Remember me

Back to the Future, Part 4?

November 5, 2019
Introduction
 
Let’s play a little "Carnac":
 
A: Marty McFly and Dave Stieb
Q: Name 2 people who benefit from "Alternate History".
 
We’re all familiar with the story of Marty McFly, the main character in the "Back to the Future" movie trilogy. Marty had an unsatisfying family life in 1985 – a jailbird uncle, siblings that were losers, a mom who drank too much, and a father who had a miserable job where his boss ("Biff") is the same jerk who bullied him in high school. What happens from there? Combine a crazy scientist and a time-traveling DeLorean and go back 30 years into the past, and, despite a few bumps in the road that nearly erased Marty from existence, you return to the present, voila, Marty has that brand new vehicle he had his heart set on, and his family is now a bunch of winners. Even the bully, rather than being the boss and tormentor of Marty’s dad, now washes his cars. 
 
All because of a little intervention and the resulting alternate history. And everything was fine until Part 2, when Marty got greedy and inadvertently gave "Future Biff" the idea to go back in time and give "Young Biff" the sports magazine that allowed him to win a fortune by betting on sporting events where he knew the outcome. But that’s another story……
 
That is the premise for this article.  I wanted to take a look at how things might be different if we applied today’s Cy Young Award standards to the past and created a bit of our own alternate history. Who might have won some awards, and who might have lost some? Would certain players be thought of differently now when we reflect on their careers?
 
Setup
 
In the last article I wrote, I referenced a tool developed by Tom Tango a few years ago that attempts to project the winners of the Cy Young awards each year by calculating a point total based on 4 simple data markers (innings pitched, earned runs allowed, strikeouts, and wins). The more points you have, the more likely you are to win the award. 
 
It’s appealing to me because:
1)      It’s proven to be very accurate
2)      It’s very simple
 
The basic formula is Cy Young Points = ​(IP/2 - ER) + SO/10 + W (there’s another version he developed that would extend to relievers as well, but for the most part, the overwhelming majority of true Cy Young Award contenders at this point are starters). It captures 4 key attributes that appear to be fundamental in voter evaluation of starting pitchers:
 
  • Workload (Innings)
  • Run prevention (Earned Runs)
  • Ability to make batters miss (Strikeouts)
  • Contributing to team victories (Wins)
 
It does not incorporate WAR, ERA+, FIP, or any other "advanced metric". It’s proven to be very accurate simply by using only those 4 basic stats. In the past 26 Cy Young Award votes (covering 13 years), the formula identifies the correct winner 23 times, and in the other 3, it has the actual winner no lower than 2nd. It also has done a really good job of predicting who the other top finishers will be. 
 
Note that, like Bill James’ Hall of Fame Monitor tool, Tango’s Cy Young Points does not try to determine who deserves to win the award – it’s merely trying to predict who will win. And, it does a really good job of that.
 
Now, one thing I like about it is that it is designed to reflect the current trends in Cy Young voting. It’s clear, especially when you see pitchers like Felix Hernandez (13-12) in 2010 and Jacob deGrom (10-9) in 2018 receiving the award despite mediocre W-L records, that wins don’t carry as much weight as they used to, and voters are putting more weight on run prevention, workloads, and strikeouts. Wins still matter to some degree, but it seems to be true that they matter more in a complementary sense, where if 2 pitchers are close in other regards, the one with more wins might get an edge. But, simply piling up a large win advantage by itself without having strong performances in other categories doesn’t carry the weight that it used to. 
 
So, what I decided to do, with a lot of help from baseballmusings.com (which has a page that lets you plug in a date to see how many Tango Cy Young Points a player had for a given point in time in a season), is to take Tango’s Cy Young Award prediction formula, which clearly captures the standards of today, and apply it to all of the years dating back to the first Cy Young Award season of 1956 to see who might have won the award instead. In effect, creating our own "alternate history".
 
Now, I must confess, this isn’t a "serious" exercise. It’s more what I would call an exercise in funtility. Not futility, mind you…….but "fun"-tility. It’s meant to be fun, to perform a little bit of "what if" analysis, even though it won’t change anything. After all, no one is going to take those awards away from anyone, nor should they. Awards provide a snapshot in time, and there’s a lot to be said for how players are evaluated by on-the-spot observers in real time, using whatever means and information they have at their disposal to confer honors.   There’s nothing that says that today’s standards are better than yesterday’s. 
 
In fact, I acknowledge that there are issues with even doing this, as pitchers’ stats were very different in years past – they typically made more starts, carried heavier workloads, didn’t strike out as many batters per inning, and you would tend to see more impressive seasonal win totals. The dynamic and the context of pitchers’ stats were very different then vs. now, so in that sense it may not be logical, even in a fun exercise, to imply that you can take a formula designed to predict award winners in today’s game and apply it to previous eras. Nevertheless, I decided to do it anyway to have a little fun and just to see what results it yielded.
 
So, it’s time to pack a bunch of baseball writers into a Delorean, set the speed for 88 miles per hour, crank up the flux capacitor to 1.21 gigawatts, and travel back to each year since 1956 to see what kind of alternate history we can create.
 
