Saturday, July 28, 2018

Jack Morris Did Not Excel At ERA+, FIP ERA, Pitching To The Score, Strikeout-To-Walk Ratio Or Pitching In Big Games Even Though He Had Very Good Fielding Behind Him

This is a post I did last December.

Jack Morris had one top 5 finish in ERA+ and none in FIP ERA.

Now you might say that ERA or ERA+ can be affected by fielders, so if a guy has bad fielders, it hurts his ERA.

But Morris had good fielders like Trammel, Whitaker and Chet Lemon behind him. For most of his time with the Tigers, they were above average in Defensive Efficiency Rating and several times were number 1 (I think 4-5 times). So it seems unlikely that his ERA could have been hurt by bad fielders.

So then when we adjust his ERA for park effects with ERA+, he managed only one top 5 finish. So he was rarely one of the elite pitchers in the league.

Not having any top finishes in FIP ERA might not be his fault if Tiger stadium was an easy park to hit HRs in. But he actually had only one top 10 finish in his career and that was a 6th. Don't forget that walks and strikeouts affect FIP ERA along with HRs. But anyway, we can always go back to ERA+, which, again, does not seem to be unfair to him since it seems like he had good fielders behind him.

He also had only one top 5 finish in strikeout to walk ratio

Now some people said he was a winner. But his career winning pct of .577 is only 205th best all-time. He also had only three top 5 finishes and never led the league, even though he pitched for some good teams.

Some have said he could pitch to the situation. But here are his OPS allowed stats for Low, Medium, and High Leverage situations

Low .694
Medium .692
High .695

Now maybe the norm for starters is to do much worse in high leverage cases. But this does not look that impressive

He was supposed to be a big game pitcher. But in 92 post-season innings he has an ERA of 3.80. Good, but not great.

The table below shows the Tigers Defensive Efficiency Rating each year from 1979-1990, the years when Morris was a regular starting pitcher with them. It also shows where they ranked in the league each year as well as the league average. Then averages for all three are shown. He had some good fielders behind him.



Year DER Rank Lg Avg
1979 0.705 5 0.700
1980 0.704 4 0.698
1981 0.740 1 0.711
1982 0.725 1 0.704
1983 0.726 1 0.699
1984 0.713 3 0.699
1985 0.717 2 0.706
1986 0.719 1 0.699
1987 0.705 3 0.697
1988 0.718 1 0.702
1989 0.690 12 0.698
1990 0.705 5 0.699

0.714 3.25 0.701

Wednesday, July 25, 2018

Does plate discipline lead to higher averages with runners in scoring position?

Last Saturday during the Astros-Angels game one of the announcers said something about how Bregman has become more disciplined this year and knows the strike zone better, so that has helped him hit better with RISP (he is 5th in the league or MLB, can't recall which).

I asked on the Yahoo Statistical Analysis group if there are any studies that show that guys with more plate discipline do better with RISP, either in general or relative to what they normally hit.
D.K. Willardson responded and did a study, looking at the relationship between OSwing % and RISP AVG minus Overall AVG (OSwing % is the percentage of pitches outside the strike zone that batters swing at). D.K. looked at Qualified Hitters 2015-2017.

The r-squared was .0372, meaning that only 3.72% of the variation in Overall AVG minus RISP AVG is explained by OSwing %. So guys who are more disciplined do improve with RISP, but not much.

Thursday, July 19, 2018

A three true outcome All-Star game

Not only was there an above average number of HRs in the game, there was a slightly above average number of walks and strikeouts.

There were 9 walks in 10 IP.  The MLB average per team is 3.3 per 9 IP. So that would be 6.6 for two teams. 9 in 10 is above average.

There were 25 strikeouts in the game or 12.5 per team. The MLB average per 9 IP is 8.5.

Half of the hits were HRs. 10 out of 20.

The average team gets 7.3 non-HR hits per game. So you would expect 14.6 in 9 innings. But in 10 innings there were only 10.
Maybe most all-star games are like this. But this game did shatter the HR record.

