Tuesday, June 30, 2015

Carlos Correa's WAR Is In Top 10 Already For Shortstops Aged 20 And Younger

From the Baseball Reference Play Index

For single seasons, From 1901 to 2015, Younger than 20, Played 50% of games at SS, (requiring WAR_bat>=1), sorted by greatest WAR Position Players. Player ages are computed as their age on June 30th. Correa has 1.1 offensive WAR and 0.7 defensive WAR (there must be a rounding issue).


Rk Player WAR Year Age PA
1Alex Rodriguez 9.4 1996 20 677
2 Travis Jackson 4 1924 20 633
3 Arky Vaughan 3.8 1932 20 555
4 Elvis Andrus 3.6 2009 20 541
5Edgar Renteria 3.2 1996 19 471
6 Alan Trammell 2.8 1978 20 504
7 Jose Reyes 2.3 2003 20 292
8 Travis Jackson 2.1 1923 19 350
9 Carlos Correa 1.7 2015 20 93
10 Starlin Castro 1.6 2010 20 506

Friday, June 26, 2015

SABR Names Dan Levitt As Winner Of This Year's Bob Davids Award

Click here to read about his achievements at the official SABR site. This seems well deserved. Dan excels at both history and sabermetrics.

Here is an excerpt from the announcement:
"Levitt is the author of four fine books: Paths to Glory: How Great Baseball Teams Got That Way (with Mark Armour; winner of the 2004 Sporting News-SABR Baseball Research Award); Ed Barrow: The Bulldog Who Built the Yankees' First Dynasty (a 2009 Seymour Medal finalist); The Battle That Forged Modern Baseball: The Federal League Challenge and Its Legacy (which won the 2013 Larry Ritter Award); and In Pursuit of Pennants: Baseball Operations from Deadball to Moneyball, published this spring with Mark Armour."
This was a good article by Dan: The Batter/Pitcher Match Up. Using the Bill James Log5 formula, Dan found that good hitters do about as well as we would expect against good pitchers. I think it is from the late 1990s.

Click here to go to Dan's Amazon page.

Here is part of what the award is about:
"The Bob Davids Award honors SABR members whose contributions to SABR and baseball reflect the ingenuity, integrity, and self-sacrifice of the founder and past president of SABR, L. Robert "Bob" Davids. Voluntary activities in the area of administration and research are among those contributions considered."

Tuesday, June 23, 2015

Where Does "The Sandberg Game" Rank In WPA?

On this date in 1984, Ryne Sandberg was 5 for 6 and hit game tying HRs in the bottom of the 9th and the bottom of the 10th (the latter with two outs and 1 on). The Cubs beat the Cardinals 12-11 in 11 innings.

But the big deal is that both of those HRs were off of Hall of Fame reliever Bruce Sutter. The game was televised nationally, I think on NBC. It was, if I recall correctly, a warm sunny day at Wrigley Field (I saw this game on TV).

Sandberg was not a big star yet, although he did win the MVP award that year. It was only his third full season and he had not yet played in an All-Star game. His career average entering the season was .261 with just 15 HRs in 327 games (all statistics used here are from Baseball Reference).

He did enter this game with a .321 average but just 7 HRs in 66 games (a pace of only 17 for a full season).

But how hard was it to homer against Sutter? He gave up 9 in 1984 while facing 477 batters. He faced Sandberg 6 times that year and gave up those 2 HRs. So when not facing Sandberg, Sutter gave up 7 HRs in 471 PAs. The year before he gave up 8 in 384 PAs. When not facing Sandberg over those two years, his HR per batter faced rate was 1.76%. This sounds about average.  The NL HR per batter faced rate over those two years was 1.81% (again, without IBBs removed).

He also issued 18 IBBs, so that HR rate might be higher than average. Sutter had a higher IBB rate than average, which was only about 1% (Sutter's was about 2%).

So yes, it was a big deal to hit those HRs. But at this point in his career, Sutter was about average in allowing HRs. In fact, not counting Sandberg, he allowed a higher HR% vs. righties (2.4%) than lefties (1.6%) over those two years.

Sandberg has said "It was one of those wild games, wind blowing out a little bit so it was going to be an offensive game from the get-go." See Wrigley 100 June 23: The Sandberg Game. There was, however, only one other HR in the game.