Back to the Past
 
Let’s begin with an example of how this works. Ironically, we begin in 1956, only 1 year after the point in time that our buddy Marty traveled into the past. 1956 marked the inaugural Cy Young Award and, as you probably know, from 1956 through 1966 only one Cy Young was awarded annually for the Major Leagues as a whole, rather than a separate award for each league.
 
Let’s take a look at 1956. The Cy Young award went to Don Newcombe, who had an impressive 27-7 record along with a 3.06 ERA for the pennant-winning Dodgers. Newcombe was also named the NL MVP for that year.
 
This is the result of the voting per baseball-reference.com:
 
Rank
Name
Tm
Vote Pts
1st Place
Share
WAR
W
L
ERA
IP
BB
SO
1
Don Newcombe
BRO
10.0
10.0
63%
4.5
27
7
3.06
268.0
46
139
2
Sal Maglie
BRO
4.0
4.0
25%
5.0
13
5
2.89
196.0
54
110
3
Whitey Ford
NYY
1.0
1.0
6%
5.2
19
6
2.47
225.2
84
141
3
Warren Spahn
MLN
1.0
1.0
6%
5.5
20
11
2.78
281.1
52
128
 
It was an interesting vote. The ballot format at that time consisted of each voter voting for just a single pitcher. Newcombe won, but it wasn’t a complete runaway, as Newcombe’s teammate Maglie also got 4 votes despite a modest 13-5 record. I’m actually kind of surprised that Maglie finished 2nd ahead of many others who had significantly higher win totals (Maglie also finished 2nd in the NL MVP vote that year, also behind Newcombe). The only things I can figure are that:
 
1)      Maglie pitched for the pennant-winning Dodgers, and that carried some weight

2)      Maglie was terrific down the stretch, going 8-1 in his last 10 starts, including tossing a no-hitter in late September, and the Dodgers ended up winning a tight race by 1 game over the Braves and 2 over the Reds. I suspect the voters were impressed by those key, clutch performances.
 
Anyway, what if we applied the Tango formula to 1956? Here would be the top contenders:
 
Name
W
L
IP
ER
SO
ERA
Tango Cy Young Points
Herb Score
20
9
249.3
70
263
2.53
100.97
Early Wynn
20
9
277.7
84
158
2.72
90.63
Warren Spahn
20
11
281.3
87
128
2.78
86.47
Whitey Ford
19
6
225.7
62
141
2.47
83.93
Don Newcombe
27
7
268.0
91
139
3.06
83.90
Johnny Antonelli
20
13
258.3
82
145
2.86
81.67
Frank Lary
21
13
294.0
103
165
3.15
81.50
Lew Burdette
19
10
256.3
77
110
2.70
81.17
Billy Pierce
20
9
276.3
102
192
3.32
75.37
Bob Lemon
20
14
255.3
86
94
3.03
71.07
Bob Friend
17
17
314.3
121
166
3.46
69.77
Jack Harshman
15
11
226.7
78
143
3.10
64.63
Ron Kline
14
18
264.0
99
125
3.38
59.50
Paul Foytack
15
13
256.0
102
184
3.59
59.40
Sal Maglie
13
5
196.0
63
110
2.89
59.00
 
As you can see, Newcombe had a big advantage in wins, and he did carry a pretty decent workload, but several others had lower ERA’s and more impressive strikeout totals. This formula would have Newcombe in a fairly tight cluster among several pitchers like Spahn, Ford (who led the Majors in ERA), Antonelli, Lary, and Burdette.
 
The winner according to this approach would be the young phenom, Herb Score. Score had 7 fewer wins than Newcombe and about 20 fewer innings pitched, but he was dominant in strikeouts (he struck out 263 batters that year, 37% more than the #2 man, Billy Pierce of the White Sox with 192) and finished just behind Ford for the lowest ERA in the Majors. Applying this formula, Score would have been predicted to win the award.
 
One other thing. As we go along, I’m going to presume that, by today’s standards, no relievers would be named as an alternate history Cy Young Award winner. 9 relievers have won the Cy Young award, but since 1992 (when Dennis Eckersley won) only 1 reliever (Eric Gagne in 2003) has won. My apologies to them, but relievers won’t be represented in our alternate history.
 
Anyway, that’s how this works. Let’s look at some results, one group at a time.
 
Group #1 – 1956-1966
 
This first group captures the era when there was only 1 Cy Young Award winner for the Major Leagues as a whole. To save some time and space, I’m not going to list all the stats for both pitchers each year, but will call out some observations as we go along.
 
Note – The green highlights indicate that a different winner would be assumed using today’s standards. No highlight means the same pitcher would be expected to win.
 