Update July 24: Diane Firstman tweeted on July 20 "it was an all-time record high of 48.9% (44 TTO / 90 PAs)."

Her blog is Value Over Replacement Grit: We are the VORG. Replacement is futile. Her poster presentation on TTOs won an award at the recent SABR convention. See SABR 48: Maddox, Firstman win 2018 convention presentation awards.

Sunday, July 15, 2018

Today is the anniversary of the "Grich game" when he almost single handedly defeated the Yankees and Ron Guidry in 1979

Click here to go the box score and play by play at Baseball Reference.

Bobby Grich went 4 for 5 (all against Guidry). That included a HR and a 2B with 5 RBIs in a 5-4 come from behind win for the Angels. Grich had a win probability added (WPA) of 1.211, the 9th highest that we know of. Guidry won the Cy Young award the year before while leading the league in ERA. He would lead the AL again in ERA in 1979. This was a home game for the Angels.

Here are Grich's ABs:

1st: Lineout with one out and none on trailing 2-0.

3rd: Single with runners on 1st and 2nd and one out to drive in a run to make the score 4-1.

5th: Single with a man on first and two outs. But next batter makes an out.

7th: Double with runners on 1st and 2nd and one out. Both runners score to make it 4-3.

9th: HR with man on 1st and two outs. Gives the Angels a 5-4 victory in front of 40,739 fans (also in 2 hours 19 minutes).

As mentioned earlier, Grich had a win probability added (WPA) of 1.211, the 9th highest that we know of (using the Baseball Reference Play Index-for games of at least .9, the earliest is 1925). Click here to see all those games. Art Shamsky had a game of 1.503 in 1966, which is the best. WPA uses historical data to estimate how much every change in the base-out-score situation changes a team's probability of winning. The later and closer the game, the more a hit increases WPA.

That famous "Sandberg game" when Ryne hit two clutch HRs off of Bruce Sutter is 42nd with 1.063. I have written about that game. It was impressive, but Sutter was no longer in is prime like Guidry was in 1979. Grich's game is far better. See my post from a few years ago Where Does "The Sandberg Game" Rank In WPA? I explain how hitting a HR off of Sutter was not especially hard at that point in his career.

Guidry as well rested coming into the game. He went 6 innings on July 10th, so he was not over worked. He did face 38 batters in this game, including 10 Ks and 2 BBs. So he could have easily thrown over 100 pitches.

He faced Grich for the 5th time that game in the 9th inning. Guidry's OPS allowed that year when facing batters for the 4th time or more in a game was .863. Pretty high. But the year before it was just .334 and the year after it was .627. So Guidry was not necessarily going to have problems in this situation. In 1977 it was .739

Here is what all AL pitchers allowed facing batters for the 4th time or more in a game in from 1977-80 with the league average OPS for all PAs after it in parentheses.

1977: .746 (.735)
1978: .719 (.711)
1979: .765 (.743)
1980: .766 (.731)

So batters did somewhat better facing a pitcher for the 4th time or more, but in the two previous years, not by much. Yes, Guidry faced Grich for the 5th time in the 9th. But he was the Cy Young award winner the previous year (with a 1.74 ERA). So it is not surprising a manager would stick with him.

Rich Gossage, one of the Yankees' best relievers, worked 3.2 innings the day before and another good righty reliever, Ron Davis had pitched 2.1. Davis also pitched on July 13th. Looking at the Yankee roster, those guys seem to have been the best possible options.

Grich was a righty and Guidry was a lefty. But Guidry did not get hit that hard by righties. Here are his HRs allowed divided by PAs for the years 1977-80:

1977: 9/606
1978: 11/757
1979: 17/776
1980: 12/729

So a manager would not have worried too much about Guidry having to face a righty.

Here is the OPS Guidry allowed vs. righties:

1977: .629
1978: .561
1979: .667
1980: .724

Here is the OPS Guidry allowed during innings 7-9:

1977: .677
1978: .428
1979: .755
1980: .659

Again, no big indicators that it would be a problem to leave Guidry in. Grich did hit 30 HRs that year but before that he had never hit 19. He did have 18 HRs in 358 PAs thru July 14, which is a pretty good total.

