. By counting the top three plays twice and Mazeroski’s walk-off homer three times, the ratings are now

1960: 6.49

1995: 5.19

And science prevails.

Of course, there is nothing magical about TOP THREE plays or LAST play. You could try using the top five plays and last five plays (believe me, I did).  But I do think that using Top-3 + Last can sometimes lead you astray. I will now present exhibits A and B to demonstrate where it can swing and miss.

Exhibit A: 1988 WS Game 1

Exhibit B: 1985 NLCS Game 6

I expect you to know them. The two biggest home runs in terms of changing the odds in post-season history courtesy of Mr. Clark and Mr. Gibson.

1985: WPA 4.48 in 83 plays and 9 innings

1988: WPA 3.94 in 82 plays and 9 innings

The 1985 game had more action with the same number of plays, which you can easily see in the line scores

StL          0              0              1              0              0              0              3              0              3              (7)

LA           1              1              0              0              2              0              0              1              0              (5)

 

Compared to

 

Oak        0              4              0              0              0              0              0              0              0              (4)

LA           2              0              0              0              0              1              0              0              2              (5)

 

The ‘85 game has a game tie in the 7th, broken tie in the 8th and lead change in the 9th

The ‘88 game has a lead change in the 2nd and a lead change in the 9th

Modified WPS says

1985: 4.48 + 1.34 + 0.01 = 5.83 (Tied for 94th best game)

1988: 3.94 + 1.43 + 0.87 = 6.28 (Tied for 58th best game)

I don’t think you can argue that the 1988 game is much better than the 1985 game; I don’t think it’s a better game at all. And it’s the last-play bonus that is to blame. Had the 1985 game been played in St. Louis then Clark’s homer would have been a walk-off and the game would have rated 6.56, well ahead of the 1988 game.

If you think about it, a last-play bonus is biased towards games won by the home team. If the home team loses, the last play will rarely amount to anything.

Only 23 times has it been at least 20%. When the home team wins, it is at least 20% 122 times.

Only 11 times has it been at least 30%. When the home team wins, it is at least 20% 96 times.

I also know this because I tried last play, last five plays, and last ten plays in trying to construct a rating system. I also tried top five plays, top ten plays, all plays over 10%, WPA – .03 per play (yielding the bizarre result of games with negative excitement).

Eventually I tried a simple power transformation on EVERY play. First, I tried summing the squares of the probabilities changes, like any good statistician would.

When I did that, the 1985 game Rated 10th and the 1988 game rated 5th. Which is the wrong order, and both games are just rated too high. Then I tried other powers…the Goldilocks approach, looking for the one that was just right.

 

Power             Rank               Rank

2.0          1985       10th         1988       5th best game

1.9          1985       12th         1988       8th Best game

1.8          1985       15th         1988       20th Best game

1.7          1985       23rd        1988       25th Best game

1.6          1985       32nd        1988       36th Best game

1.5          1985       38th        1988       51st Best game

1.4          1985       53rd        1988       76th Best game

1.3          1985       61st         1988       104th Best game

1.2          1985       79th        1988       133rd Best game

1.1          1985       100th      1988       158th Best game

1.0          1985       116th      1988      185th Best game

Everything above 1.7 was eliminated since it rated 1988 better than 1985

 

Here’s some shorthand I’m going to use:

Game 6 of the 1985 NLCS: STL 7, LA 5 in 9 innings — WPA 4.48 (9-4-2-1)

Game 1 of the 1988 WS: LA 5, SF 4 in 9 innings — WPA 3.98 (5-2-2-1)

The 1985 game had 9 plays rated>= 0.1, 4 plays rated>=0.2, 2 plays rated>=0.3 and 1 play rated >=0.5

The 1988 game had 5 plays rated>= 0.1, 2 plays rated>=0.2, 2 plays rated>=0.3 and 1 play rated >=0.5

For a sense of scale, the average game is WPA 2.67 (4.89-0.88-0.33-0.03)

(You can check the examples listed below on BBRef to get more detail on each game)

 

Checking 1.7, both exhibits rated higher than

Game 2 of the 2017 WS: HOU 7, LA 6 in 11 innings — WPA 5.30 (10-5-3-0)

Game 1 of the 2015 WS: KC 5, NYM 4 in 14 innings — WPA 6.36 (16-3-1-0)

1.7 weights the big plays too much

 

Checking 1.6, both test games rated higher than

Game 6 of the 1986 WS: NYM 6, BOS 5 in 10 innings — WPA 5.14 (16-3-3-0)

Game 6 of the 1986 NLCS: NYM 7, HOU 6 in 16 innings — WPA 5.80 (11-3-2-0)

1.6 weights the big plays too much

 

Checking 1.5,

the 1985 game rated higher than

Game 6 of the 1986 WS: NYM 6, BOS 5 in 10 innings — WPA 5.14 (16-3-3-0)

The 1988 game rated higher than

Game 4 of the 2001 WS: NYY 4, ARI 3 in 10 innings — WPA 4.58 (10-3-2-0)

1.5 weights the big plays too much, but it’s getting hard to find clear mistakes

 

Checking 1.4,

the 1985 game rated higher than

Game 3 of the 1976 NLCS: CIN 7, PHI 6 in 9 innings — WPA 4.72 (14-3-2-0)

Lead changes in the 7th, 8th and 9th innings.

The 1988 game rated higher than

Game 4 of the 1986 ALCS: CAL 4, BOS 3 in 11 innings — WPA 4.64 (7-4-2-0)

1.4 weights the big plays too much, but I’m now splitting hairs

 

Checking 1.3, I like this one. Let me check 1.2

 

Checking 1.2,

the 1985 game rated lower than

Game 2 of the 1996 ALDS: NYY 5, TEX 4 in 12 innings — WPA 5.02 (8-2-0-0)

Game 2 of the 1990 WS: CIN 5, OAK 4 in 10 innings — WPA 4.50 (10-2-0-0)

1.2 weights the big plays too little. Famous games are losing to games without any highlights.

 

So, I think 1.3 is the sweet spot.

My rating score is = Sum((change in probability between plays)^1.3) *2

The *2 at the end is purely cosmetic. It allows the very best game to score close to ten.

 

With base WPA, Gibson’s homer (.87) is worth about 25x a normal play (.035). With WPS it’s worth bout 75x a normal play. Raising all the plays to the 1.3 power means that Gibson’s homer is now worth about 65x a typical play.

With base WPA, Clark’s homer (.74) is worth about 21x a normal play (.035). With WPS it’s worth bout 42x a normal play. Raising all the plays to the 1.3 power means that Clark’s homer is now worth about 53x a typical play.

With a little algebra,

WPA:  Gibson = 1.18 * Clark

WPS: Gibson = 1.76 * Clark

Power 1.3: Gibson = 1.23 * Clark

A nice property of the transformation is that when the change in odds doubles, the play is worth ~ two and half times a much (2.46x)

 

EXCITEMENT IS NOT LINEAR

 

A 10% play is now worth 2.46 times as much as 5% play

A 20% play is now worth 2.46 times as much as 10% play

A 50% play is now worth 2.46 times as much as a 25% play

The system has a single parameter applied to ALL plays, so a game isn’t screwed if it has four great plays or the best play comes in the 8th inning. Ranking games this way, here are the five games better than, and worse than, my two test cases.

 

Series Road Team home team IP  (WPA^1.3)
*2
 WPA Top
Play
 # Plays  P>= .1  P>= .2  P>=.3  P>=.5
2014
ALCS G1
Royals 8 Orioles 6 10 5   5.14 35.0%         96        13         3         2        –
1935
WS G3
Tigers 6 Cubs 5 11 4.97   5.02 36.0%         96        15         5         1        –
1976
NLCS G3
Phillies 6 Reds 7 9 4.95   4.72 46.0%         82        14         3         2        –
2015
ALDS2 G2
Rangers 6 Blue Jays 4 14 4.93   5.46 37.0%       115         7         2         1        –
1997
ALCS G4
Orioles 7 Indians 8 9 4.92   4.92 38.0%         88        16         4         1        –
1985
NLCS G6
Cardinals 7 Dodgers 5 9 4.92   4.48 74.0%         83         9         4         2         1
1975
NLCS G3
Reds 5 Pirates 3 10 4.88   4.52 55.0%         81        14         3         3         1
1933
WS G4
Giants 2 Senators 1 11 4.87   4.94 55.0%         92         9         3         1         1
2011
ALCS G2
Tigers 3 Rangers 7 11 4.86   5.10 34.0%         92        13         3         1        –
2012
ALDS2 G2
Athletics 4 Tigers 5 9 4.86   4.86 41.0%         85        11         4         1        –
1999
NLCS G6
Mets 9 Braves 10 11 4.85   5.12 26.0%       108        14         3        –        –

 

 

Series Road home team IP  (WPA^1.3)
*2
 WPA Top
Play
 # Plays  P>= .1  P>= .2  P>=.3  P>=.5
1952
WS G5
Dodgers 6 Yankees 5 10 4.51   4.70 44.0%         92        10         4         1        –
1923
WS G1
Giants 5 Yankees 4 9 4.51   4.54 40.0%         78        12         2         2        –
1984
NLCS G4
Cubs 5 Padres 7 9 4.51   4.54 37.0%         83        10         4         2        –
1992
WS G2
Blue Jays 5 Braves 4 9 4.5   4.40 65.0%         85        11         1         1         1
1998
ALCS G2
Indians 4 Yankees 1 12 4.48   4.78 33.0%         96        11         3         1        –
1988
WS G1
Athletics 4 Dodgers 5 9 4.47   3.98 87.0%         82         5         2         2         1
2000
NLCS G2
Mets 6 Cardinals 5 9 4.46   4.66 32.0%         91        13         3         2        –
2016
NLDS2 G5
Dodgers 4 Nationals 3 9 4.46   4.66 21.0%         84        14         1        –        –
1977
WS G1
Dodgers 3 Yankees 4 12 4.45   4.80 30.0%         97        11         2         1        –
1954
WS G1
Indians 2 Giants 5 10 4.43   4.74 29.0%         89        11         1        –        –
1958
WS G1
Yankees 3 Braves 4 10 4.43   4.56 40.0%         88        10         3         2        –

 

 

I hope you’ll look at these and see that while they have different shapes, they all contain a similar ‘volume’ of excitement.