So what was Sandberg's WPA in that game? 1.063. It is the 35th highest since 1941, according to the Baseball Reference Play Index. Here are the top 40:


Rk Player Date WPA
1Art Shamsky8/12/1966 1.503
2 Dolph Camilli1941-09-01 (1) 1.398
3Jim Pagliaroni9/21/1965 1.287
4Brian Daubach8/21/2000 1.253
5Nelson Cruz9/7/2014 1.22
6Bobby Grich7/15/1979 1.211
7Mel Hall6/27/1984 1.206
8 Carlos May1973-09-03 (1) 1.204
9Willie Mays5/26/1962 1.204
10 Del Ennis4/26/1949 1.197
11Jim Hickman5/28/1970 1.181
12Hank Aaron9/10/1971 1.159
13 Dwight Evans6/23/1990 1.145
14Dante Bichette6/10/1998 1.141
15 Ken Boyer1961-09-14 (2) 1.14
16 Cody Ross6/7/2008 1.139
17 Will Clark6/22/1988 1.133
18 George Brett5/28/1979 1.126
19 Bobby Bonds4/29/1969 1.126
20 Ryan Langerhans9/7/2005 1.119
21 Barry Bonds8/12/1991 1.119
22Brandon Inge8/24/2003 1.113
23Hank Aaron8/18/1959 1.112
24Bill Melton1970-08-02 (2) 1.107
25Tim Harkness6/26/1963 1.107
26Wilin Rosario9/18/2014 1.087
27Mike Young5/28/1987 1.087
28 Fred McGriff8/24/1996 1.084
29Roy Sievers7/16/1958 1.083
30 Bob Allison4/16/1963 1.082
31Josh Hamilton7/9/2011 1.081
32 Ted Williams1946-07-14 (1) 1.072
33 Kevin Mitchell8/13/1987 1.064
34 Willie Montanez7/11/1976 1.064
35Ryne Sandberg6/23/1984 1.063
36 Jerry Buchek1967-09-22 (2) 1.063
37 Jose Bautista5/22/2013 1.062
38 Eric Soderholm5/13/1972 1.059
39Raul Mondesi4/5/1999 1.056
40 Steve Finley9/10/1996 1.056

Saturday, June 20, 2015

Top 10 Average Game Scores This Year, 8 Start Minimum

After Saturday's no-hitter (June 20), which was a 97, Scherzer is at 70.29. That would be 7th since 1938 for guys with 30+ starts.


Max Scherzer 68.2
Chris Sale 65.3
Dallas Keuchel 64.9
Zack Greinke 64.3
Sonny Gray 64.1
Chris Archer 63.6
Francisco Liriano 63.1
Gerrit Cole 63
Jacob deGrom 62.7
Johnny Cueto 61.8

Here is that top 10


Bob Gibson 76.1 1968
Luis Tiant 72.4 1968
Sandy Koufax 71.8 1965
Tom Seaver 70.9 1971
Pedro Martinez 70.8 1997
Dwight Gooden 70.4 1985
Vida Blue 69.9 1971
Sandy Koufax 69.8 1963
Steve Carlton 69.3 1972
Hal Newhouser 69.3 1946

Wednesday, June 17, 2015

SB% Lower So Far This Year

Here are the AL %'s in recent years

2003 70.04%
2004 68.62%
2005 70.49%
2006 71.46%
2007 73.19%
2008 72.96%
2009 73.98%
2010 73.59%
2011 72.10%
2012 74.98%
2013 73.57%
2014 73.91%
2015 69.37%

The AL has had at least 73% ever year since 2007 except 2008 which was very close at 72.96%. But for some reason this year is below 70%.

Now the NL

2003 68.87%
2004 71.71%
2005 70.67%
2006 71.29%
2007 75.56%
2008 73.04%
2009 70.74%
2010 71.20%
2011 72.34%
2012 73.13%
2013 71.92%
2014 71.62%
2015 70.10%

Maybe not as dramatic as the AL, but the NL had 75.56% in 2007 and this year could be the lowest since 2003.

