Morris vs. Blyleven

The Hall of Fame results will be announced next week, and the subject of Jack Morris continues unabated. I admit that it’s hard for me to take the Morris Hall of Fame campaign seriously, but Chris Jaffe predicts he’ll get 65 percent of the vote. That’s astonishing to me.

A lot of people have written a lot about Morris, and I’m not going to repeat it all here. But one of the things I’ve heard from a Morris defender is this: Now that Bert Blyleven‘s in the Hall, Morris should be, too. I’d like to put that notion to rest.

First, there are the surface stats: Blyleven pitched 4,970 innings, struck out 3,701 batters and had a 3.31 ERA. When you compare his ERA to his contemporaries (and adjust for ballpark), his ERA+ is 118. That means that the adjusted league-average ERA was 18 percent higher than his.

Morris pitched 3,824 innings, struck out 2,478 batters (5.8 per nine innings vs. Blyleven’s mark of 6.7 per nine innings) and had a 3.90 ERA. His ERA+ was 105, which means that the league-average ERA (adjusted for ballpark) was five percent higher than his. Morris was a fine pitcher, but he was no Blyleven.

Some people then say, “Fine, but Morris was a winner.” It’s true that, statistically, Morris had a better winning percentage than Blyleven. Morris won 254 games and lost 186 for a winning percentage of .577, while Blyleven won 287 games and lost 250 for a winning percentage of .534. The implication is that Morris was a “winner,” and Blyleven wasn’t. This perception is enhanced by Morris’ fantastic performance in the seventh game of the 1991 World Series.

This is ridiculous. Minnesota in the early 1970s (Blyleven’s prime) was a .500 team. Detroit in the early 1980s (Morris’ prime) was a .540 team (I’m just guesstimating here, but I don’t think I’m off by much). Morris typically pitched for much stronger offenses than Blyleven did. The difference in their win/loss records has everything to do with context and nothing to do with pitching performance.

Still, the perception persists. I’ve heard it said that Morris was better than Blyleven at “pitching to the score.” This typically means that, sure, Morris sometimes gave up six runs, but only when Detroit scored 10. He matched his performance to his surroundings.

Better baseball analysts than I, such as Bill James and Joe Sheehan, have already picked this argument apart. But I’d like to do it one more time, the easy way. I’d like to show the difference in each pitcher’s Win Probability Added (WPA).

WPA is discussed here. The upshot is that players are given credit for helping their teams win games based on what they did on every single play. You can find WPA stats at both Fangraphs and Baseball Reference; it’s easy to find and easy to understand. Plus, it captures performance and context. It reflects someone “pitching to the score.”

Let me give you a few examples from 1974, one of Blyleven’s finest years.

On April 20, Blyleven lost a 1-0 game against the Rangers, giving up the winning run in the bottom of the ninth. He pitched extremely well but ultimately gave up the winning run, resulting in a WPA of just 0.15. His contribution was very positive, and his WPA was positive, but it didn’t result in a win. Given that a winning team compiles 0.5 WPA points in total at the end of a game, this was a good-not-great total.

On the other hand, on Sept. 25, the Twins beat the A’s 1-0 with Blyleven pitching a complete-game shutout. His WPA total was a fantastic 0.77, one of the best totals you can achieve in a game. He pitched extremely well AND he gave up fewer runs than his team scored. As a result, his WPA total was 0.62 higher in this game than on April 20.

On June 24, the Twins beat the Rangers, 8-4. Blyleven pitched just six innings and left with a 4-3 lead. Bill Hands shut down the Rangers, and the Twins scored four more runs to put the game away. Blyleven’s WPA in this game was 0.01—average.

Lastly, on July 13, the Twins won a 2-1 game against the Indians in 11 innings. Blyleven pitched nine innings and gave up just one run, keeping the Twins in the game until they won it with Bill Campbell on the mound. Blyleven’s WPA total was 0.41, an excellent total matching his performance with the context of the game and ultimately resulting in a win. Note that even if the Twins had lost the game, Blyleven’s WPA would have remained 0.41, as he was off the mound when the final outcome was determined.