Another way to evaluate the method is to look at games with the same WPA. Going back to where I began in this article, here are the seven games with a base WPA of 4.52 (No promises that BBRef has not revised the scores since I captured the data…). They are each tied for the 108th highest WPA. But after using the 1.3 power factoring, you get this:

  Game Outcome RANK (WPA^1.3)*2  WPA   #
Plays
 Top 5
Plays
 # plays
30-70%
 P>=
.1
 P>=
.2
 P>=
.3
 P>=
.5
1960 WS G7 Pit 10 NYY 9 in 9 52              5.10   4.52     77   1.74         25    15      4      3      1
1975 NLCS G3 Cin 5 Pit 3 in 10 63              4.88   4.52     81   1.60         49    14      3      3      1
1911 WS G3 A’s 3 Giants 2 in 11 110              4.41   4.52     86   1.10         58    15      3      1     –
1998 NLCS G1 SD 3 Atl 2 in 10 117              4.36   4.52     84   1.10         59    11      2      1     –
2011 NLDS2 G5 Ari 3 Mil 2 in 10 119              4.35   4.52     85   1.05         71    13      2      1     –
1926 WS G5 NYY 3 StL 2 in 10 130              4.20   4.52     86   0.84         66    16      1     –     –
1995 NLCS G2 Atl 6 Cin 2 in 10 139              4.12   4.52     95   0.75         70    13     –     –     –

 

1960 gets the love it deserves, moving up 56 spots to the 52nd best game. That despite of having the fewest plays in the 30%-70% victory range. Games with more plays do worse since that means they have smaller impact plays on average. Think of the Top 5 plays as the highlight reel for the game. 1995 NLCS Game 2 has no play >0.2 and therefore drops 31 spots in the rankings.

Adjusted WPS? Weighted WPS? Power WPS? I really do need to give it a proper name.

 

A Final example, from among the greatest Playoff games ever.

2000 NLDS G3: Mets 3, Giants 2 in 13 innings — ModWPS Rank = 11, PowerWPS Rank = 22

1986 ALCS G5: Red Sox 7, Angels 6 in 11 innings — ModWPS Rank = 22, PowerWPS Rank = 12

1980 NLCS G5: Phillies 8, Astros 7 in 10 innings — ModWPS Rank = 25, PowerWPS Rank = 14

 

The 2000 game had the higher WPS, partly because it had more plays. ModWPS likes it more due to the additional action and walk-off homer, which the better top-three plays in 80/86 could not overcome.

 

year        WPS      Plays      Last Play    Top-3     ModWPS

2000       6.34        109         0.42          0.98                 7.74

1986       5.86       97           0.05             1.42                 7.33

1980       6.06        93           0.04            1.11                 7.21

 

So why do I think 1986/1980 are better?

Because, the deeper you go beyond the top three, the better the other two are revealed to be.

 

2000                                       1986                                       1980

1.28                                        1.94                                        1.61                        Sum of Top-5 Plays

42-31-25-16-14                  73-35-34-32-20                  40-38-35-26-24     Top-5 Plays

1.88                                        2.77                                        2.43                        Sum of Top-10 Plays

16-3-2-0                               14-5-4-1                               17-6-3-0               10%-20%-30%-50% plays

 

Or simply check the line scores.

2000

0 0 0 2 0 0 0 0 0 0 0 0 0 (2) Giants

0 0 0 0 0 1 0 1 0 0 0 0 1 (3) Mets

1986

0 2 0 0 0 0 0 0 4 0 1         (7) Red Sox

0 0 1 0 0 2 2 0 1 0 0         (6) Angels

1980

0 2 0 0 0 0 0 5 0 1             (8) Phillies

1 0 0 0 0 1 3 2 0 0             (7) Astros

 

The 2000 game IS a fabulous game. But the 1986 and 1980 games are more epic, with all the late-inning heroics. The 2000 game has exactly the required three big plays and the walk-off. It checks all the boxes.

I do kinda feel bad writing this. It sounds like I’m just picking on modified WPS here. LOOK AT WHAT ELSE IT GOT WRONG…

But as I said before, Power WPS is barely better. And to show that it’s better at all, I need to show those rare cases where it makes a better call. And it was an excellent benchmark, comparing differences between it and my sixty-eleven schemes helped me identify the flaws in sixty-ten of them.

Of course, even this is not the perfect system. Any play-by-play method will still fail to capture the in-play action. A bases-empty foul pop-out rates exactly the same as a bases-empty thrown-out-at-home-trying-to-stretch-a-triple. But it is the best we can do for now.

Whereas I used to guess my line score method captured maybe 70% of the excitement of a game, PBP ratings must be capturing upwards of 90%. Which means greater confidence in game rankings and playoff series ratings.

Anyway, if anyone has any thoughts, feedback, or questions I’d love to hear them. If no one can shoot the idea full of holes, or even one hole; then comes ranking and lists of games and series.


Relationship Between OBP and Runs Scored in College Baseball

There is a segment of the population of the United States which meets the following criteria:  between the age of 18-21, devout FanGraphs reader, and was mesmerized by the movie “Moneyball.”  I have read the book and watched the movie a number of times, as well as dedicating time to understanding the guiding principles in the book and how they relate to professional baseball.  The relationship between on-base percentage and scoring runs in Major League Baseball is well established, but has anyone ever taken the time to examine the relationship at the collegiate level?

Collegiate baseball is volatile — roster makeups change dramatically each year, no player is around more than five years, not to mention there are hundreds of teams competing against one another. In terms of groundbreaking sabermetric principles, this study is not intended to turn over any new stones, but rather present information which may have been overlooked up to this point, which is the relationship between on-base percentage and runs scored in collegiate baseball.

To conduct this study, I compiled a list of Southeastern Conference team statistics from the 2014-2017 seasons (Runs Scored, On Base Percentage, Runs Against, and Opponents’ On Base Percentage).  I then performed linear regression on the distribution by implementing a line of best fit.  Some teams’ seasons were excluded due to inability to access that season’s data, and I felt like removing the 2014 Auburn season on the grounds that it was an outlier affecting the output (235 runs, 0.360 OBP).  Below is the resulting math:  the R2, and the resulting predictive equation:

Runs Scored = ( 3,537. x OBP ) – 933.6791

R² = 0.722849

I am by no means a seasoned statistician, but in my interpretation of the R2 value, the relationship between Runs Scored and OBP in this is moderately strong, with a team’s OBP accounting for roughly 72.3% of the variation in Runs Scored in a season.  Simply, OBP is statistically significant in determining the offensive potency of a team.

At the professional level, the R2 is found to be around 0.90.  The competitive edge the Oakland A’s used in “Moneyball” was using this correlation to purchase the services of “undervalued” players.  But what about in college?  Colleges certainly cannot purchase their players, but the above information can be useful to college programs.

For example, the average Runs Scored per season of the sample I used was roughly 347.8.  If an SEC team wanted to set the goal of being “above average” offensively, they would be able to determine, roughly, what their target OBP should be by using the resulting predictive equation from the Linear Fit:

Does this mean if an SEC program produces an OBP of .362 they would score 348 runs precisely? Obviously not. Could they end up scoring exactly 348 runs? Yes, but variation exists, and statistics is the study of variation.  Here are a few seasons in which teams posted an OBP at or around 0.362, and the resulting run totals:

The average of those six seasons’ run totals was 347.5, which is pretty darn close to 348, and even closer to the average of 347.8 runs derived from the sample.

Another use for this information is lineup construction and tactical strategy in-game.  The people in charge of baseball programs do not need instruction on how to construct their roster and manage their team, but who would disagree with a strategy of maximizing your team’s ability to get on base?

The purpose of this study was to examine the relationship between On Base Percentage and Runs Scored in college baseball, and how the relationship compares to its professional counterpart.  To conclude, the relationship between OBP and runs exists at the collegiate level, and carries considerable weight and value if teams are willing to get creative in utilizing its ability.

 

Disclaimer: I am a beginner-level statistician, and if you have any suggestions or critiques of this article, please feel free to share them with me.

Theodore Hooper is a Student Assistant, Player Video/Scouting, for the University of Tennessee baseball program.  He can be reached at thooper3@vols.utk.edu or on LinkedIn at https://www.linkedin.com/in/theodore-hooper/


Shohei Ohtani, Stephen Strasburg, and Literal “Can’t Miss”

Because I believed Jeff Sullivan that there was a 2% chance Ohtani would sign Friday, I wrote this article, that now reads a bit weird, but I’m not going to change it because I have to get back to the job that I do for money instead of for fun. Any complaints about said weirdness should be addressed to Jeff Sullivan.

The phrase “sure thing” is thrown around a lot in sports to describe things that are not actually sure things. Atlanta was going to win the Super Bowl. Up 25 points, it was a sure thing until it wasn’t. Shohei Ohtani is not a sure thing to be a superstar in MLB, or even an All-Star. He’s not even a sure thing to be an average big-league player. History is littered with players that were star players one day and also-rans the next. But what Ohtani is a sure thing to do is pay off his price tag. Barring a tragic accident or something else outside of the realm of baseball, the team that signs Ohtani will surely earn more than $20.5 million extra during the 2018 baseball season that they would not have earned without him.

To understand why this is a certainty, you can look at Stephen Strasburg in 2010. The Nationals were the worst team in baseball in 2009, finishing with 59 wins for the second consecutive season. They were not much better in 2010, adding 10 more wins to that total, but still finishing 5th in the NL East. Their overall attendance for 2010 was bad, which is to be expected. 1,828,066 were said to have paid money to see the Nationals play that year, or about 22,500 per game. That was only about 10,000 higher, on the season, than they managed in 2009. Strasburg’s first game was electric (I was there). It was the kind of atmosphere that made no sense at all for a team that was mired in long-term terribleness. The game, played on a Tuesday, sold out with attendance only rivaled that season by opening day, up to that point.

But that only tells a small part of the story. The two Tuesday games the Nationals had played at home prior to this game averaged about 16,000 fans. It’s pretty safe to say that this game, against the Pirates, would have been in a similar range without Strasburg. But the Strasburg effect was much wider reaching. First, once it became known that Strasburg was pitching, the Nationals created a ticket package to sell many of the unsold tickets. The package included the Strasburg game, plus three additional games. For those that came to the stadium without a ticket, their choices were between that and the suddenly busy scalpers.