Year
Winner
Would Now Finish
Winner by Today's Standards
1956
Don Newcombe 
5th
Herb Score
1957
Warren Spahn 
2nd
Jim Bunning
1958
Bob Turley 
6th
Whitey Ford
1959
Early Wynn 
6th
Sam Jones
1960
Vern Law 
6th
Don Drysdale
1961
Whitey Ford 
1st
Whitey Ford 
1962
Don Drysdale 
1st
Don Drysdale 
1963
Sandy Koufax 
1st
Sandy Koufax 
1964
Dean Chance 
1st
Dean Chance 
1965
Sandy Koufax 
1st
Sandy Koufax 
1966
Sandy Koufax 
1st
Sandy Koufax 
 
The first 5 seasons under this premise would have resulted in different pitchers winning, with most of the actual winners finishing 5th or 6th
 
1958 is an interesting one – Turley took home the award with a 21-7, 2.97 season. Turley’s teammate Ford did not receive any Cy Young votes, and his 14-7 record doesn’t compare to Turley’s, but Ford did lead the Majors (by a wide margin) with a 2.01 ERA.
 
In 1959, Jones basically pitched better than Wynn, but Wynn pitched for the AL pennant-winning White Sox, and that I’m sure helped him when it came time to hand out the award.   1960 was kind of similar – Law had a much better record (20-9) than Drysdale (15-14), but Drysdale, by any other reasonable standard, was a better pitcher that year. As you can see, the early Cy Young winners were generally on pennant-winning teams, as clearly that was a factor that helped those candidates.
 
The last 6 seasons of this first group were pretty straightforward, as the winners all had pretty dominant seasons across the board (including 3 spectacular Koufax seasons) , and the formula implies that they all would still win under today’s standards. Here are some of the impressive stat lines for that group:
 
Year
Winner
Team
W-L
ERA
IP
SO
1961
Whitey Ford
New York (A)
25-4
3.21
283
209
1962
Don Drysdale
Los Angeles (N)
25-9
2.83
314.1
232
1963
Sandy Koufax
Los Angeles (N)
25-5
1.88
311
306
1964
Dean Chance
Los Angeles (A)
20-9
1.65
278.1
207
1965
Sandy Koufax
Los Angeles (N)
26-8
2.04
335.2
382
1966
Sandy Koufax
Los Angeles (N)
27-9
1.73
323
317
 
Group #2 – 1967-1979
 
The next group begins with 1967, the first season that the leagues named separate Cy Young Award winners. Over these 13 seasons (26 awards), the formula would imply that over half (14) would go to someone else now. Note that 1969 had a shared award (AL, Cuellar and McLain tied).
 
 
Year
Lg
Winner
Would Now Finish
Winner by Today's Standards
1967
AL
Jim Lonborg
5th
Gary Peters
1967
NL
Mike McCormick
5th
Jim Bunning
1968
AL
Denny McLain
1st
Denny McLain
1968
NL
Bob Gibson
1st
Bob Gibson
1969
AL
Mike Cuellar
1st
Mike Cuellar
1969
AL
Denny McLain
2nd
 
1969
NL
Tom Seaver
5th
Bob Gibson
1970
AL
Jim Perry
3rd
Sam McDowell
1970
NL
Bob Gibson
2nd
Tom Seaver
1971
AL
Vida Blue
1st
Vida Blue
1971
NL
Fergie Jenkins
2nd
Tom Seaver
1972
AL
Gaylord Perry
1st
Gaylord Perry
1972
NL
Steve Carlton
1st
Steve Carlton
1973
AL
Jim Palmer
3rd
Nolan Ryan
1973
NL
Tom Seaver
1st
Tom Seaver
1974
AL
Catfish Hunter
3rd
Nolan Ryan
1974
NL
Mike Marshall
x
Phil Niekro
1975
AL
Jim Palmer
1st
Jim Palmer
1975
NL
Tom Seaver
1st
Tom Seaver
1976
AL
Jim Palmer
2nd
Frank Tanana
1976
NL
Randy Jones
3rd
J.R. Richard
1977
AL
Sparky Lyle
x
Nolan Ryan
1977
NL
Steve Carlton
1st
Steve Carlton
1978
AL
Ron Guidry
1st
Ron Guidry
1978
NL
Gaylord Perry
3rd
Phil Niekro
1979
AL
Mike Flanagan
1st
Mike Flanagan
1979
NL
Bruce Sutter
x
J.R. Richard
 
The 1970’s marked the first appearance of relievers as Cy Young winners. Marshall, Lyle, and Sutter each took home an award during this period.
 
Who would be the main beneficiaries during this period? Well, Nolan Ryan, in large part to the impressive strikeout totals, would be presumed to have won 3 awards at the expense of Jim Palmer (1973), Catfish Hunter (1974), and Sparky Lyle (1977). Is that reasonable? Actually, I think it is. For one thing, Ryan was among the top candidates anyway in all 3 of those ballots, finishing 2nd, 3rd, and 3rd, respectively (with the ’73 and ’77 voting actually being pretty tight).   His W-L records weren’t very sparkling, but the Angels were pretty bad in those days, and that likely wouldn’t have been held against him today, at least not to the same degree. His strikeout totals were, of course, awesome, and he was among the league ERA leaders in all of those years.  
 
Who else? Well, J.R. Richard comes up as a 2-time winner in 1976 and 1979, although the formula implies a close race in 1976 among Richard, Randy Jones, Tom Seaver (despite a 14-11 record), and Jerry Koosman, so anyone could have won it. It does show that 1979 would have been a convincing win for Richard (and his 313 K’s), although Phil Niekro logged over 340 innings that season.
 