Here are his OPS vs. lefties:

1976: .794
1977: .807 (in only 50 PAs, he was hurt)
1978: .796
1979: .909 (but it was .900 vs. righties)
1980: .547

So it looks like there would have been no reason for a manager to especially fear Grich when facing a lefty. Good, but not devastating numbers.

Update 7-16-18: Grich had only faced Guidry 7 times before this game in his career and only once before that season. Grich was 2 for 7 with a single, a double and 3 strikeouts. Sandberg had faced Sutter 12 times including 4 times that season.

Friday, July 13, 2018

Was Bobby Murcer one of the most underrated players of the '70s?

That is what someone said on Twitter today. But who underrated him? How do we measure that?

I have tried to estimate this before and links are provided below. If you read them and click on links to the complete rankings, Murcer does not appear to have been underrated.

Here I just looked at where he ranked in WAR among position players and in MVP shares among position players. Here are the WAR leaders from the 1970s using the Baseball Reference Play Index

WAR leaders of the 1970s.

Murcer is 38th.

I also copied and pasted all the MVP votes from 1970-1979 from both leagues into Excel and found the cumulative total for each player.  I took out any pitchers. Murcer is 36th, very close to his rank in WAR. So it seems that he was rated just about right by the MVP voters. He had three top 10 finishes (a 7th, a 5th and a 9th).


Name Share
Willie Stargell  3.16
Joe Morgan  2.99
Johnny Bench  2.67
Jim Rice  2.1
Steve Garvey  2.06
Reggie Jackson  2.05
Pete Rose  1.98
Dave Parker  1.97
George Foster  1.94
Rod Carew  1.7
Greg Luzinski  1.52
George Brett  1.51
Thurman Munson  1.5
Ken Singleton  1.41
Fred Lynn  1.4
Billy Williams  1.32
Sal Bando  1.32
Mike Schmidt  1.22
Joe Rudi  1.08
Don Baylor  1.04
Lou Brock  1.01
Bobby Bonds  1
Joe Torre  0.99
Reggie Smith  0.99
Dick Allen  0.98
Amos Otis  0.93
Brooks Robinson  0.81
Jeff Burroughs  0.79
John Mayberry  0.78
Frank Robinson  0.75
Hank Aaron  0.73
Boog Powell  0.72
Mickey Rivers  0.7
Ted Simmons  0.67
Tony Perez  0.67
Bobby Murcer  0.66

Was Willie Mays The Most Underrated Player In History? Or Was It Wade Boggs? Is Albert Pujols The Most Overrated? (Revised)

Using A Player's WAR To Predict First Year Hall Of Fame Vote Percentage (and possibly estimate "underratedness")

Thursday, July 5, 2018

Mariners specialize in high leverage situations

Now you might think it is just a good bullpen. But they hit much better in high leverage situations, too. Overall, the Mariners have an OPS of .742 and have allowed .709. That .033 differential is modest. Using my regression generated formula of

Pct = .5 + 1.3*OPSDIFF

That gives them about a .543 pct when it is actually .632. So they have won about 8 more games than expected (they are now 55-32).

Here is what their hitters are doing. Data from Baseball Reference.

Split BA OBP SLG OPS
High Leverage 0.298 0.357 0.505 0.862
Medium Leverage 0.268 0.330 0.414 0.744
Low Leverage 0.235 0.296 0.388 0.683

Now what the pitchers allow.

Split BA OBP SLG OPS
High Leverage 0.250 0.313 0.361 0.674
Medium Leverage 0.241 0.298 0.407 0.705
Low Leverage 0.246 0.306 0.426 0.731

Now the OPS differentials. So they play like the 1927 Yankees in High Leverage situations (the Yanks that year had an overall differential of .196).


Split OPS OPS Diff
High Leverage 0.862 0.674 0.188
Medium Leverage 0.744 0.705 0.039
Low Leverage 0.683 0.731 -0.048