The point is that WPA totals reflect both performance and context. It does a much better job of uncovering how well pitchers pitched to the situation than won-loss record does. (As an example, Campbell picked up the win in that July 13 game; Blyleven got a no-decision). Short of examining each game one-by-one (which James and Sheehan did), it’s the single best way of capturing how well pitchers “pitched to the score.”

Here are the WPA totals for Morris and Blyleven, by age:

 Age     Morris  Blyleven
  19               0.3
  20               3.4
  21               1.7
  22      -0.1     3.5
  23      -0.9     3.7
  24       2.3     4.2
  25       1.3     4.1
  26       1.8     2.1
  27       1.4     0.3
  28       1.4    -0.2
  29       1.2    -1.0
  30       3.0     1.9
  31       3.2    -0.3
  32       3.6     0.5
  33       0.1     3.4
  34      -1.9     2.5
  35      -2.7     0.0
  36       2.1     1.6
  37       2.1    -2.9
  38      -2.0     4.5
  39      -1.0    -1.1
  41              -1.7
Total     15.0    30.6

It’s not close. Blyleven’s career total is twice Morris’, and his peaks were higher, too. Blyleven had seven years with a WPA total above 3.0; Morris had three. Blyleven topped 4.0 three times; Morris never did.

Unlike some folks I know, I’m willing to give Morris some credit for his fantastic seventh game heroics. But that still doesn’t make the comparison between these two great pitchers close. Only one is a Hall of Famer.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Melvin Mora career highlights
Next: 30th anniversary: Tony C’s heart attack »

Comments

  1. Dave Studeman said...

    tc, as I showed in my post, Morris’ peak doesn’t compare to Blyleven’s. Arguing that they were similar is ridiculous.

    If you believe, after all the evidence, that wins and losses are a better measure of a pitcher’s contribution to a team’s winning than Win Probability, I don’t know what to say.

  2. David said...

    Personally, I don’t like WPA for pitchers too much.  It gives all defensive credit for any outcome to the pitcher, which I think is a really dangerous way to go about distributing credit.  Really, it it not too different from ERA, or to a lesser extent, W/L.  Still, I would agree with the conclusion.

    As for peak value, I would call Blyleven’s peak his first 10 years (if we’re going with a consecutive 10-year period as suggested above).  We can use ‘79-‘88 for Morris, and ‘70-‘79 for Blyleven.

    Morris:
    173-112 (.607); 2471 IP; 1627 Ks (5.9 K/9); 3.86 RA; 114 ERA+; 33.4 rWAR; Tigers were .552 in that period.

    Blyleven:
    148-128 (.536); 2624.7 IP; 2082 Ks (7.1 K/9); 3.25 RA; 130 ERA+; 56.3 rWAR; Blyleven’s teams were .523 in that period.

    For the Blyleven analysis, I used the Twins’ record through June 4th, 1976, since his first start with the Rangers was June 5th, and I used the Rangers’ record thenceforth.

    Anyway, Blyleven’s winning percentage wasn’t much better than his teams’ winning percentages . . . but then again, they weren’t that great of teams.  Morris played on NO losing teams in his 10-year peak.  And he was very good, but Blyleven’s numbers are clearly far better.

    For the record, during their ten-year peaks, Blyleven was on one World-Series-winning team, and his postseason record in three games (two starts) was 2-0 with a 1.28 ERA.  Morris was ALSO on one World-Series-winning team, and was 3-1 with a 3.00 ERA.

    I’d go with Blyleven over a career, or a 10-year peak, or a single season (Blyleven in 1973 vs. Morris in 1983).  It has to be Bert.  Personally, I don’t think there’s a whole lot to discuss here.

  3. Derek Ambrosino said...

    Preach, Dave!