The weekend before Strasburg’s debut, the Nationals drew almost 91,000 fans against the Reds. That was about 6,000 more than they drew during their previous weekend series against the Orioles. But the Orioles are not really a fair comparison because they are Beltway rivals of the Nationals. If you go back the weekend series before that, when the Nationals were playing the Marlins, just over 63,000 came to the ballpark. Now, some of this might have been random or based on other factors, but fans were anticipating Strasburg’s debut and buying tickets in anticipation of it — fans including the humble author of this post. Some even speculated that the Nationals intentionally misled fans in order to juice the gate.

Strasburg’s second start was in Cleveland, who also seemed to have benefited from a large Strasburg-related spike in attendance for that game. His third game was a Friday and was sold out. Again, even weekend games prior to Strasburg were sparsely attended. The Saturday game after his start was nearly sold out, likely due to fans incorrectly believing it was going to be the day Strasburg pitched. His 3rd home game was the first that did not sell out. But there is more to the story (again, I was there). This game was played during the week, during the day, and it was literally at or above 100 degrees that day. There were still almost 32,000 fans there. After pitching on the road, Strasburg again sold out the stadium for his 6th start and had a large, but not sold-out game for his 7th. His next home start, another Tuesday game, again sold out. I’ve probably gone on long enough, if not too long. If you didn’t know how much Strasburg boosted attendance, you do now.

And here is the question: As much hype as there was in D.C. during the summer of 2010, do we really think that Ohtani’s hype won’t meet or perhaps even far exceed it? I certainly do. Additionally, the team that signs Ohtani will have a much easier time plotting and planning exactly how they can wring the most dollars possible out of their fans. In 2010, the Nationals created a multigame package on the fly, but now teams are exploring more and more ways of earning extra dollars through dynamic pricing. If a team like Seattle were to get Ohtani and announces that he’ll be pitching the second game of the season, the added revenue they would be able to earn would be astounding. Last season, the Mariners’ attendance dropped from 44,856 on opening day to 18,527 the next day. With Ohtani, it’s safe to say they’d sell out their second game.

If you’ve stayed with me this long, you might be wondering “yeah, I knew all of this already, so what?” And yes, most FanGraphs readers probably already believed that Ohtani is going to really juice attendance wherever he ends up. But then, why did three teams not even bother to try to acquire him? There simply is no justification. If you are the Marlins, you can do both things. You can say you are cutting payroll by $50 million, but at the same time if Ohtani for some reason picks the Marlins (he wouldn’t), you find the $20 million to pay him. You will get it back and more. Take out a loan. Heck, take out a payday loan with onerous and unfair terms and you will still end up ahead. It simply makes no sense. He is, from a financial standpoint, literally a sure thing.


Identifying Impact Hitters: Proof of Concept

Earlier this season I set out to build a tool similar in nature to my dSCORE tool, except this one was meant to identify swing-change hitters. Along the course of its construction and early-alpha testing, it morphed into something different, and maybe something more useful. What I ended up with was a tool called cHit (“change Hit”, named for swing changers but really I was just too lazy to bother coming up with a more apt acronym for what the tool actually does). cHit, in its current beta form, aims to identify hitters that tend to profile for “impact production” — simply defined as hit balls hard, and hit them in the air. Other research has identified those as ideal for XBH, so I really didn’t need to reinvent the wheel. Although I’d really like to pull in Statcast data offerings in a more refined form of this tool, simple batted ball data offered here on FanGraphs does the trick nicely.

The inner workings of this tool takes six different data points (BB%, GB%, FB%, Hard%, Soft%, Spd), compares each individual player’s stat against a league midpoint for that stat, then buffs it using a multiplier that serves to normalize each stat based on its importance to ISO. I chose ISO as it’s a pretty clean catch-all for power output.

Now here’s the trick of this tool: it’s not going to identify “good” hitters from “bad” hitters. Quality sticks like Jean Segura, Dee Gordon, Cesar Hernandez, and others show up at the bottom of the results because their game doesn’t base itself on the long ball. They do just fine for themselves hitting softer liners or ground balls and using their legs for production. Frankly, chances are if a player at the bottom of the list has a high Speed component, they’ve got a decent chance of success despite a low cHit. Nuance needs to be accounted for by the user.

Here’s how I use it to identify swing-changers (and/or regression candidates): I pulled in data for previous years, back to 2014. I compared 2017 data to 2016 data (I’ll add in comparisons for previous years in later iterations) and simply checked to see who were cHit risers or fallers. The results were telling — players we have on record as swing changers show up with significant positive gains, and players that endured some significant regression fell.

There’s an unintended, possible third use for this tool: identifying injured hitters. Gregory Polanco, Freddie Freeman, and Matt Holliday all suffered/played through injury this year, and they all fell precipitously in the rankings. I’ll need a larger sample size to see whether injuries and a fall in cHit are related or if that’s just noise.

Data!