Speaking of Phil Niekro, he shows as winning 2 awards, in ’74 over Marshall, and ’78 over Perry. In both years, his impressive innings pitched totals would have weighed heavily in his favor..
 
Tom Seaver won 3 actual Cy Young’s in the ‘60’s and ‘70’s, but this projects him actually winning 4, losing his 1969 award to Bob Gibson, but picking up 2 incremental wins in ’70 and ’71 (Seaver led the league in ERA and K’s both of those years).
 
I’m not convinced that Seaver would have actually lost the ’69 award – the formula shows a pretty strong 6-man race with Seaver, Bob Gibson, Juan Marichal, Bill Singer, Larry Dierker, and Gaylor Perry, and Seaver mostly rates behind the others because of fewer K’s and a lighter workload, but he certainly had the best W-L record of the group. And, of course, it was the year of the Amazin’ Mets. I suspect he would have still won, which would have translated to 5 total Cy Youngs in this alternate history.
 
Other winners in this time frame would include Gary Peters, Jim Bunning (to add to the mythical award he would have won in ‘57), Sam McDowell, and Frank Tanana.
 
Group #3 – 1980-1989
 
A fun decade – this approach implies that 12 of the 20 awards would go to other people, with most of the changes occurring in the first half of the decade.
 
Year
Lg
Winner
Would Now Finish
Winner by Today's Standards
1980
AL
Steve Stone 
3rd
Mike Norris
1980
NL
Steve Carlton 
1st
Steve Carlton 
1981
AL
Rollie Fingers 
x
Steve McCatty
1981
NL
Fernando Valenzuela 
2nd
Steve Carlton 
1982
AL
Pete Vuckovich 
6th
Dave Stieb
1982
NL
Steve Carlton 
2nd
Steve Rogers
1983
AL
La Marr Hoyt 
7th
Jack Morris
1983
NL
John Denny 
2nd
Mario Soto
1984
AL
Willie Hernandez 
x
Dave Stieb
1984
NL
Rick Sutcliffe 
10th
Dwight Gooden
1985
AL
Bret Saberhagen 
3rd
Dave Stieb
1985
NL
Dwight Gooden 
1st
Dwight Gooden 
1986
AL
Roger Clemens 
1st
Roger Clemens 
1986
NL
Mike Scott 
1st
Mike Scott 
1987
AL
Roger Clemens 
1st
Roger Clemens 
1987
NL
Steve Bedrosian 
x
Orel Hershiser
1988
AL
Frank Viola 
1st
Frank Viola 
1988
NL
Orel Hershiser 
1st
Orel Hershiser 
1989
AL
Bret Saberhagen 
1st
Bret Saberhagen 
1989
NL
Mark Davis 
x
Orel Hershiser
 
Observations on the 1980’s:
 
·         This decade is already notorious for some controversial selections. Cy Young voters continued to fall in love with closers, continuing what began in the ‘70’s, by awarding the Cy Young 4 times to relievers.

This was also the decade that Bill James exploded on the scene (his Baseball Abstract reached a much larger audience in the early 1980’s), and you started to see a lot of debate around the traditional methods of evaluation. Steve Stone, Pete Vuckovich, and Lamar Hoyt had seasons that had analysts like Bill questioning the wisdom of the selections. 

·         Dave Stieb and Orel Hershiser both show as benefitting greatly from today’s standards. 

This formula implies Stieb taking home the award in ’82, ’84, and ’85, replacing Vuckovich, Hernandez, and Saberhagen. Of course, those 3 likely benefitted from being key contributors on pennant-winning teams.

Stieb actually finished 4th in the voting in ’82, and tied for 7th both in ’84 and ’85. But, he finished first in WAR for AL pitchers in ’82, ’83, and ’84, and was 2nd in’85, and was also top 5 in ERA each of those years. I think he was clearly the best pitcher in the AL during that time frame, and probably the best in the Majors. 

Hershiser took home an award in 1988 for his outstanding 23-8, 2.26 performance that year, but this formula implies that in both ’87 and ’89 (rather than closers Bedrosian and Davis), Hershiser could have won as well, which would have made him a 3-time winner, with all 3 coming in a row.  It’s easy to see why he didn’t win either of those seasons – his record in ’87 was 16-16 with a 3.06 ERA, and in ’89 he was 15-15, despite a sparkling 2.31 ERA, and in those days, it was very unlikely that a pitcher with a .500 W-L record would win the award. Despite those records, Hershiser actually did place pretty high in the balloting both years – he finished 4th in 1987 (getting 2 first place votes), and 4th again in 1989 (picking up 1 first place vote, and finishing 2nd in the NL in ERA). It’s not difficult to see a re-vote going in his favor, especially using today’s standards.