    The Morris for the HOF movement is crazy. Those who support Morris often tout the fact that he won more games than any pitcher in the 80’s. …Great, so he topped the league in an extremely flawed stat over an entirely arbitrary period of time. …John Lackey won 12 games last year while setting a record for highest ERA among an SP with like 60 IP or something, meanwhile his own teammate, Josh Beckett finished 5th in the league in ERA and ERA, 3rd in pitcher WAR, and amassed more than 7 more WAR than Lackey.

    And, then we get the randomness of the 90s. Great, Morris is the winningest pitcher of 90s. And, Mark Grace had the most hits over the same arbitrary period of time. And, the gestation period of an elephant is 22 months. …As long as we’re spouting of random factoids with no relevance to anything at hand.

    Who won the most games from 1977 – 1986? Anybody?… The answer is – who cares.

    And, finally, if the random arbitrary cut-off of Jack Morris’s winning decade is relevant, why aren’t other random, contrived cut-offs relevant?

    Say hello to the last man to hit .400 – Wade Boggs, who did so over a 162 game stretch spanning mid-1985 to mid-1986.

    Sorry for the snark, but this “debate” gets me worked up. Bert Blyleven may be a bit of a perv with the telestrator pen, but he’s one of the top 25 starting pitchers to ever play the game. Jack Morris is nowhere close to his caliber and a borderline HOF-er at best.

  4. Dave Studeman said...

    Agreed that WPA gives pitcher credit for things that fielders do, but so what? Most pitching stats do. And, for this particular exercise, Blyleven struck out more batters than Morris.  If anything, removing this bias makes Morris’ case even less compelling.

    Comparing team W/L records to pitcher W/L records is a poor man’s WPA.  It’s more valid than straight W/L records, but still misleading.

  5. tc said...

    @Dave Studeman- I know that you don’t think that W-L matters, that it is a surface stat, and that is because you believe that quantifiable stats are the only measurement that should be considered and for you W-L speaks more to the team effort than to the individual. I hope I got that right. But to me, no matter what the quality of the team, the guy on the mounds job is to keep his team in the game and funnily enough, the best pitchers do just that. Walter Johnson was a classic example of this. He won despite being on terrible teams for the most part. His W-L pct. was significantly higher than the W-L pct. of the teams he played for, .599 to .480, 119 pts. higher. Morris, during his peak was @ .607 and his team was .553. Time and time again, the best pitchers are on the hill for a higher pct. W-L than their teams. Now you can dismiss W-L as an unquantifiable stat, but what I’m saying is that all of those other stats are what add up to W-L, and in the end, the best pitchers wind up winning games, consistently. It’s the stats persons job to figure out how they do it. OK, so Blyleven struck out more batters, so what? Is striking out batters the object of the game? The question to a stat person is, how did Morris do it, not well he doesn’t fit our set of stats so therefore he was just lucky to be on a good team. That wouldn’t explain Walter Johnson very well. I suspect that Sparky loved Morris because he shouldered the load for the team, pitched deep into the game and saved the bullpen for the lesser pitchers on the staff. I don’t know this for a fact, but I suspect Morris derided pitchers who pitched for stats. His teammates probably saw this guy as their Clint Eastwood, someone who no matter how tough it got, fought through his best and worst and still kept the team in position for the win. I know you don’t respect opinions like mine because you are a pure science person, if you can’t measure it, it doesn’t exist, but my view is there are a lot of pitchers on good teams who were no where near as consistent in getting the team into the win column as Morris, so why was that? What stats can explain that phenomenon and if you can’t explain it then you need to look deeper.

  6. tc said...

    I believe the point to Morris’ record during the ‘80s is not some arbitrary time frame but meant to represent his peak. To get in the HOF you must either have a great peak or a great overall and I guess Blyleven got in on his overall. If you measure Morris’ peak from say ‘79 to ‘88, you come up with 173-112 W-L @ .607%. Not bad. His ERA, if I figured it right was 3.55 and he gave up 8.0 HPG. He won, in total, 20 games 3 times and almost did it 2 other times. I believe old Tiger stadium was a hitters park and that should go to his credit as well. Given all of that, I guess if you open the door to Blyleven you might as well at least consider Morris. He even had a few pretty good off peak years. However,I watched ‘em both play and personally, I don’t think either one belongs.