Monday, June 15, 2015

For Pitchers That Change Teams, What Affects Their BABIP More? Their Own BABIP From The Year Before Or The BABIP For The Rest Of Their New Team?

This is related to the previous post Year to year correlation of BABIP for pitchers that changed teams and those that did not.

I ran another regression for the guys who changed teams. The dependent variable was their BABIP in year 2 and the independent variables were their BABIP in year 1 and the BABIP for the rest of their new team in year 2 (so that each guy's BABIP was removed from the team total). Here is the equation

BABIP2 = 0.232*OtherBABIP2 + 0.123*BABIP1 + 0.189

OtherBABIP2 means the BABIP for the rest of their new team in year 2

The r-squared is .022. Standard error is .02004. T-values are

OtherBABIP2 1.18
BAbip1 1.25

So neither is significant. But .232/.123 is 1.88. So the effect from the BABIP from the rest of the team (the guy's new team) is almost double from his previous year's BABIP.

The correlation between BABIP2 and OtherBABIP2 is .086, almost as high as .094, correlation between BABIP2 and BABIP1 for this same group of guys who changed teams (all the guys who had 150+ IP in the years 2003-14).

If we suppose that fielders have no influence on BABIP, then it means that the contribution from "the rest of the team" is due solely to the other pitchers. Then why is the effect  from other pitchers so much stronger than the pitcher himself (or at least what he did the year before)?

That would not make any sense. So it seems like there would have to be a big role for the fielders. If you think it is a "small" role, but not zero, the previous paragraph still applies.

The previous post showed that for guys who stayed on the same team, the impact of last year's BABIP on this year's BABIP was 2.5 times higher than it was for guys who changed teams.

And what is the big change when you switch teams? The fielders. The parks, too. The pitchers have a whole new set of fielders behind them and they pitched half their games (or around that) in a completely different park than the year before.

I don't know how much the makeup of a team's fielders change from year to year. My guess is that it is less than 50%. So guys who stay on the same team don't see much change in fielders and most of the time no change in park.

So it looks like the big difference between guys who change teams and guys who don't is the fielding. You could say guys staying on the same team see a change in fielding. But that is probably small compared to the guys who change teams.

And you could say that for guys who change teams the change in parks is a big deal. But you only pitch about half the time in one park, whereas it probably the same mix of fielders behind you all the time. So again, that points to the fielders.

Monday, June 8, 2015

Year to year correlation of BABIP for pitchers that changed teams and those that did not

I found all the guys who had 150+ IP in the years 2003-14.  If a guy pitched for two teams in a year, that guy in that year was disqualified. I found the year to year correlations for SO, BBs, HRs and BABIP (SO, BBs, HRs were all as a percentage of batters faced, which excluded IBBs and I included HBP in BBs). I used the Baseball Reference Play Index. (June 11-I made some corrections as mentioned in my first comment)

There were 488 cases of a guy being on the same team the next year (of course, if John Smith was on the Reds from 2011-13, we have one case of his 2011-12 seasons and another case of his 2012-13 seasons)

There were 107 cases of guys being on a different team the next year. Here are the correlations for each group

 Same team

 BABIP    0.234
 BB          0.704
 HR          0.384
 SO          0.811

 Different team

 BABIP    0.094 (p value is .33)
 BB          0.692
 HR          0.305
 SO          0.733

The p-value in all cases except for the one listed are under .01. So the only one which is not significant (and it is not even close) is BABIP for guys that changed teams.

Now the correlations are lower for the guys that switched teams for all four stats. But the difference for BABIP is by far the biggest.

It is not clear what causes this. The pitchers have a whole new set of fielders behind them and they pitched half their games (or around that) in a completely different park than the year before. It seems like if the pitchers had the most control over what happens on balls in play, the correlation for guys that change teams would be closer to that of guys who stayed on the same team.

Phil Birnbaum suggested I run a regression in which BABIP in year 2 depends on BABIP in year 1. Here it is for guys that stayed on the same team.

BABIP2 = .234*BABIP1 + .226

No the regression for guys that were on a different team

BABIP2 = .093*BABIP + .268

The effect is 2.53 times higher if you stay on the same team. That is, if you stayed on the same team, your BABIP in year 1 is about 2.5 times stronger in predicting your BABIP in year 2 than if you switched teams.