cHit 2017
Name Team Age AB cHit Score BB% GB% FB% Hard% Soft% Spd ISO
Joey Gallo Rangers 23 449 27.56 14.10% 27.90% 54.20% 46.40% 14.70% 5.5 0.327
J.D. Martinez – – – 29 432 23.52 10.80% 38.30% 43.20% 49.00% 14.00% 4.7 0.387
Matt Carpenter Cardinals 31 497 22.46 17.50% 26.90% 50.80% 42.20% 12.10% 3.1 0.209
Aaron Judge Yankees 25 542 21.56 18.70% 34.90% 43.20% 45.30% 11.20% 4.8 0.343
Lucas Duda – – – 31 423 19.69 12.20% 30.30% 48.60% 42.10% 14.50% 0.5 0.279
Cody Bellinger Dodgers 21 480 19.26 11.70% 35.30% 47.10% 43.00% 14.00% 5.5 0.315
Miguel Sano Twins 24 424 17.73 11.20% 38.90% 40.50% 44.80% 13.50% 2.9 0.243
Jay Bruce – – – 30 555 16.50 9.20% 32.50% 46.70% 40.30% 11.70% 2.6 0.254
Trevor Story Rockies 24 503 16.39 8.80% 33.70% 47.90% 40.30% 14.40% 4.7 0.219
Justin Turner Dodgers 32 457 16.16 10.90% 31.40% 47.80% 38.90% 9.80% 3.3 0.208
Khris Davis Athletics 29 566 15.64 11.20% 38.40% 42.30% 42.10% 13.50% 3.4 0.281
Brandon Belt Giants 29 382 15.38 14.60% 29.70% 46.90% 38.40% 14.00% 4.2 0.228
Nick Castellanos Tigers 25 614 14.94 6.20% 37.30% 38.20% 43.40% 11.50% 4.6 0.218
Eric Thames Brewers 30 469 14.52 13.60% 38.40% 41.30% 41.50% 16.00% 4.6 0.271
Justin Upton – – – 29 557 14.43 11.70% 36.80% 43.70% 41.00% 19.80% 4 0.268
Justin Smoak Blue Jays 30 560 14.38 11.50% 34.30% 44.50% 39.40% 13.10% 1.7 0.259
Wil Myers Padres 26 567 14.32 10.80% 37.50% 42.90% 41.40% 19.50% 5.3 0.220
Paul Goldschmidt Diamondbacks 29 558 14.31 14.10% 46.30% 34.90% 44.30% 11.30% 5.6 0.265
Chris Davis Orioles 31 456 14.28 11.60% 36.70% 39.80% 41.50% 12.80% 2.7 0.208
Kyle Seager Mariners 29 578 13.57 8.90% 31.30% 51.60% 35.70% 13.10% 2.2 0.201
Nelson Cruz Mariners 36 556 13.35 10.90% 40.40% 41.80% 40.70% 14.70% 1.7 0.261
Mike Zunino Mariners 26 387 13.31 9.00% 32.00% 45.60% 38.60% 17.50% 1.9 0.258
Mike Trout Angels 25 402 13.16 18.50% 36.70% 44.90% 38.30% 19.00% 6.2 0.323
Corey Seager Dodgers 23 539 13.08 10.90% 42.10% 33.10% 44.00% 12.90% 2.7 0.184
Logan Morrison Rays 29 512 12.74 13.50% 33.30% 46.20% 37.40% 17.50% 2.4 0.270
Randal Grichuk Cardinals 25 412 12.61 5.90% 35.90% 42.70% 40.20% 18.20% 5.2 0.235
Salvador Perez Royals 27 471 12.50 3.40% 33.30% 47.00% 38.10% 16.50% 2.4 0.227
Michael Conforto Mets 24 373 12.42 13.00% 37.80% 37.80% 41.60% 20.20% 3.6 0.276
Matt Davidson White Sox 26 414 12.19 4.30% 36.20% 46.50% 38.20% 15.80% 1.8 0.232
Mike Napoli Rangers 35 425 12.15 10.10% 33.20% 52.10% 35.50% 21.90% 2.7 0.235
Miguel Cabrera Tigers 34 469 12.03 10.20% 39.80% 32.90% 42.50% 9.90% 1.1 0.149
Brandon Moss Royals 33 362 11.83 9.20% 33.10% 44.50% 37.30% 13.60% 2.3 0.221
Curtis Granderson – – – 36 449 11.69 13.50% 32.60% 48.80% 35.30% 17.60% 4.8 0.241
Ian Kinsler Tigers 35 551 11.64 9.00% 32.90% 46.50% 37.00% 18.70% 5.6 0.176
Edwin Encarnacion Indians 34 554 11.01 15.50% 37.10% 41.80% 37.60% 15.50% 2.7 0.245
Manny Machado Orioles 24 630 10.79 7.20% 42.10% 42.10% 39.50% 18.50% 3.3 0.213
Freddie Freeman Braves 27 440 10.72 12.60% 34.90% 40.60% 37.50% 12.40% 4.3 0.280
Nolan Arenado Rockies 26 606 10.60 9.10% 34.00% 44.90% 36.70% 17.60% 4.1 0.277
Anthony Rendon Nationals 27 508 10.41 13.90% 34.00% 47.20% 34.30% 13.00% 3.5 0.232
Yonder Alonso – – – 30 451 10.34 13.10% 33.90% 43.20% 36.00% 13.20% 2.4 0.235
Kyle Schwarber Cubs 24 422 10.24 12.10% 38.30% 46.50% 36.40% 21.30% 2.8 0.256
Carlos Gomez Rangers 31 368 10.19 7.30% 39.10% 40.30% 39.00% 16.50% 5 0.207
Luis Valbuena Angels 31 347 9.81 12.00% 38.40% 47.30% 35.80% 22.00% 1.3 0.233
Dexter Fowler Cardinals 31 420 9.61 12.80% 39.40% 38.20% 38.10% 12.70% 5.9 0.224
Jed Lowrie Athletics 33 567 9.40 11.30% 29.40% 43.50% 34.50% 12.10% 2.7 0.171
Giancarlo Stanton Marlins 27 597 8.96 12.30% 44.60% 39.40% 38.90% 20.80% 2.3 0.350
Jose Abreu White Sox 30 621 8.95 5.20% 45.30% 36.40% 40.50% 15.80% 4.4 0.248
Josh Donaldson Blue Jays 31 415 8.92 15.30% 41.00% 42.30% 36.30% 17.30% 1.6 0.289
Joey Votto Reds 33 559 8.87 19.00% 39.00% 38.00% 36.30% 10.40% 2.8 0.258
Victor Martinez Tigers 38 392 8.75 8.30% 42.10% 34.20% 39.90% 12.40% 0.9 0.117
Charlie Blackmon Rockies 31 644 8.63 9.00% 40.70% 37.00% 39.00% 17.10% 6.4 0.270
Mitch Moreland Red Sox 31 508 8.43 9.90% 43.40% 36.20% 38.90% 13.50% 1.7 0.197
Scott Schebler Reds 26 473 8.29 7.30% 45.60% 38.20% 39.40% 19.30% 3.9 0.252
Paul DeJong Cardinals 23 417 8.19 4.70% 33.70% 42.90% 36.40% 21.40% 2.5 0.247
Ryan Zimmerman Nationals 32 524 8.18 7.60% 46.40% 33.70% 40.50% 14.10% 2.2 0.269
Mookie Betts Red Sox 24 628 7.76 10.80% 40.40% 42.80% 35.70% 18.20% 5.5 0.194
Rougned Odor Rangers 23 607 7.61 4.90% 41.50% 42.20% 36.80% 18.50% 5.6 0.193
Francisco Lindor Indians 23 651 7.42 8.30% 39.20% 42.40% 35.20% 14.30% 5.1 0.232
Brad Miller Rays 27 338 7.39 15.50% 47.40% 36.10% 38.40% 18.10% 4.6 0.136
Daniel Murphy Nationals 32 534 6.97 8.80% 33.50% 38.90% 35.70% 16.70% 3.8 0.221
Travis Shaw Brewers 27 538 6.87 9.90% 42.50% 37.60% 37.10% 15.80% 4.5 0.240
Jake Lamb Diamondbacks 26 536 6.86 13.70% 41.10% 38.30% 35.70% 12.90% 4.4 0.239
Todd Frazier – – – 31 474 6.75 14.40% 34.20% 47.50% 32.20% 23.20% 3.1 0.215
Yasmani Grandal Dodgers 28 438 6.63 8.30% 43.50% 40.00% 36.50% 17.60% 1.1 0.212
Brian Dozier Twins 30 617 6.60 11.10% 38.40% 42.60% 34.10% 15.90% 5.2 0.227
Adam Duvall Reds 28 587 6.55 6.00% 33.20% 48.60% 31.80% 17.50% 3.9 0.232
Hunter Renfroe Padres 25 445 6.52 5.60% 37.90% 45.40% 34.60% 23.50% 3.2 0.236
Justin Bour Marlins 29 377 6.40 11.00% 43.40% 33.60% 38.80% 19.60% 1.6 0.247
Carlos Correa Astros 22 422 6.33 11.00% 47.90% 31.70% 39.50% 15.00% 3.2 0.235
Marcell Ozuna Marlins 26 613 6.09 9.40% 47.10% 33.50% 39.10% 18.30% 2.3 0.237
Domingo Santana Brewers 24 525 5.85 12.00% 44.90% 27.70% 39.70% 11.70% 4 0.227
Kris Bryant Cubs 25 549 5.83 14.30% 37.70% 42.40% 32.80% 14.80% 4.4 0.242
Gary Sanchez Yankees 24 471 5.47 7.60% 42.30% 36.60% 36.90% 18.60% 2.6 0.253
Asdrubal Cabrera Mets 31 479 5.46 9.30% 43.50% 36.20% 36.80% 17.20% 2.5 0.154
Austin Hedges Padres 24 387 5.37 5.50% 36.60% 45.70% 33.10% 22.30% 2.7 0.183
Logan Forsythe Dodgers 30 361 5.33 15.70% 44.00% 33.10% 36.60% 13.20% 2.8 0.102
Yadier Molina Cardinals 34 501 5.25 5.20% 42.20% 37.40% 36.40% 16.50% 3.9 0.166
Bryce Harper Nationals 24 420 5.07 13.80% 40.40% 37.60% 34.30% 13.30% 3.7 0.276
Neil Walker – – – 31 385 5.01 12.30% 36.20% 41.70% 32.80% 17.70% 2.8 0.174
Aaron Altherr Phillies 26 372 5.01 7.80% 43.10% 37.50% 36.40% 20.10% 5.5 0.245
Andrew McCutchen Pirates 30 570 4.90 11.20% 40.70% 37.40% 35.20% 17.50% 4.3 0.207
Eduardo Escobar Twins 28 457 4.86 6.60% 33.70% 45.30% 31.40% 16.00% 5.1 0.195
Anthony Rizzo Cubs 27 572 4.79 13.20% 40.70% 39.20% 34.40% 19.80% 4.4 0.234
Ryan Braun Brewers 33 380 4.73 8.90% 49.20% 31.90% 39.00% 19.20% 5.3 0.218
Kendrys Morales Blue Jays 34 557 4.56 7.10% 48.40% 33.20% 37.90% 15.20% 1.1 0.196
Jose Ramirez Indians 24 585 4.54 8.10% 38.90% 39.70% 34.00% 16.70% 6 0.265
Mike Moustakas Royals 28 555 4.51 5.70% 34.80% 45.70% 31.90% 21.20% 1.1 0.249
Andrew Benintendi Red Sox 22 573 4.50 10.60% 40.10% 38.40% 34.30% 16.60% 4.5 0.154
Jose Bautista Blue Jays 36 587 4.47 12.20% 37.70% 45.80% 31.40% 21.70% 3.4 0.164
Jason Castro Twins 30 356 4.36 11.10% 41.90% 33.50% 36.00% 14.00% 1.5 0.146
Albert Pujols Angels 37 593 4.12 5.80% 43.50% 38.10% 35.10% 15.90% 2.1 0.145
Hanley Ramirez Red Sox 33 496 4.04 9.20% 41.80% 37.10% 35.30% 20.00% 1.5 0.188
Tommy Joseph Phillies 25 495 3.99 6.20% 41.70% 39.00% 35.00% 20.90% 2.2 0.192
Tim Beckham – – – 27 533 3.99 6.30% 48.80% 29.50% 39.10% 15.50% 4.4 0.176
Jonathan Schoop Orioles 25 622 3.90 5.20% 41.90% 37.20% 36.10% 23.00% 2.2 0.211
George Springer Astros 27 548 3.58 10.20% 48.30% 33.80% 36.70% 17.90% 3.1 0.239
Carlos Beltran Astros 40 467 3.54 6.50% 43.10% 40.40% 33.70% 17.50% 1.8 0.152
Alex Bregman Astros 23 556 3.52 8.80% 38.40% 39.90% 33.00% 18.00% 5.9 0.191
Carlos Santana Indians 31 571 3.49 13.20% 40.80% 39.30% 33.00% 18.40% 4 0.196
Eugenio Suarez Reds 25 534 3.33 13.30% 38.90% 37.10% 33.80% 20.70% 3.1 0.200
Scooter Gennett Reds 27 461 3.29 6.00% 41.30% 37.60% 34.40% 17.20% 4.3 0.236
Mark Reynolds Rockies 33 520 3.26 11.60% 42.10% 36.30% 34.50% 19.00% 2.7 0.219
Josh Reddick Astros 30 477 3.23 8.00% 33.60% 42.30% 31.10% 17.20% 4.8 0.170
Mitch Haniger Mariners 26 369 2.97 7.60% 44.00% 36.70% 34.70% 17.70% 4.3 0.209
Ian Happ Cubs 22 364 2.92 9.40% 40.20% 39.70% 32.80% 18.70% 5.7 0.261
Josh Harrison Pirates 29 486 2.90 5.20% 36.50% 40.80% 32.40% 18.70% 4.9 0.160
Keon Broxton Brewers 27 414 2.78 8.60% 45.10% 34.60% 35.30% 17.00% 7.4 0.200
Matt Joyce Athletics 32 469 2.69 12.10% 37.80% 42.80% 30.30% 16.30% 3.2 0.230