  7. Dave Studeman said...

    tc, your post is silly, as are your criticisms of me.  When you refer to wins and losses, you’re using measurements too.  They’re just bad measurements.  You’re practicing “science” as much as I am.  The difference is that you’re practicing bad science.

    WPA addresses your point of “winning” exactly.  It looks at the outcome of every single at bat, given the score, inning and situation, and measures how well the pitcher moved his team toward victory.  Pure wins and losses don’t do that.  Again, go read the work of Sheehan and James for a game-by-game account of whether Morris truly was a “winner” in a manner that escapes measurement.

    Striking out batters is important because it is an indication of how much pitchers relied on their fielders.  I notice that not many Morris advocates mention Trammell, Whitaker, Parrish and Lemon when discussing Morris.  They should certainly be given some of the credit you’re heaping on Morris.  Blyleven rarely had a fielding team that strong behind him.

    However, you’re not acknowledging that credit, nor are you acknowledging the worthiness of my proposed better measurements, because you’re biased to believe that Jack Morris was a winner.

  8. Derek Ambrosino said...

    Given all of that, I guess if you open the door to Blyleven you might as well at least consider Morris.

    [Begrduingly] opening the door for Blyleven… Really? Then we might was well at least consider Morris? Really?

    Yeah, the HOF “opened the door” to the player with the 13th highest pitching WAR in history – right behind Randy Johnson and ahead of guys like Nolan Ryan, Steve Carlton, and Bob Gibson. …Now, that’s just 1 number, but so it W-L record, and my number is a lot more sound than yours.

    FTR, I don’t believe Bert to be the 13th greatest pitcher of all time, but he’d starting making a strong case in the early 20s. Morris on the other had ranks 141 in career pitcher WAR – about as far behind the immortal Brad Radke as Blyleven is behind the big unit.

    And, as far as the dynamics of Tiger Stadium – that’s accounted for in ERA+. As Dave mentioned, Morris posted a career 105 ERA+  Blyleven 118. So, relative to their peers and adjusted for home park, Blyleven was 13% better at preventing the other team from scoring (earned) runs, over 30% more innings than Morris pitched. That’s a tremendously large gap in production.

  9. tc said...

    If you took what I said to be a criticism of you, I apologize. I didn’t mean it that way. My point, putting it as simply as I can is that science should explain success and not discount it. There have been many different paths to a successful pitching career but success in baseball for a pitcher is measured by whether he helps the team win. Jack Morris was not just some lump on the mound of a good team. As I said before, there have been plenty of pitchers on good teams who have been nowhere near as successful as Morris at reaching the win column and the answer should be something more than that he was just fortunate. Don’t bother responding, I’ll bag it with this apology. I thought there was an honest difference of opinion and I meant nothing more than that.

  10. Dave Studeman said...

    tc, I don’t mind being criticized.  In fact, I like it.  You’re right that we do have an honest difference of opinion. Nothing to apologize for.

    I never said Jack Morris was a “lump on the mound of a good team,” nor did I say he was “just fortunate.”  I’m not “discounting” what Morris did.

    I’m simply saying that Blyleven did much more than Morris did to help his teams win.  I’m saying that pitcher wins and losses are a poor way to measure that effect, and that WPA is much better.

  11. Derek Ambrosino said...

    For a long time, my big 3 EGREGIOUS HOF snubs were Santo, Allen, and Blyleven. 2 of those 3 have been righted; Allen is LONG overdue. For 11 seasons, his OPS+ never dropped below 145. His defense was atrocious, and he was probably born too soon to DH, but he was an absolutely elite offensive force. His career was also hurt and delayed too by all the racism he faced in the minors, which continued in his early pro days. He was clearly ready considerably before he was brought up, considering he led the league in offensive WAR his first full season.