Friday, June 5, 2015

Team OPS Differentials Thru 6-4-15

The season is about a third over.


TEAM OPS OPSA DIFF Pct
LA Dodgers 0.789 0.668 0.121 0.574
St. Louis 0.730 0.655 0.075 0.667
Kansas City 0.738 0.668 0.070 0.588
Houston 0.722 0.666 0.056 0.618
Oakland 0.710 0.658 0.052 0.411
Pittsburgh 0.698 0.655 0.043 0.547
Tampa Bay 0.697 0.655 0.042 0.527
San Francisco 0.733 0.699 0.034 0.545
NY Yankees 0.723 0.696 0.027 0.537
Cleveland 0.734 0.709 0.025 0.491
Toronto 0.765 0.740 0.025 0.455
Detroit 0.737 0.715 0.022 0.509
Washington 0.723 0.701 0.022 0.537
LA Angels 0.684 0.666 0.018 0.519
Chicago Cubs 0.699 0.690 0.009 0.538
NY Mets 0.674 0.671 0.003 0.545
Cincinnati 0.707 0.716 -0.009 0.442
Texas 0.731 0.741 -0.010 0.519
Baltimore 0.707 0.718 -0.011 0.453
Arizona 0.733 0.747 -0.014 0.472
Seattle 0.689 0.703 -0.014 0.444
Miami 0.683 0.699 -0.016 0.407
Colorado 0.750 0.780 -0.030 0.462
Boston 0.683 0.728 -0.045 0.436
Atlanta 0.689 0.738 -0.049 0.491
Minnesota 0.687 0.738 -0.051 0.604
San Diego 0.676 0.758 -0.082 0.491
Milwaukee 0.659 0.760 -0.101 0.333
Philadelphia 0.635 0.739 -0.104 0.382
Chicago Sox 0.651 0.756 -0.105 0.462

It looks like the big an0molies are the Twins and the A's. Twins are 11-7 in 1-run games (about what they normally do). The A's are 3-15. Both their pitchers and hitters are really bad when it is late & close. See tables below.

The Twins hitters doing much better with runners on but not so good when it is late & close:


Twins  Hitters


Split BA OBP SLG OPS
RISP 0.299 0.371 0.451 0.823
None on 0.232 0.276 0.355 0.631
Men On 0.280 0.343 0.421 0.764
Late & Close 0.199 0.264 0.280 0.544

Now the Twins pitchers. Seems to be the opposite of the hitters. So-so with runners on but great when it is late & close.


Twins  Pitchers


Split BA OBP SLG OPS
RISP 0.264 0.345 0.404 0.749
Men On 0.273 0.340 0.417 0.757
None on 0.270 0.305 0.418 0.723
Late & Close 0.211 0.257 0.289 0.545

Now the A's hitters. They seem okay with runners on but are pretty bad when it is late & close.


A's Hitters


Split BA OBP SLG OPS
RISP 0.254 0.310 0.414 0.724
None on 0.258 0.320 0.387 0.706
Men On 0.254 0.318 0.397 0.715
Late & Close 0.241 0.294 0.337 0.631

Now the A's pitchers. They are doing poorly in RISP and late & close situations


A's Pitchers


Split BA OBP SLG OPS
RISP 0.260 0.345 0.410 0.755
Men On 0.241 0.323 0.371 0.693
None on 0.235 0.291 0.341 0.632
Late & Close 0.276 0.350 0.429 0.779

Tuesday, June 2, 2015

How Overmatched Were The Mets Against The Orioles In 1969?

Very. The Orioles had the 4th highest OPS differential ever (from 1914-2014). It was .136. The Mets had just .017 (and several NL teams had better that year). So the Orioles had an advantage of .119.

The only time there was a greater difference before 1969 for the teams meeting in the World Series was in 1944 when the Cardinals had an OPS differential of .130 and the Browns had just .005. So the Cards had an advantage of .125. But that was a war year.

The 1927 Yankees had the greatest OPS differential of .196 and their opponent, the Pirates, had .084 (which is very good). So the Yanks had an advantage of .112.

Thru 1969, these are the only cases of an advantage being at least .100. There were also only three other cases reaching even .080. 42 cases had an advantage of less than .050. And that is out of 56 World Series.