Derek Dietrich Marlins 27 406 2.65 7.80% 36.50% 40.70% 32.10% 20.50% 3.9 0.175
Ryon Healy Athletics 25 576 2.56 3.80% 42.80% 38.20% 33.90% 16.50% 1.4 0.181
Evan Longoria Rays 31 613 2.50 6.80% 43.40% 36.80% 34.30% 18.00% 3.8 0.163
Zack Cozart Reds 31 438 2.49 12.20% 38.20% 42.30% 30.80% 19.50% 5.3 0.251
Robinson Cano Mariners 34 592 2.48 7.60% 50.00% 30.60% 36.90% 12.80% 2 0.172
Max Kepler Twins 24 511 2.39 8.30% 42.80% 39.50% 32.90% 18.70% 4.2 0.182
Steven Souza Jr. Rays 28 523 2.22 13.60% 44.60% 34.30% 34.10% 16.50% 4.8 0.220
Michael Taylor Nationals 26 399 2.17 6.70% 42.90% 36.70% 34.00% 18.10% 5.9 0.216
Yulieski Gurriel Astros 33 529 2.12 3.90% 46.20% 35.20% 35.10% 15.90% 2.8 0.187
Corey Dickerson Rays 28 588 1.24 5.60% 41.80% 35.80% 33.60% 18.70% 4 0.207
Whit Merrifield Royals 28 587 1.01 4.60% 37.70% 40.50% 30.60% 15.40% 6.7 0.172
Chris Taylor Dodgers 26 514 0.88 8.80% 41.50% 35.80% 32.40% 15.80% 6.4 0.208
A.J. Pollock Diamondbacks 29 425 0.81 7.50% 44.60% 32.10% 35.00% 19.80% 7.5 0.205
Marwin Gonzalez Astros 28 455 0.71 9.50% 43.90% 36.20% 32.70% 18.60% 3.2 0.226
Yangervis Solarte Padres 29 466 0.62 7.20% 41.60% 42.10% 31.10% 25.20% 2.4 0.161
Shin-Soo Choo Rangers 34 544 0.57 12.10% 48.80% 26.20% 36.10% 12.20% 4.7 0.162
Buster Posey Giants 30 494 0.50 10.70% 43.60% 33.00% 33.00% 14.10% 2.8 0.142
Jedd Gyorko Cardinals 28 426 0.48 9.80% 40.50% 39.30% 30.80% 19.20% 3.8 0.200
Yasiel Puig Dodgers 26 499 0.30 11.20% 48.30% 35.60% 32.90% 18.30% 4.4 0.224
Eddie Rosario Twins 25 542 0.12 5.90% 42.40% 37.40% 31.70% 16.70% 3.9 0.218
J.T. Realmuto Marlins 26 532 -0.01 6.20% 47.80% 34.30% 33.30% 14.90% 5 0.173
Jorge Bonifacio Royals 24 384 -0.20 8.30% 39.30% 34.80% 32.20% 20.20% 2.9 0.177
Gerardo Parra Rockies 30 392 -0.27 4.70% 46.80% 30.30% 34.70% 14.40% 3 0.143
Willson Contreras Cubs 25 377 -0.34 10.50% 53.30% 29.30% 35.50% 17.00% 2.4 0.223
Kole Calhoun Angels 29 569 -0.37 10.90% 43.90% 35.00% 31.80% 17.00% 3.7 0.148
Robbie Grossman Twins 27 382 -0.43 14.70% 40.70% 34.40% 30.90% 16.00% 3.5 0.134
Matt Holliday Yankees 37 373 -0.46 10.80% 47.70% 37.50% 31.80% 21.20% 2.1 0.201
Mark Trumbo Orioles 31 559 -0.47 7.00% 43.30% 40.60% 30.40% 20.90% 2.5 0.163
Stephen Piscotty Cardinals 26 341 -0.80 13.00% 49.20% 33.20% 32.70% 17.90% 2.7 0.132
Tommy Pham Cardinals 29 444 -0.86 13.40% 51.70% 26.10% 35.50% 15.40% 6 0.214
Joe Mauer Twins 34 525 -0.92 11.10% 51.50% 23.60% 36.40% 12.80% 2.4 0.112
Jackie Bradley Jr. Red Sox 27 482 -0.94 8.90% 49.00% 32.60% 33.30% 17.50% 4.5 0.158
Brandon Crawford Giants 30 518 -0.98 7.40% 46.20% 34.40% 32.60% 19.30% 2.5 0.151
Nomar Mazara Rangers 22 554 -1.13 8.90% 46.50% 34.20% 32.60% 20.90% 2.6 0.170
Ben Zobrist Cubs 36 435 -1.35 10.90% 51.10% 33.30% 32.30% 14.90% 3.6 0.143
Javier Baez Cubs 24 469 -1.36 5.90% 48.60% 36.00% 32.40% 21.30% 5.3 0.207
Jorge Polanco Twins 23 488 -1.42 7.50% 37.90% 42.80% 27.70% 19.90% 4.9 0.154
Avisail Garcia White Sox 26 518 -1.70 5.90% 52.20% 27.50% 35.30% 15.70% 4.3 0.176
Matt Kemp Braves 32 438 -1.76 5.80% 48.50% 28.20% 34.70% 17.40% 1.7 0.187
Maikel Franco Phillies 24 575 -2.04 6.60% 45.40% 36.70% 30.90% 20.80% 1.5 0.179
Nick Markakis Braves 33 593 -2.17 10.10% 48.60% 29.20% 33.10% 15.60% 1.9 0.110
Tucker Barnhart Reds 26 370 -2.46 9.90% 46.00% 27.80% 33.20% 16.50% 3.4 0.132
Trey Mancini Orioles 25 543 -2.48 5.60% 51.00% 29.70% 34.10% 19.60% 3.2 0.195
Christian Yelich Marlins 25 602 -2.51 11.50% 55.40% 25.20% 35.20% 15.90% 5.2 0.156
Lorenzo Cain Royals 31 584 -2.79 8.40% 44.40% 32.90% 31.10% 18.70% 6.5 0.140
Josh Bell Pirates 24 549 -2.87 10.60% 51.10% 31.20% 32.60% 20.60% 3.5 0.211
Jose Reyes Mets 34 501 -3.00 8.90% 37.20% 43.10% 26.70% 26.10% 7.2 0.168
Carlos Gonzalez Rockies 31 470 -3.04 10.50% 48.60% 31.70% 31.90% 20.50% 3.2 0.162
Adam Jones Orioles 31 597 -3.27 4.30% 44.80% 34.30% 30.90% 20.10% 2.7 0.181
Byron Buxton Twins 23 462 -3.57 7.40% 38.70% 38.00% 27.60% 18.20% 8.2 0.160
Kevin Kiermaier Rays 27 380 -3.81 7.40% 49.60% 32.10% 31.80% 22.00% 5.9 0.174
Chase Headley Yankees 33 512 -3.90 10.20% 43.50% 31.70% 30.00% 17.10% 4.3 0.133
Xander Bogaerts Red Sox 24 571 -4.31 8.80% 48.90% 30.50% 31.40% 19.70% 6.7 0.130
Jordy Mercer Pirates 30 502 -4.33 9.10% 48.30% 30.90% 31.00% 19.00% 2.9 0.151
Brandon Drury Diamondbacks 24 445 -4.44 5.80% 48.80% 29.40% 31.70% 16.60% 2.4 0.180
Alex Gordon Royals 33 476 -4.69 8.30% 42.60% 33.00% 29.20% 19.40% 4.3 0.107
Ben Gamel Mariners 25 509 -4.84 6.50% 44.90% 33.30% 29.40% 18.70% 4.9 0.138
Hernan Perez Brewers 26 432 -4.85 4.40% 48.30% 33.50% 30.40% 21.20% 5.3 0.155
Matt Wieters Nationals 31 422 -4.94 8.20% 42.50% 36.40% 27.40% 18.10% 2 0.118
Brett Gardner Yankees 33 594 -5.07 10.60% 44.50% 33.20% 28.80% 20.00% 6 0.163
Odubel Herrera Phillies 25 526 -5.10 5.50% 44.10% 34.70% 29.40% 24.40% 4.3 0.171
Freddy Galvis Phillies 27 608 -5.11 6.80% 36.70% 39.20% 25.50% 18.10% 5.3 0.127
Elvis Andrus Rangers 28 643 -5.13 5.50% 48.50% 31.50% 30.50% 18.70% 5.7 0.174
Danny Valencia Mariners 32 450 -5.93 8.00% 47.90% 31.00% 29.80% 20.50% 3.3 0.156
Kevin Pillar Blue Jays 28 587 -6.25 5.20% 43.10% 36.40% 27.30% 22.50% 4.4 0.148
Dansby Swanson Braves 23 488 -6.35 10.70% 47.40% 29.40% 29.30% 18.00% 3.2 0.092
Jose Altuve Astros 27 590 -6.45 8.80% 47.00% 32.70% 28.20% 19.00% 6.4 0.202
Alcides Escobar Royals 30 599 -6.47 2.40% 40.80% 37.40% 26.80% 22.80% 4.3 0.107
Andrelton Simmons Angels 27 589 -6.62 7.30% 49.50% 31.50% 29.30% 20.60% 5 0.143
Didi Gregorius Yankees 27 534 -6.91 4.40% 36.20% 43.80% 23.10% 24.40% 2.7 0.191
Ryan Goins Blue Jays 29 418 -6.94 6.80% 50.30% 34.80% 27.70% 19.60% 2.7 0.120
Gregory Polanco Pirates 25 379 -7.00 6.60% 42.20% 37.50% 25.90% 22.80% 3.7 0.140
David Peralta Diamondbacks 29 525 -7.02 7.50% 55.10% 26.50% 31.80% 21.20% 4.6 0.150
Kolten Wong Cardinals 26 354 -7.11 10.00% 48.10% 31.80% 28.20% 20.80% 5.4 0.127
Orlando Arcia Brewers 22 506 -7.74 6.60% 51.60% 28.50% 30.20% 22.90% 4.1 0.130
Martin Maldonado Angels 30 429 -7.80 3.20% 48.50% 36.60% 26.70% 21.60% 2.3 0.147
Cory Spangenberg Padres 26 444 -7.85 7.00% 49.30% 27.80% 29.20% 16.90% 5 0.137
Joe Panik Giants 26 511 -7.96 8.00% 44.00% 34.10% 26.10% 20.10% 4.2 0.133
David Freese Pirates 34 426 -8.08 11.50% 57.00% 22.60% 31.90% 19.40% 1 0.108
Melky Cabrera – – – 32 620 -8.14 5.40% 48.90% 29.00% 28.90% 19.00% 2.3 0.137
Hunter Pence Giants 34 493 -8.28 7.40% 57.20% 29.40% 29.40% 18.50% 3.6 0.126
Manuel Margot Padres 22 487 -8.30 6.60% 40.50% 36.30% 25.40% 25.90% 6.1 0.146
Trea Turner Nationals 24 412 -8.61 6.70% 51.70% 33.50% 26.70% 18.00% 8.9 0.167
Jonathan Villar Brewers 26 403 -8.85 6.90% 57.40% 21.90% 33.20% 27.00% 5.4 0.132
Starlin Castro Yankees 27 443 -9.19 4.90% 51.80% 28.00% 29.20% 21.80% 3.5 0.153
Denard Span Giants 33 497 -9.30 7.40% 45.00% 33.60% 25.10% 18.60% 5.5 0.155
Jacoby Ellsbury Yankees 33 356 -9.73 10.00% 45.90% 31.00% 26.10% 22.70% 7.7 0.138
Delino DeShields Rangers 24 376 -9.93 10.00% 45.10% 34.80% 23.90% 20.10% 7.1 0.098
Adam Frazier Pirates 25 406 -9.98 7.90% 47.90% 26.80% 27.50% 17.90% 5.7 0.123
DJ LeMahieu Rockies 28 609 -10.42 8.70% 55.60% 19.70% 30.60% 15.40% 3.9 0.099
Yolmer Sanchez White Sox 25 484 -10.53 6.60% 44.50% 33.90% 24.00% 19.30% 5.3 0.147
Jason Heyward Cubs 27 432 -10.54 8.50% 47.40% 32.70% 25.50% 25.80% 4.3 0.130
Tim Anderson White Sox 24 587 -10.66 2.10% 52.70% 28.00% 28.30% 21.30% 6.2 0.145
Jean Segura Mariners 27 524 -10.79 6.00% 54.30% 26.40% 28.30% 19.70% 5.5 0.128
Cameron Maybin – – – 30 395 -10.88 11.30% 57.70% 27.90% 27.40% 20.10% 6.9 0.137
Dustin Pedroia Red Sox 33 406 -10.90 10.60% 48.80% 28.80% 25.90% 20.10% 2.2 0.099
Jose Iglesias Tigers 27 463 -10.91 4.30% 50.40% 26.40% 28.40% 23.40% 4.2 0.114
Eric Hosmer Royals 27 603 -11.30 9.80% 55.60% 22.20% 29.50% 21.80% 3.4 0.179
Eduardo Nunez – – – 30 467 -12.27 3.70% 53.40% 29.10% 26.70% 24.50% 4.8 0.148
Jon Jay Cubs 32 379 -12.53 8.50% 47.10% 23.90% 25.30% 11.50% 5.3 0.079
Brandon Phillips – – – 36 572 -12.97 3.50% 49.50% 28.30% 25.50% 21.70% 4.1 0.131
Guillermo Heredia Mariners 26 386 -15.19 6.30% 47.40% 34.90% 20.40% 23.80% 2.2 0.088
Ender Inciarte Braves 26 662 -15.36 6.80% 47.00% 29.10% 22.10% 20.90% 5.4 0.106
Jonathan Lucroy – – – 31 423 -16.18 9.60% 53.50% 27.90% 22.30% 20.50% 3.1 0.106
Jose Peraza Reds 23 487 -16.45 3.90% 47.10% 31.30% 21.40% 26.60% 5.8 0.066
Cesar Hernandez Phillies 27 511 -18.08 10.60% 52.80% 24.60% 22.10% 23.50% 6 0.127
Billy Hamilton Reds 26 582 -21.80 7.00% 45.80% 30.60% 16.00% 25.00% 9 0.088
Dee Gordon Marlins 29 653 -28.88 3.60% 57.60% 19.60% 16.10% 24.70% 8.5 0.067