  12. Dave Studeman said...

    Yes.  Buy the Hardball Times Annual.  See link in the upper left-hand column.

    FWIW, David Freese’s heroics were in the sixth game of the World Series, so his totals don’t compare to Morris’.

  13. lexomatic said...

    “Say hello to the last man to hit .400 – Wade Boggs, who did so over a 162 game stretch spanning mid-1985 to mid-1986”
    Actually Derek, I believe Tony Fernandez had 2 consecutive halves hitting over .400 – 2nd half in ‘98, 1st half in 99. Could be ‘97-8 though, I forget.

  14. Todd said...

    The Annual has championship WPA for all players ever? Or for this postseason?

    I went ahead and tried hand calculating what Freese had for just those last two games. If that 166 number is what I need to use, then he had 166 * .194 + (166/2) * .953 = 111.3 for those last two games. Less then Morris’ 141, but still really big.

  15. Dave Studeman said...

    If I understand your math, you’re comparing Freese’s Champ WPA in two games to Morris’ Champ WPA in one.  The entire point of the original post was to give some perspective to Morris’ single game impact.

  16. Todd said...

    So, based on the Morris for HoF as a result of game 7 1991 argument, how does David Freese’s case look after this year?

    Alternatively, is there anywhere I can look up championship WPA, rather then having to hand calculate it?

  17. Paul E said...

    Regarding the he had the most wins in the 1980’s or most hits in the 1990’s type of argument, a little bit of trivia:

    For the period 1964 -1973, amongst players with at least 800 games played, the leader in OPS+ was…… Dick Allen @ 165 – followed by 9 (NINE) Hall of Famers. Let me qualify this by saying I believe Allen was, at the least, a Hall of Fame talent. And, frequently, the best player on the field.

    I just don’t think that type of argument regarding a specific period for a player’s prime is the “be all and end all” of arguments for a HoF candidacy. And, in Morris’ case it seems to be the only argument I’ve ever heard. It’s just not enough.

  18. Derek Ambrosino said...

    Lexo,

    Fernandez flirted with .400 for a while in ‘99. In a stretch from late ‘98 through mid ‘99, he hit .388 over a string of 124 games and 438 ABs. That’s the closest I can get with Tony.

    Using the mathematical first and second halves of the two seasons, Fernandez hit .342 in the second half of ‘98 and .372 for the first half of ‘99.

    Boggs hit .395 in the second half of ‘85 and .363 in the first half of ‘86, but from June 9, 1985 through June 6, 1986, Boggs hit exactly .400 over the span of 162 games.

    From July 1, 1993 through July 1, 1994 Tony Gwynn hit .398.

  19. tim said...

    tc – let’s say in a crazy run of games, a pitcher throws 10 games of 7 or 8 innings, allowing 0-1 runs per start with only a few hits per game, and his team’s offense is so anemic it scores no runs and they lose, or they score a couple in the 8th-9th. that pitcher would end up going 0-4 or something like that. his era would be microscopic, but his W-L would look like junk. which is the better measure for his performance during that run? he was a victim of his team’s offensive struggles.

  20. BobDD said...

    Somebody demanded that stats explain why Morris won so many games – that stats should be able to show how he got that record.  Easy enough.

    Jack Morris’s career run support, relative to league average – 107.1% (100 being average missy).  So Morris should win 7% more games than his pitching ability.  254 suddenly becomes 236.  There’s more but already our questioner was right – stats can explain that rather simply after all.

    Jack Morris was very good and any team would have been ecstatic to have him, after all he was about 88.x% as good as Blyleven and that certainly has a lot of value.

  21. BobDD said...

    My math was correct, but my logic was off.  That 7% was how many more runs his teams scored; not how many more wins.  Actual wins would have been affected by more than that 7% amount (only logical assumption used here), but I do not have a proven formula to share.  So I apologize for overrating my methods and the resultant overrating of Jack Morris.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>