So the Orioles came into the 1969 Series with an incredible edge over the Mets but the Mets still won.

If we go by OPS differential, 1969 was the greatest upset ever. The next biggest was

1931: A's had a .109 differential, Cardinals .045. A's had a .064 advantage yet Cards won.

So with an advantage of .119, the Orioles losing to the 1969 Mets was not just the biggest upset (at least since 1914), it was far and away the biggest.

One year before 1914 that was a big upset was 1906 when the White Sox beat the Cubs. But we don't have opponents SLG for before those years. But we can make an educated guess about how big of an advantage the Cubs had.

The Cubs had an OPS of .667 on offense while the Sox had .588. That gives the Cubs an advantage of .079.

The pitchers of each team allowed an OBP of .280. For the Cubs to have an edge of .119 in OPS differential, they would have to have allowed an SLG of .040 less than the Sox.

The Cubs did allow their opponents a batting average of .207 while it was .239 for the Sox. Now if each allowed the same isolated power, then the Cubs would have an SLG which is .032 lower. That would bring their advantage up to .111 (.079 + .032). That would make 1906 the 4th biggest upset.

The Cubs allowed 12 HRs and the White Sox 11. That suggests that they had the same isolated power allowed. But if the Cubs had allowed an ISO that was .008 less than the Sox, it would bump their advantage up to .119 and the upset would be just as big as the one in 1969 (SLG = AVG + ISO).

Since ISO is extra bases divided by ABs, it would mean that the Cubs would have to have allowed about 39 fewer extra bases (based on 4,918 ABs, which we can get by dividing their hits allowed by .207).

That would mean something like the Cubs allowing 20 fewer 2Bs and 10 fewer 3Bs. That would be 40 extra bases. That seams plausible. But we don't know for sure. There is just a possibility that 1906 was a bigger upset than 1969.

Monday, June 1, 2015

Harper Has Fourth Best Monthly OPS+ Ever For A Player 22 Or Younger

Here is the top 20 with 75+ PAs for the month using the Baseball Reference Play Index:


Rk Player Month Year PA OPS+
1 Ted Williams Sept/Oct 1941 101 297
2 Ted Williams July 1941 80 284
3 Ted Williams August 1941 157 279
4 Bryce Harper May 2015 109 279
5 Joe DiMaggio July 1937 139 277
6 Ted Williams Sept/Oct 1939 127 265
7 Eddie Mathews Sept/Oct 1954 97 263
8 Boog Powell June 1964 86 259
9 Cesar Cedeno June 1972 127 251
10 Mike Trout June 2014 102 248
11 Ted Williams June 1941 125 243
12 Bernie Carbo May 1970 80 239
13 Mike Trout July 2012 112 238
14 Pete Reiser Sept/Oct 1941 114 235
15 Ken Griffey July 1991 94 235
16 Johnny Bench June 1970 104 235
17 Johnny Bench May 1969 77 233
18 Al Kaline August 1956 131 233
19 Jimmie Foxx May 1929 121 232
20 Hank Aaron July 1956 130 231

Now for 23 and under. Harper is 6th. He is in excellent company on each list


Rk Player Month Year PA OPS+
1 Ted Williams Sept/Oct 1942 90 348
2 John Mayberry June 1972 76 299
3 Ted Williams Sept/Oct 1941 101 297
4 Ted Williams July 1941 80 284
5 Ted Williams August 1941 157 279
6 Bryce Harper May 2015 109 279
7 Reggie Jackson June 1969 126 277
8 Joe DiMaggio July 1937 139 277
9 Ted Williams May 1942 125 268
10 Ted Williams Sept/Oct 1939 127 265
11 Dick Wakefield Sept/Oct 1944 127 264
12 Eddie Mathews Sept/Oct 1954 97 263
13 Boog Powell June 1964 86 259
14 Eddie Mathews June 1955 118 256
15 John Mayberry Sept/Oct 1972 131 252
16 Alex Rodriguez May 1999 75 251
17 Rudy York August 1937 126 251
18 Cesar Cedeno June 1972 127 251
19 Mike Trout June 2014 102 248
20 Jose Canseco Sept/Oct 1988 108 247