Okay, so here’s the breakdown. I pulled all 2017 hitters with 400 at-bats or more so I could capture some significant hitters that didn’t have qualifying numbers of ABs due to injury. Ball-bludgeon extraordinaire Joey Gallo is a pretty solid name to have heading up this list, as he’s pretty much the human definition of what this tool is trying to identify. JD Martinez, Aaron Judge, Cody Bellinger, Miguel Sano, Trevor Story, and Justin Turner all in the top 10 is pretty much all the proof-of-concept I needed.

Interesting notes:

Brandon Belt at 12 — Someone needs to tell the Giants to trade him to literally any other team, stat.

Giancarlo Stanton at 46 — Surprisingly, the MVP fell off from his stats in 2016. His grounders and soft contact rose by 3 or more percentage points, and shaved off the equivalent from hard and fly balls. His output was fueled by adding almost 200 ABs to his season — he could actually get better if he can stay healthy and add those hard flies back in!

Francisco Lindor at 58 — The interesting part of this is even though Lindor is still a decent way down the list, he actually was the biggest gainer from last season to this, adding 9.52 points to his cHit. We knew he was gunning for flies from the outset of the season, and it looks like his mission was accomplished.

Mike Moustakas at 87 — Frankly, being bookended by Jose Ramirez and Andrew Benintendi should, in a vacuum, should be great company. But this is a prime example of how cHit requires users to not take the numbers at face value. Ramirez and Benintendi aren’t slug-first hitters like Moose. They’ve got significantly better Speed scores, plus aren’t as prone to soft contact. I’d be very wary of Moose regressing, as he seems to rely on sneaking some less-than ideal homers over fences. If he goes to San Francisco I could see his value crater (see Belt, Brandon).

Eric Hosmer at 206 — Nope, negative, pass, I’m trying to sign quality hitters here <— Suggested responses for GMs when approached this offseason by Scott Boras on behalf of Hosmer.

Final Notes:

  •  Batted-ball distribution data is noticeably absent. In one of my iterations I added in those stats, and found that they actually regressed the accuracy of the formula. It doesn’t matter where you hit the ball, as long as you hit it hard.
  • Medium% and LD% are noisy stats. They also regressed the formula.
  • I may look to replace BB% in future iterations. For now though, it does a decent job of capturing plate discipline and selectivity.
  • K% doesn’t seem to have much of an impact on cHit (see Gallo, Joey).
  • R-squared numbers over the last four years of data hold pretty steady between .65 and .75, which is really encouraging. Also, the bigger the pool of data per year (number of batters analyzed), the higher R-squared goes; which is ultimately the most encouraging result of this whole endeavor.

Input is greatly appreciated! I’m not a mathematician in any stretch of the imagination, so if there’s a better way of going about this I’d love to hear it. I’ll do a writeup about my swing-change findings at a later date.


Looking for Evidence of a Change to the Ball

We saw an unprecedented jump in home runs in the last few years. What made it so strange was that most of it happened after the 2015 All-Star break. There is an increased awareness of launch angle and bat path, and 2015 was the first year there was a public in-game feedback, but still you would expect such an adjustment to take longer, especially since in-season swing changes are really hard to do — maybe with a whole offseason to work on it, it might have been slightly more believable.

There have been multi-factor explanations like a great rookie class of power hitters in the second half of 2015, changed approach, and other stuff like a slightly smaller zone, but really you would not expect such a multi-factor cause to happen that quickly and distinctly between two season halves. That made most sabermetric writers, including most of the FanGraphs staff, believe in a single-factor cause, most likely the ball.

There is some evidence for a changed ball, and there is also anecdotal evidence of minor-league players called up claiming the MLB ball flies farther. However, MLB so far has rejected that, and supported that with the credible name of professor Alan Nathan, albeit without really publishing the data, which further increased the suspicion.

We also did see an increase in launch angle: In 2015 in the first half, the LA of the league was 9.6, and in the second half it was 10.3, which further slightly increased in the first half of 2016 (10.4) and 2017 (10.8). The biggest jump, however, occurred between the season halves of 2015. So were the players really able to increase their LA with a single focus cue without really having much time to work on swing mechanics by just aiming higher after getting the first-half feedback? Those are the most talented athletes in the world, but still that sounds incredible.

But of course just increased elevation doesn’t explain the surge. The number of balls hit between 20 and 35 degrees (usual HR range) increased from roughly 8200 in the first half of 2015 to roughly 8600 in the first half of 2016, but the number of HRs increased from 2521 to 3082. Since less than half of the FBs between 20 and 35 go out of the park (I don’t have the exact number but I estimate 30% from the numbers I have), the 600 more batted balls in that range don’t explain 500 more HRs. That means, apart from more FBs, those also got out more, and the league saw a jump in HR/FB rate (9.5% in 2014 and 12.8 in 2016).

To research that, I looked into some Statcast stats. All stats here are just first halves of the respective seasons, because the first half of 2015 was the last “normal” HR half. Also I want to lessen weather effects.

This table shows that balls between 20 and 35 degrees do indeed fly farther and also go faster off the bat.
Average distance (20-35 LA)

2015 326 89.9
2016 331 91.6
2017 332 91.3

So does this jump in HR/FB prove a juiced ball? Not necessarily. To explain this, we have to get into swing mechanics. The attack angle is the vector of the bat’s sweetspot just before contact. Generally you can hit higher LAs (launch angles) by just hitting the bottom of the ball, but while some backspin is good, too much of it will slow down the ball. Generally the more LA and attack angle match, the higher the exit velo. That means players that try to swing up more might shift their highest velos to higher LAs. So while players couldn’t really change their swings that fast, just the intent of higher LA might have unconsciously caused a higher attack angle and thus more “flush hit” fly balls.

Evidence for the ball not being a factor is that average league EV is actually down a tiny bit. However, if the attack-angle theory is true, you would also expect that the EV of balls between 0 and 10 degrees would lower a little bit, and that hasn’t really happened.

Avg EV EV (0-10 LA)
2015 87.1 93.3
2016 87.8 93.3
2017 86.9 93.1

Another theory came from Tom Tango. He assumed that harder swinging and increased attack angles lead to higher peak EVs but also more weak mis-hits.

We do indeed see a big increase of balls hit above 105 MPH, but on the other side (and there have to be weaker hits to explain that overall EV is not up) there is an effect of more weak-hit balls in 2017, but not so in 2016.

EV >85 Balls 105
2015 96.2 19210 2960
2016 96.9 19075 3917
2017 96.7 20436 3635

To see if there is an aerodynamic effect — one theory of the juiced ball is reduced air drag due to lower seams — I looked at the average distance of balls hit at 20-25 degree LA in different velocity buckets.

EV Range 95-100 100-105 105-110
2015 366 391 415
2016 362 387 408
2017 363 391 411

You can’t really see an effect here. Balls hit at the same EV (which is measured right after exit so that air drag hasn’t done its work yet) don’t fly farther in 2016 or 2017 than they did in the first half of 2015. That means there likely isn’t really an effect of aerodynamics, at least not a big one.

So the reason for increased HRs seems to be mostly that fly balls fly faster and farther for whatever reason. We don’t see an across-the-board increase of EV, however, but simple explanations like a shift of max EVs to other launch angles don’t seem to really work either, as LAs from 0-10 (and also lower than minus 5 for that matter) haven’t really changed in their EV.