So, here’s a question – what would Hershiser’s and Stieb’s Hall of Fame case looks like if they each had 3 Cy Young Awards under their belts? I think Hershiser has the better case….his career record’s a little stronger, plus he has the consecutive inning streak thing, as well as an outstanding postseason record, which would probably give him a leg up on Steib, but 3 Cy Youngs would give both of them some serious brownie points. There are pitchers like Bret Saberhagen and Johan Santana who have 2 Cy Youngs and aren’t in, but keeping out someone with 3? I think voters would give them serious consideration.

·         By this methodology, Jack Morris, Mike Norris, Steve McCatty, Steve Rogers and Mario Soto would all now have a Cy Young award on their resumes that they could boast about.
 
Group #4 – 1990-1999
 
Year
Lg
Winner
Would Now Finish
Winner by Today's Standards
1990
AL
Bob Welch 
4th
Roger Clemens 
1990
NL
Doug Drabek 
3rd
Frank Viola 
1991
AL
Roger Clemens 
1st
Roger Clemens 
1991
NL
Tom Glavine 
1st
Tom Glavine 
1992
AL
Dennis Eckersley 
x
Roger Clemens 
1992
NL
Greg Maddux 
1st
Greg Maddux 
1993
AL
Jack McDowell 
7th
Kevin Appier
1993
NL
Greg Maddux 
1st
Greg Maddux 
1994
AL
David Cone 
1st
David Cone 
1994
NL
Greg Maddux 
1st
Greg Maddux 
1995
AL
Randy Johnson 
1st
Randy Johnson 
1995
NL
Greg Maddux 
1st
Greg Maddux 
1996
AL
Pat Hentgen 
1st
Pat Hentgen 
1996
NL
John Smoltz 
2nd
Kevin Brown
1997
AL
Roger Clemens 
1st
Roger Clemens 
1997
NL
Pedro Martinez 
1st
Pedro Martinez 
1998
AL
Roger Clemens 
1st
Roger Clemens 
1998
NL
Tom Glavine 
3rd
Kevin Brown
1999
AL
Pedro Martinez 
1st
Pedro Martinez 
1999
NL
Randy Johnson 
1st
Randy Johnson 
 
In contrast to the ‘80’s the 90’s look relatively uneventful, as only 6 of the 20 awards would theoretically have gone to other candidates. I think, in large part, that’s due to the presence of so many great pitchers who dominated the decade. Where as the 1980’s lacked true dominant pitchers, the ‘90’s gave us legendary Hall of Fame pitchers  like Roger Clemens, Greg Maddux, Randy Johnson, Pedro Martinez, John Smoltz, and Tom Glavine. Those 6 pitchers claimed 14 of the 20 Cy Young awards, and actually 15 if you include Dennis Eckersley, who took home the award as a reliever in 1992. Even if you apply the formula to this era, those same pitchers would still have taken home 14 of the 20 awards. Great pitchers, great performances. 
 
About the only significant shake up would be that Kevin Brown is identified as someone who could have taken home a couple of awards, in 1996 with the Marlins (17-11, 1.89) and 1998 with the Padres (18-7, 2.38).   Would Brown have received much more support than the 2.1% he captured in his only time on the Hall of Fame ballot if he had a couple of Cy Youngs in his pocket? 
 
1996, if put to a vote today, would be interesting – you have Smoltz with the stellar 24-8, 2.94 record vs. Brown. Smoltz would have had the advantage in wins by 7, but also in innings by about 20 (253 vs. 233), and a big margin in K’s (276 vs. 159). However, Brown’s ERA was a full run per game lower. In fact Brown’s ERA of 1.89 was about 80 points better than the #2 man in the NL (Maddux with 2.72), which was kind of reminiscent of deGrom’s margin in 2018 (1.70 vs. the 2.32 of the #2 man, Aaron Nola). Nowadays, a margin that big tends to bode well for the ERA leader.

The formula also implies that, in addition to 3 other Cy Youngs that Roger Clemens won in the decade, he could have taken home a couple more (in 1990 and 1992).
 
Group #5 – 2000-2009
 
Year
Lg
Winner
Would Now Finish
Winner by Today's Standards
2000
AL
Pedro Martinez 
1st
Pedro Martinez 
2000
NL
Randy Johnson 
1st
Randy Johnson 
2001
AL
Roger Clemens 
4th
Mike Mussina
2001
NL
Randy Johnson 
1st
Randy Johnson 
2002
AL
Barry Zito 
2nd
Pedro Martinez 
2002
NL
Randy Johnson 
1st
Randy Johnson 
2003
AL
Roy Halladay 
4th
Pedro Martinez 
2003
NL
Eric Gagne 
x
Mark Prior
2004
AL
Johan Santana 
1st
Johan Santana 
2004
NL
Roger Clemens 
3rd
Randy Johnson 
2005
AL
Bartolo Colon 
3rd
Johan Santana 
2005
NL
Chris Carpenter 
3rd
Roger Clemens 
2006
AL
Johan Santana 
1st
Johan Santana 
2006
NL
Brandon Webb 
1st
Brandon Webb 
2007
AL
C.C. Sabathia 
1st
C.C. Sabathia 
2007
NL
Jake Peavy 
1st
Jake Peavy 
2008
AL
Cliff Lee 
1st
Cliff Lee 
2008
NL
Tim Lincecum 
1st
Tim Lincecum 
2009
AL
Zack Greinke 
1st
Zack Greinke 
2009
NL
Tim Lincecum 
1st
Tim Lincecum 
 
In the first decade of the 2000’s, there’s a bit of upheaval in the first half of decade, but then fairly stable over the last half.
 