It remains mysterious what did actually happen. We do know LAs have increased some, but that doesn’t explain the whole story. But I couldn’t find real evidence for a changed ball in Statcast either. Could a super fast on-the-fly adjustment of the league between season halves based on the Statcast date really be the driving factor here?

Intellectually I really want to believe the juiced-ball theory, as it is the most elegant explanation for such a quick turnaround, but maybe it isn’t that easy.


Yasiel Puig Was a Terrible, Terrible Baserunner

Yasiel Puig had an impressive rebound season in 2017. He responded to disappointing, injury-marred seasons in 2015-16 with a solid 2.9 WAR this year. Puig greatly improved his plate discipline, increasing his selectivity and his contact rates en route to an 11.2 BB% and 17.5 K%. He has been known for his free-swinging ways since entering the league, but he may have changed that reputation this past season. Puig was not the reckless hitter he had been in the past. However, he may have decided to channel that recklessness to the base paths.

Puig is a good athlete, but has never been much of a base-stealer. In his first two years, he converted a poor 22/37 of his steal attempts, and mostly quit trying to steal in 2015-16. Despite the failures of his base-stealing, he had actually been a slight positive on the bases in his career, accumulating 0.5 runs above average in 2013-16, per FanGraphs’ base-running metric. Puig reverted to his aggressive base-stealing in 2017, and his 15 stolen bases indicated success with the approach. His 71.4% conversion rate was not exceptional, but not horrible. But his base-running had no semblance of success.

Puig was the sixth-worst player on the bases in 2017, accumulating -7.6 runs. He was surrounded by names like Albert Pujols, Miguel Cabrera, and Edwin Encarnacion. Not exactly names you want to be grouped with when talking about base-running.

FanGraphs’ base-running metric encompasses three things: wSB, wGDP, and UBR. wSB measures the run value a player produced based off attempting steals. Puig produced a mediocre mark of 0.1, which lines up with his stolen-base numbers. wGDP measures the ability of a player to avoid double plays. Puig ranked 13th-worst in 2017 with -2.4 runs produced, but wGDP is more related to avoiding ground balls with men on base and beating out throws to first. UBR (Ultimate Base Running), measures the value of a player with respect to non-stealing base-running, like taking an extra base. Puig produced -5.3 runs per UBR, sixth-worst in the league.

Let’s focus on that UBR. Providing context, that figure is a whole lot worse than sixth-lowest in the league. Puig had a speed score of 4.4 in 2017, placing him 80th in the league, among players with at least 400 PA. The five players directly ahead of Puig in UBR had an average speed score of 2.3. That would rank 185th. Considering speed, Puig was likely the worst base-runner in the league. He did things like this, which you probably remember from the World Series:

Puig was probably the worst base-runner in 2017. But how bad was he on a historical level?

Of all individual seasons (min 400 PAs) since 2002, when UBR was introduced, Puig’s UBR ranks lower than the 3rd percentile out of 3393 seasons. Of those individual seasons with a speed score within one standard deviation of Puig’s, his UBR ranks lower than the 1st percentile.

Here is a plot of every one of those seasons, with each player’s stolen-base total versus their UBR. Puig in 2017 is highlighted in yellow.

Obviously, players with higher stolen-base totals are generally faster, and thus produce more value on the bases. As with anything, though, there are outliers. Puig is definitely an outlier. Only one player with as many stolen bases has produced an UBR lower than Puig: Juan Encarnacion in 2003. Here is another chart, with speed score plotted against UBR. Puig again is in yellow.

Puig is again an extreme outlier, even historically. Considering his athleticism, Puig had one of the worst base-running seasons of the last 15 years. This does not mean a ton. Puig has not always been a terrible base-runner, and he was still a quite effective player in 2017, woes on the basepaths aside. He can easily turn it around and produce a solid base-running season with the physical gifts he has. However, in 2017, Puig’s base-running was really, really terrible.


On Starling Marte and Steroids

Each baseball fan has a set of specific events throughout time they remember fondly. Some exist in said group because of their emotional impact on your fandom. Others remain on the peripheral of importance because of a random characteristic that still stands out.

Those peripheral events, for me, are often those I’ve seen on live television. I don’t think of these events often, nor do I keep a record of them, or have some strict guideline for what sticks in my head, but when a story in the present day sparks my memory, a picture often emerges. My teenage years watching baseball were done one of two ways: sitting on the ground in front of my laptop with MLB.tv fading in-and-out, or scouring local stations for a good matchup. These two primary settings allowed for many one-off memories to accumulate.

When I began to think about Pittsburgh Pirates’ outfielder Starling Marte — due to this offseason’s stagnation — I thought back to the first pitch he saw in his major-league career. Just over five years ago, 23-year-old Starling Marte took the first pitch Dallas Keuchel threw on July 26 out of Minute Maid Park. The rarity of that event — a prospect’s debut, leading off a game, first-pitch home run — forces me to remember that bomb whenever Marte steps into a batter’s box. Because I happened to see it live, that memory has stuck.

For the wider population of fans, what now supersedes that milestone is Marte’s run-in with performance-enhancing drugs.  Suspended for 80 games during the 2017 season, this mistake by Marte will couple itself with any other success he has.

Predicting how Marte would fare upon his return during this layoff in 2017 raised some interesting, PED-related questions. Would his power drop? Would his speed deteriorate? What about his overall durability?

Nestled within all those asks is what exactly the effect of PEDs on an athlete’s body is after stopping use. Much more intriguing is this question: does any use at all matter as much as stopping that use? In other words, do the effects of PED use in the first place help prolong success?

I mention this because Marte joins Dee Gordon as the more prominent speed-first users of prohibited substances in the recent years. The drugs Gordon and Marte took were different from my understanding — nandrolone versus a stacked dose with clostebol — but maybe some intrigue exists in the stats before and after use?

The overall comparison doesn’t show us much. Even in what I highlighted with darkened gridlines — slugging percentage and wRC+ — has more noise within it than signal. Two main questions exist, among many others, that don’t have answers.

  • What portion of the “before” PED use window contains tainted statistics?
  • What portion of the drop is due specifically to the lack of steroids in the body?

But perhaps our intentions with those questions are incorrect. Think back to the question I asked before showing this dataset: do the effects of PED use in the first place help prolong success?

What if the muscle memory and learning that takes place while a player is under the influence of the drug extends beyond the window where a player can run a positive test?

With some high-level Googling, I found one instance where this idea might be a reasonable rabbit hole to dig into (BBC News). Certainty around this topic, however, is impossible, given all the variables. Some selection bias brings us the average fan to Nelson Cruz and Bartolo Colon as examples of this idea. But assuming two players with demonstrable skills outside of steroid use represent a wider population is not an appropriate assumption. We’re left in limbo regarding how much one positive test early on can affect one’s long-term production.

***

Let’s leave the uncertainty around long-term effects of Marte’s steroid use alone for now and focus on what has happened in Marte’s career.

The attribute his value has been tied to for most of his career, like Dee Gordon, is speed. But for Marte, age-induced deterioration of that attribute may be underway as he heads into his 29-year-old season with the Pirates.

It wasn’t too long ago we were concerned about the viability of McCutchen’s long-term impact, yet speed has a much greater weight on the impact of Marte as a player than McCutchen. I remain perplexed as to how Marte intends to turn around this decline in sprint speed as he starts to fall away from elite towards the 27.0 feet-per-second average the standard MLB player possesses.

Marte can still produce with his bat, but after seeing this decrease in peak sprint speed, I wonder if he becomes less reliant on his wheels to buoy his BABIP and the resulting average he’ll post. The Pirates’ outfielder might need to adjust.

To counteract this potential speed regression, Marte might want to adjust back to his approach from 2015, where he popped 19 home runs.

What we do know from that year presides in his tendency to pull the ball above his career average, which resulted in the majority of his home runs landing somewhere near the corner in a park’s left-field seats. He was also more aggressive than he had ever been in his career in 2015, but since, Marte has reverted to a contact-based approach, raising his zone-contact rate by two percent and overall contact rate by three percent.

With all this said, the form and substance of Marte’s swing has been largely the same since the early days of his career. Each of the four videos embedded within the GIF below are base hits to left field for Marte. Instead of focusing on the moments just before contact — where most hitters look identical — focus on his pre-pitch rhythm and timing.

Marte has a unique pulse when it comes to the timing mechanism in his hands, as his bat moves towards the first-base line twice prior to his load. The speed at which he executes this varies slightly based on the pitch, but his front foot’s inward turn and hip rotation remain unaltered from this selection of swing in our four-year sample.

My worry is that pushing Marte towards the 2015 version of himself, with pull-happy tendencies and a little bit more aggression, may not lead to the power result we want. With his speed possibly deteriorating, the balls he rolls over on with his sights set on the bleachers will turn into hits less often. We might want Marte to trade some of his contact for power, but my inclination is that such a trade, at present, is not one-for-one and would result in a net-negative effect.

This contact approach of Marte’s may be the new normal, and I remain worried about what the ceiling of productivity can be if he doesn’t find a second wind in the speed department. Marte can still be an asset to the Pirates, and isn’t a financial burden, but it might be too late to expect 2015’s power-speed combo that had the chance to nudge Marte towards the elite bracket of outfielders in baseball.

Bill Brink of Pittsburgh Post Gazette reports that Marte is making up his lost at-bats in the Dominican Winter League for Leones del Escogido. The results, so far, in a small sample have not been great:

.197/.244/.316 in 76 at-bats, with a 21:3 strikeout-to-walk ratio.

Marte’s evolution as a hitter will become clearer as our post-PED sample size increases. The Pirates’ outfield, once considered the best in baseball with McCutchen, Marte, and Polanco, now finds itself in a pickle, especially if Cutch is traded, Marte’s speed continues to trend south, and Polanco can’t stay healthy.


An Exercise in Generating Similarity Scores

In the process of writing an article, one of the more frustrating things to do is generate comparisons to a given player. Whether I’m trying to figure out who most closely aligns with Rougned Odor or Miguel Sano, it’s a time-consuming and inexact process to find good comparisons. So I tried to simplify the process and make it more exact — using similarity scores.