The big news here is that the formula implies that Roger Clemens would have lost a couple of instances where he won, but that he would have picked up a new one that he didn’t win. 
 
In addition, Johan Santana would have picked up a 3rd award (in 2005) which would have given him 3 in a row from 2004-2006.    Now, that would have really made things interesting come Hall of Fame ballot time. Santana only received 2.4% of the vote in his lone appearance on the ballot (in 2018), in large part because he had a relatively short career. What if he had won 3 consecutive Cy Youngs, though? Would he have received significantly greater support? 
 
Also notable – Pedro Martinez, who already has 3 actual Cy Young awards, could have picked up a couple of more, giving him 5 in total for his career.
 
Group #5 – 2010-2019
 
Year
Lg
Winner
Would Now Finish
Winner by Today's Standards
2010
AL
Felix Hernandez 
1st
Felix Hernandez 
2010
NL
Roy Halladay 
1st
Roy Halladay 
2011
AL
Justin Verlander 
1st
Justin Verlander 
2011
NL
Clayton Kershaw 
1st
Clayton Kershaw 
2012
AL
David Price 
2nd
Justin Verlander 
2012
NL
R.A. Dickey 
1st
R.A. Dickey 
2013
AL
Max Scherzer 
1st
Max Scherzer 
2013
NL
Clayton Kershaw 
1st
Clayton Kershaw 
2014
AL
Corey Kluber 
2nd
Felix Hernandez 
2014
NL
Clayton Kershaw 
1st
Clayton Kershaw 
2015
AL
Dallas Keuchel 
1st
Dallas Keuchel 
2015
NL
Jake Arrieta 
1st
Jake Arrieta 
2016
AL
Rick Porcello 
2nd
Justin Verlander 
2016
NL
Max Scherzer 
1st
Max Scherzer 
2017
AL
Corey Kluber 
1st
Corey Kluber 
2017
NL
Max Scherzer 
1st
Max Scherzer 
2018
AL
Blake Snell 
1st
Blake Snell 
2018
NL
Jacob deGrom 
1st
Jacob deGrom 
 
As you would expect, the current decade is pretty stable as the formula is pretty much spot on. But for what it’s worth, it implies that Verlander, who had a unanimous selection for the AL Cy Young (and also won the AL MVP) in 2011, could have easily picked up a couple of more wins, in 2012 and 2016. Not to mention that he is a strong candidate to win the 2019 AL Cy Young as well.
 
The other alternate history change would have awarded the 2014 Cy Young to Felix Hernandez rather than Corey Kluber. As it turns out, that was a very close vote – Kluber had the Wins and strikeout edge, but Hernandez led the league with a 2.14 ERA.   It could have gone either way.
 
Wrapping it Up
 
Here are a couple of summary tables to wrap it up…..
 
This first one shows, by decade, how many times a different winner was implied (the 1970’s and the 1980’s being the most volatile). In total, 41% of the awards would have gone to a different pitcher in our alternate universe:
 
Decade
Awards Given
 Different Winner
% of Different Winners
1950s
4
4
100%
1960s
14
5
36%
1970s
20
11
55%
1980s
20
12
60%
1990s
20
6
30%
2000s
20
7
35%
2010s
18
3
17%
Grand Total
116
48
41%
 
Here’s another interesting (but a bit long) table – this shows which pitchers would benefit most from the alternate history, sorted by the net gain/loss, and headed by Nolan Ryan and Dave Stieb, who each would have improved from 0 wins to 3 wins. Jim Palmer, on the other hand, would have lost 2 of his 3 awards.
 
26 pitchers who were single-time winners in their careers would now have lost them. 19 pitchers who were shut out during their actual careers would win at least one, including Stieb and Ryan with 3 each, and Niekro, Brown, Bunning and Richard with 2 each.
 
Pitcher
Cy Young Awards in Alternate History
Cy Young Awards in Actual History
Net Gain/Loss
Nolan Ryan
3
0
3
Dave Stieb
3
0
3
Pedro Martinez 
5
3
2
Orel Hershiser
3
1
2
Justin Verlander 
3
1
2
Phil Niekro
2
0
2
Kevin Brown
2
0
2
Jim Bunning
2
0
2
J.R. Richard
2
0
2
Roger Clemens 
8
7
1
Randy Johnson 
6
5
1
Tom Seaver
4
3
1
Johan Santana 
3
2
1
Whitey Ford
2
1
1
Don Drysdale
2
1
1
Dwight Gooden 
2
1
1
Felix Hernandez 
2
1
1
Frank Viola 
2
1
1
Jack Morris
1
0
1
Kevin Appier
1
0
1
Herb Score
1
0
1
Mario Soto
1
0
1
Mark Prior
1
0
1
Mike Mussina
1
0
1
Mike Norris
1
0
1
Sam Jones
1
0
1
Sam McDowell
1
0
1
Steve McCatty
1
0
1
Steve Rogers
1
0
1
Gary Peters
1
0
1
Frank Tanana
1
0
1
Steve Carlton 
4
4
0
Greg Maddux 
4
4
0
Sandy Koufax 
3
3
0
Clayton Kershaw 
3
3
0
Max Scherzer 
3
3
0
Tim Lincecum 
2
2
0
Bob Gibson
2
2
0
Zack Greinke 
1
1
0
Jacob deGrom 
1
1
0
Jake Arrieta 
1
1
0
Jake Peavy 
1
1
0
Dean Chance 
1
1
0
David Cone 
1
1
0
Mike Cuellar 
1
1
0
Mike Flanagan 
1
1
0
Mike Scott 
1
1
0
Dallas Keuchel 
1
1
0
Pat Hentgen 
1
1
0
C.C. Sabathia 
1
1
0
R.A. Dickey 
1
1
0
Blake Snell 
1
1
0
Ron Guidry 
1
1
0
Cliff Lee 
1
1
0
Brandon Webb 
1
1
0
Vida Blue 
1
1
0
Denny McLain 
1
2
-1
Corey Kluber 
1
2
-1
Gaylord Perry 
1
2
-1
Roy Halladay 
1
2
-1
Bret Saberhagen 
1
2
-1
Tom Glavine 
1
2
-1
Chris Carpenter 
0
1
-1
Dennis Eckersley 
0
1
-1
Don Newcombe 
0
1
-1
Doug Drabek 
0
1
-1
Early Wynn 
0
1
-1
Eric Gagne 
0
1
-1
Fergie Jenkins 
0
1
-1
Fernando Valenzuela 
0
1
-1
Bob Turley 
0
1
-1
Jack McDowell 
0
1
-1
Jim Lonborg 
0
1
-1
Jim Perry 
0
1
-1
David Price 
0
1
-1
John Denny 
0
1
-1
John Smoltz 
0
1
-1
La Marr Hoyt 
0
1
-1
Mark Davis 
0
1
-1
Bruce Sutter 
0
1
-1
Mike Marshall 
0
1
-1
Mike McCormick 
0
1
-1
Pete Vuckovich 
0
1
-1
Bartolo Colon 
0
1
-1
Randy Jones 
0
1
-1
Rick Porcello 
0
1
-1
Rick Sutcliffe 
0
1
-1
Barry Zito 
0
1
-1
Rollie Fingers 
0
1
-1
Sparky Lyle 
0
1
-1
Steve Bedrosian 
0
1
-1
Steve Stone 
0
1
-1
Catfish Hunter 
0
1
-1
Vern Law 
0
1
-1
Warren Spahn 
0
1
-1
Willie Hernandez 
0
1
-1
Bob Welch 
0
1
-1
Jim Palmer 
1
3
-2
 
Thanks for reading,
Dan 
 
 

COMMENTS (13 Comments, most recent shown first)

briangunn
This is an awesome exercise - thank you. Would love to see what would happen if you recalculated MVP voting using modern standards.
12:32 PM Nov 7th
 
MarisFan61
Dan: I'd have the same comment.
For this kind of thing, I consider those terms synonymous.

I'm curious about what Tom thinks about what the margin needs to be before he'd said it's a 'prediction' or a 'projection' or any such.

Let me be definitive. :-)
I think the number for Cole and Verlander indicate an essential TIE.

And lemme put it this way: Suppose rather than 99.6 vs. 98.5, it were 99.6 vs. 99.4.
(I'm purposely picking a second number that would round off to the lower integer.)

Do you think for a second that Tom would say the difference is either a predictor or projector or any such?

I don't. And if he would, he would be nuts. :-)

So, the issue is where we draw the line -- and I'm wondering where he does, and in fact, where you would.
(You haven't said!)
10:51 PM Nov 6th
 
DMBBHF
Maris,

How about if we say that his formula "projects" Cole to be the winner, rather than "predicts"? It's true that "Predicting" does imply a higher level of confidence, so maybe it would be fairer to say "projects" or "forecasts" Cole to be the winner.

Thanks,
Dan
10:31 PM Nov 6th
 
MarisFan61
Dan,
Just to make sure that you and the readers realize, re what you said about this year's A.L.:
"Tango's formula predicts Cole in the AL and deGrom in the NL for 2019."

Not necessarily.

It's correct only if Tom (or anybody at all) thinks that the formula is "predictive" even if the margin is tiny.

I'd be curious if Tom thinks a margin such as this one (99.6 vs 98.5) is predictive.
I think it isn't.

Mind you, I'm not saying it won't be Cole. In fact at this point I'd guess it will be. At first, following the end of the regular season, I was thinking Verlander is more likely, for reasons I explained on Reader Posts. But Cole is commonly being said to be the favorite, maybe even the presumptive winner, and so I'd guess it will be him.
But on a theoretical basis, I doubt that such a differential on such a formula could be well predictive, and I have doubts that the author of the formula would think it is.