An Introduction to Similarity Scores

The concept of a similarity score was first introduced by Bill James in his book The Politics of Glory (later republished as Whatever Happened to the Hall of Fame?) as a way of comparing players who were not in the Hall of Fame to those who were, to determine which non-HOFers deserved a spot in Cooperstown. For example, since Phil Rizzuto’s most similar players per James’ metric are not in the HOF, Rizzuto’s case for enshrinement is questionable.

James’ similarity scores work as such: given one player, to compare them to another player, start at 1000 and subtract one point for every difference of 20 games played between the two players. Then, subtract one point for every difference of 75 at-bats. Subtract a point for every difference of 10 runs scored…and so on.

James’ methodology is flawed and inexact, and he’s aware of it: “Similarity scores are a method of asking, imperfectly but at least objectively, whether two players are truly similar, or whether the distance between them is considerable” (WHHF, Chapter 7). But it doesn’t have to be perfect and exact. James is simply looking to find which players are most alike and compare their other numbers, not their similarity scores.

Yes, there are other similarity-score metrics that have built upon James’ methodology, ones that turn those similarities into projections: PECOTA, ZiPS, and KUBIAK come to mind. I’m not interested in making a clone of those because these metrics are obsessed with the accuracy of their score and spitting out a useful number. I’m more interested in the spirit of James’ metric: it doesn’t care for accuracy, only for finding similarities.

Approaching the Similarity Problem

There is a very distinct difference between what James wants to do and I what I want to do, however. James is interested in result-based metrics like hits, doubles, singles, etc. I’m more interested in finding player similarities based on peripherals, specifically a batted-ball profile. Thus, I need to develop some methodology for finding players with similar batted-ball profiles.

In determining a player’s batted-ball profile, I’m going to use three measures of batted-ball frequencies — launch angle, spay angle, and quality of contact. For launch angle, I will use GB%/LD%/FB%; for spray angle, I will use Pull%/Cent%/Oppo%; and for quality of contact, I will use Soft%, Med%, Hard%, and HR/FB (more on why I’m using HR/FB later).

In addition to the batted-ball profiles, I can get a complete picture of a player’s offensive profile by looking at their BB% and K%. To do this, I will create two separate similarity scores — one that measures similarity based solely upon batted balls, and another based upon batted balls and K% and BB%. All of our measures for these tendencies will come from FanGraphs.

Essentially, I want to find which player is closest to which overall in terms of ALL of the metrics that I’m using. The term “closest” is usually used to convey position, and it serves us well in describing what I want to do.

Gettin’ Geometrical

In order to find the most similar player, I’m going to treat every metric (GB%, LD%, FB%, Pull%, and so on) as an axis in a positioning system. Each player has a unique “position” along that axis based on their number in that corresponding metric. Then, I want to find the player nearest to a given player’s position within our coordinates system — that player will be the most similar to our given player.

I can visualize this up to the third dimension. Imagine that I want to find how similar Dee Gordon and Daniel Murphy are in terms of batted balls. I could first plot their LD% values and find the differences.

1-D visualization of Daniel Murphy's and Dee Gordon's batted ball profiles

So the distance between Murphy and Gordon, based on this, is 4.8%. Next, I could introduce the second axis into our geometry, GB%.

2-D visualization of Daniel Murphy's and Dee Gordon's batted ball profiles

The distance between the two players is given by the Pythagorean formula for distance — sqrt(ΔX^2 + ΔY^2), where X is LD% and Y is GB%. To take this visualization to a third dimension and incorporate FB%…

3-d visualization of Daniel Murphy's and Dee Gordon's batted ball profiles

… I would add another term to the distance calculation — sqrt(ΔX^2 + ΔY^2 + ΔZ^2). And so on, for each subsequent term. You’ll just have to use your imagination to plot the next 14 data points because Euclidian geometry can’t handle dimensions greater than three without some really weird projections, but essentially, once I find the distance between those two points in our 10 or 12-dimensional coordinate system, I have an idea how similar they are. Then, if I want to find the most similar batter to Daniel Murphy, I would find the distance between him and every other player in a given sample, and find the smallest distance between him and another player.

If you’ve taken a computer science course before, this problem might sound awfully familiar to you — it’s a nearest-neighbor search problem. The NNS problem is about finding the best way to determine the closest neighbor point to a given point in some space, given a set of points and their position in that space. The “naive” solution, or the brute-force solution, would be to find the distance between our player and every other player in our dataset, then sort the distances. However, there exists a more optimized solution to the NNS problem, called a k-d tree, which progressively splits our n-dimensional space into smaller and smaller subspaces and then finds the nearest neighbor. I’ll use the k-d tree approach to tackling this.

Why It’s Important to Normalize

I used raw data values above in an example calculation of the distance between two players. However, I would like to issue caution against using those raw values because of the scale that some of these numbers fall upon.

Consider that in 2017, the difference between the largest LD% and smallest LD% among qualified hitters was only 14.2%. For GB%, however, that figure was 30.7%! Clearly, there is a greater spread with GB% than there is with LD% — and a difference in GB% of 1% is much less significant than a difference in LD% of 1%. But in using the raw values, I weight that 1% difference the same, so LD% is not treated as being of equal importance to GB%.

To resolve this issue, I need to “normalize” the values. To normalize a series of values is to place differing sets of data all on the same scale. LD% and GB% will now have roughly the same range, but each will retain their distribution and the individual LD% and GB% scores, relative to each other, will remain unchanged.

Now, here’s the really big assumption that I’m going to make. After normalizing the values, I won’t scale any particular metric further. Why? Because personally, I don’t believe that in determining similarity, a player’s LD% is any more important than the other metrics I’m measuring. This is my personal assumption, and it may not be true — there’s not really a way to tell otherwise. If I believed LD% was really important, I might apply some scaling factor and weigh it differently than the rest of the values, but I won’t, simply out of personal preference.

Putting it All Together

I’ve identified what needs to happen, now it’s just a matter of making it happen.

So, go ahead, get to work. I expect this on my desk by Monday. Snap to it!

Oh, you’re still here.

If you want to compare answers, I went ahead and wrote up an R package containing the function that performs this search (as well as a few other dog tricks). I can do this in two ways, either using solely batted-ball data or using batted-ball data with K% and BB%. For the rest of this section, I’ll use the second method.

Taking FanGraphs batted-ball data and the name of the target player, the function returns a number of players with similar batted-ball profiles, as well as a score for how similar they are to that player.

For similarity scores, use the following rule of thumb:

0-1 -> The same player having similar seasons.

1-2 -> Players that are very much alike.

2-3 -> Players who are similar in profile.

3-4 -> Players sharing some qualities, but are distinct.

4+ -> Distinct players with distinct offensive profiles.

Note that because of normalization, similarity scores can vary based on the dataset used. Similarity scores shouldn’t be used as strict numbers — their only use should be to rank players based on how similar they are to each other.

To show the tool in action, let’s get someone at random, generate similarity scores for them, and provide their comparisons.

Here’s the offensive data for Elvis Andrus in 2017, his five neighbors in 12-dimensional space (all from 2017), and their similarity scores.

Elvis Andrus Most Similar Batters (2017)

The lower the similarity score, the better, and the guy with the lowest similarity score, J.T. Realmuto, is almost a dead ringer for Andrus in terms of batted-ball data. Mercer, Gurriel, Pujols, and Cabrera aren’t too far off as well.

After extensively testing it, the tool seems to work really well in finding batters with similar profiles — Yonder Alonso is very similar to Justin Smoak, Alex Bregman is similar to Andrew McCutchen, Evan Longoria is similar to Xander Bogaerts, etc.

Keep in mind, however, that not every batter has a good comparison waiting in the wings. Consider poor, lonely Aaron Judge, whose nearest neighbor is the second furthest away of any other player in baseball in 2017 — Chris Davis is closest to him with a similarity score of 3.773. Only DJ LeMahieu had a further nearest-neighbor (similarity score of 3.921!).

The HR/FB Dilemma

While I’m on the subject of Aaron Judge, let’s talk really quickly about HR/FB and why it’s included in the function.

When I first implemented my search function, I designed it to only include batted-ball data and not BB%, K%, and HR/FB. I ran it on a couple players to eye-test it and make sure that it made sense. But when I ran it on Aaron Judge, something stuck out like a sore thumb.

Aaron Judge Similarity Scores

Players 2-5 I could easily see as reasonable comparisons to Judge’s batted balls. But Nick Castellanos? Nick Castellanos? The perpetual sleeper pick?

But there he was, and his batted balls were eerily similar to Judge’s.

Aaron Judge Most Similar Batters (2017)

Judge hits a few more fly balls, Castellanos hits a few more liners, but aside from that, they’re practically twins!

Except that there’s not. Here’s that same chart with HR/FB thrown in.

Aaron Judge Most Similar Batters (2017)

There’s one big difference between Judge and Castellanos, aside from their plate discipline — exit velocity. Judge averages 100+ MPH EV on fly balls and line drives, the highest in the majors. Castellanos posted a meek 93.2 MPH AEV on fly balls and line drives, and that’s with a juiced radar gun in Comerica Park. Indeed, after incorporating HR/FB into the equation, Castellanos drops to the 14th-most similar player to Judge.

HR/FB is partially considered a stat that measures luck, and sure, Judge was getting lucky with some of his home runs, especially with Yankee Stadium’s homer-friendly dimensions. But luck can only carry you so far along the road to 50+ HR, and Judge was making great contact the whole season through, and his HR/FB is representative of that.

In that vein, I feel that it is necessary to include a stat that has a significant randomness component, which is very much in contrast with the rest of the metrics used in making this tool, but it is still a necessary inclusion nevertheless for the skill-based component of that stat.

Using this Tool

If you want to use this tool, you are more than welcome to do so! The code for this tool can be found on GitHub here, along with instructions on how to download it and use it in R. I’m going to mess around with it and keep developing it and hopefully do some cool things with it, so watch this space…

Although I’ve done some bug testing (thanks, Matt!), this code is still far from perfect. I’ve done, like, zero error-catching with it. If in using it, you encounter any issues, please @ me on twitter (@John_Edwards_) and let me know so I can fix them ASAP. Feel free to @ me with any suggestions, improvements, or features as well. Otherwise, use it responsibly!