To be clear: What I'm wondering isn't which guy Tom thinks will win.
It's whether he thinks as a general matter that such a differential is enough to call it a 'prediction.'
9:28 PM Nov 6th
 
DMBBHF
Thanks for all the comments.

Maris,

I really wasn't looking at the margins when I was pulling the results for each year, but it's an interesting question. Maybe Tom will see this and give his thoughts.

LesLein,

Tango's formula predicts Cole in the AL and deGrom in the NL for 2019.

Vanasek,

In the years 1956-1966, if there had been 1 award for each league, the formula would have identified:

1956 AL - Herb Score
1956 NL - Warren Spahn
1957 AL - Jim Bunning
1957 NL - Warren Spahn
1958 AL - Whitey Ford
1958 NL - Warren Spahn
1959 AL - Hoyt Wilhelm
1959 NL - Sam Jones
1960 AL - Jim Bunning
1960 NL - Don Drysdale
1961 AL - Whitey Ford
1961 NL - Jim O'Toole
1962 AL - Hank Aguirre
1962 NL - Don Drysdale
1963 AL - Camilo Pascual
1963 NL - Sandy Koufax
1964 AL - Dean Chance
1964 NL - Don Drysdale
1965 AL - Sam McDowell
1965 NL - Sandy Koufax
1966 AL - Jim Kaat
1966 NL - Sandy Koufax


Thanks,
Dan

8:26 PM Nov 6th
 
Gfletch
Here's a thought based on a superficial impression (just looking at the lists you presented comparing each season's winners based on Tango's formula vs who truly did win).

I wouldn't really object to either set of winners. Both sets look like they have identified reasonable choices, reasonably impressive pitchers. That suggests to me that our sabermetric formulas, though they have acquired the veneer of scientific credibility (and often deserve such a judgment), are in some ways just belief systems.

5:13 PM Nov 6th
 
Vanasek
Who would have won the awards if you had applied your formula in 1956-66 for each league instead of only one award winner for all of baseball? I'm guessing that someone like Jim Kaat, might have picked up a Cy Young during the Sandy Kofax years.
3:42 PM Nov 6th
 
LesLein
Who should win this year?
1:03 PM Nov 6th
 
LesLein
Bill James wrote in 1983 that Carlton deserved the award. He didn’t have a single cheap win. His manager left him in games where others would have been taken out. His ERA was misleading, not his won-lost record.
1:02 PM Nov 6th
 
Guy123
Very interesting analysis. Nice work.

I do wonder whether Ryan would actually have won 3 times under today's standards. Tango's model does not include BBs, and I'm sure that has almost no affect on the model's predictive power in the large majority of cases. However, Ryan was an extreme outlier on this dimension. He not only led the AL walks in all three of these seasons (1973, 1974, 1977), his walk rate was vastly higher than any other CYA contender.

One way to consider BBs is to look at pitchers' FIP-, a metric that considers K, BB, and HR. Ryan's rank in FIP- was 2nd, T5th, and T4th in these three seasons. Ryan also never led the AL in WAR (either version) in these seasons. I think modern CYA voters would take account of his huge walk totals, likely costing Ryan at least 2 of these hypothetical awards.
12:24 PM Nov 6th
 
MarisFan61
(typo -- helps me know to avoid it!)
12:37 AM Nov 6th
 
MarisFan61
P.S. Thanks for saying what happens in Part 2 (or whatever it's called!) of Back to the Future.
Help me know to avoid it. :-) I like how Part 1 came out.
(Recently saw it on TV, first time since the original time. I liked it even better this time. Very charming movie.)
12:36 AM Nov 6th
 
MarisFan61
Nice article, interesting as always!
I especially love anything that pumps up players that I think deserve more than they've gotten. That includes Stieb. I considered him among the top couple of pitchers of the early '80's (with Morris.....BTW, grammar question: is this an example where indeed it's correct to say "among" even when it's just 2? Never mind, I don't expect you to answer that).


I do have a real question :-) not much related to your most prominent stuff but which is about how you did it; really more about Tango's method.

It's a basic question about using the method. Unless I missed something, you don't refer to it except a little bit by implication, in mentioning how close the N.L. 1969 race comes out by Tango's formula.

How much of a margin does there need to be in order for the formula to be well correlated with the actual Cy Young results? How much of a margin is needed for us to say that the formula really gives an answer?

What made me think of that: As I started reading your article and saw the formula, I figured, hey, let's see what insight it gives for this year's A.L. choice between Cole and Verlander.

If I did it right, that's pretty razor thin: Cole 99.6 (actually a teeny hair more, but we don't get into that many decimals), Verlander 98.5.

Gutwise, that seems essentially a tie -- no hit neither way.
Would you agree? Do you know if Tom would agree?

And, back to your article: Since you didn't say anything about this thing of what the margin needs to be, does that mean none of these years (except maybe that NL '69 thing) came out so close?​
12:23 AM Nov 6th
 
 
©2024 Be Jolly, Inc. All Rights Reserved.|Powered by Sports Info Solutions|Terms & Conditions|Privacy Policy