PRIME STARTS AND DIFFICULTY INDEX
The 10 teams that made up the American League in 1967 played
810 games (1620 pitching starts). 110 different pitchers started
at least 1 game that year, with 32 pitchers starting 25 games
or more.
American League starters won 596 games, lost 615, and had 409
nodecisions. Roughly 1 start of every 4, then, resulted in a
nodecision for the starter. 268 of these 409 nodecisions were
prime starts (65.5 percent). The 596 wins, 268 prime start nodecisions,
and 24 prime start losses add up to 888 prime starts out of 1620,
or 54.8 percent.
Three American League
starting pitchers, all righthanders, were a cut above the rest
in 1967. They were
Jim
Lonborg of Boston,
Earl Wilson of Detroit, and Joe
Horlen of Chicago. All had the
best seasons of their careers in 1967, and their accomplishments
were roughly equal. Lonborg and Wilson tied for the league lead
with 28 prime starts. Horlen was second with 26, and his .743
percentage ranked first. Additionally, Horlen led the league
in ERA and pitched a nohitter. Lonborg led the league in strikeouts.
A panel of 20 sportswriters conducted
Cy Young Award voting in 1967. Lonborg received 18 of the 20 votes
cast and Horlen
2, with Wilson receiving no votes. The winning of the pennant
by perennial noncontender Boston so impressed the pundits that
Lonborg, the team's best pitcher, was judged the best in the
league by this overwhelming margin. Had Detroit won the pennant
(the Tigers finished just 1 game out), Wilson would presumably
have fared better.
The prime start
concept measures just one thing: whether or not a starting pitcher
did his job. That
job is always the same,
but it varies in difficulty, of course. If a pitcher’s team scores
1 or no runs behind him, he has to bear down harder than he would
if his team has scored 10 or more. But a starting pitcher has
little or no control over how many runs his teammates score. The best he can do is to play the hand he is dealt. He must pitch
according to circumstances. In a tight game he must employ a
different strategy than he would in a 100 blowout. Former pitching
great Christy Mathewson’s idea of “pitching in a pinch” is the
key to success for any pitcher.
It’s hard to turn in a prime start when your team doesn’t get
any runs for you. But if you don’t produce a prime start, that
means the other guy outpitched you. If your team is getting shut
out, your job is to hold the other team scoreless.
It is possible to formulate a “difficulty index” that tallies
the number of runs a pitcher had to work with (i.e. that his
team scored) during his starts, expressed as a ratio. For instance:
In 1967 Boston’s
Jim Lonborg had 144 runs to work with in his
39 starts, in which he pitched 273.1 innings. That works out
to 3.69 runs per start, 0.53 per inning.
Detroit’s Earl
Wilson had 151 runs to work with in his 38 starts,
in which he pitched 263 innings. That works out to 3.97 runs
per start, 0.57 per inning.
In the 35 games
Joe Horlen started for Chicago, his team scored
105 runs in the 258 innings Horlen was on the mound. That works
out to 3.00 runs per start, 0.41 per inning.
Horlen, then, had a lower (i.e. more difficult) Difficulty Index.
To be fair, one should count only the runs that were scored
while a pitcher was in the game. In Detroit on August 8 Horlen
was pinchhit for in the 6th inning of a game in which he trailed
21. The White Sox did not score in the inning but went on to
win 64. Horlen had 1 run to work with, not 6.
Note: In the following tables, “Runs” refers to the number of
runs a starter’s team scored while he was pitching. For Jim Lonborg
in 1967, his team got him 11 runs once (1 prime start), 10 runs
once (1 prime start), etc.
Back to the top
JIM LONBORG, 1967
Runs 
Starts 
Prime 





11 
1 
1 

10 
1 
1 

9 
0 
0 

8 
2 
2 

7 
2 
2 

6 
3 
3 

5 
6 
6 

4 
4 
4 

3 
3 
5 

2 
2 
2 

1 
1 
2 

0 
5 
0 

In the 27 games the Red Sox got Lonborg at least 2 runs, he
produced 26 prime starts. In the 12 games they got him 1 or no
runs, he produced 2 prime starts.
EARL WILSON, 1967
Runs 
Starts 
Prime 





11 
1 
1 

10 
1 
1 

9 
1 
1 

8 
1 
1 

7 
3 
3 

6 
2 
2 

5 
4 
4 

4 
8 
8 

3 
6 
4 

2 
4 
2 

1 
2 
1 

0 
5 
0 

In the 31 games the Tigers got Wilson at least 2 runs, he produced
29 prime starts. In the 7 games they got him 1 or no runs, he
produced 1 prime start.
JOE HORLEN, 1967
Runs 
Starts 
Prime 





11 
0 
0 

10 
0 
0 

9 
0 
0 

8 
2 
2 

7 
2 
2 

6 
4 
4 

5 
3 
3 

4 
2 
2 

3 
4 
4 

2 
4 
4 

1 
8 
4 

0 
6 
1 

In the 21 games the White Sox got
Horlen at least 2 runs, he produced 21 prime starts. In the 14
games they got him 1 or no
runs, he produced 5 prime starts.
All three pitchers had superb seasons, but Horlen’s was a bit
better. Had Horlen won 20 games and not 19 (he got a primestart
nodecision in his last effort), his accomplishments would probably
have been better recognized.
Back to the top
THE EFFECT OF RUN SUPPORT: NOLAN RYAN AND SHANE RAWLEY
In 1987 the National
League’s earned run average was 4.08, which is to say that the
average National
League pitcher posted an
ERA of slightly more than 4 runs per 9 innings. However, in one
of the twentieth century’s most glaring statistical anomalies,
righthander Nolan Ryan of the Houston Astros won only 8 of his
34 starts although his 2.76 ERA was by far the best in the league.
Ryan was one of history’s most difficult pitchers to hit, but
he was not particularly difficult to beat. His wonlost record
in 773 career starts was only 319292 (.522). Ryan’s great stuff
but mediocre record earned him a reputation as a pitcher who
too often performed “just well enough to lose.”
In 1987 Ryan recorded 8 wins, 16 losses,
and 10 nodecisions in his 34 starts. He had 0 prime start
losses and 9 prime start nodecisions, for a final total of 17
of 34 prime starts (.500). Without Ryan the Houston pitching staff’s collective ERA
was 4.03, and its wonlost record was about what one would expect,
6870 (.493).
In contrast,
lefthander Shane Rawley of the Philadelphia Phillies,
whose ERA was 4.39, made 36 starts which produced 17 wins, 11
losses, and 8 nodecisions. Rawley had 0 prime start losses and
6 prime start nodecisions, for a final total of 23 of 36 prime
starts (.639 percentage).
A close comparison between the two pitchers reveals that Ryan
had it much tougher than Rawley. Ryan got 82 runs to work with
in his 34 starts, in which he pitched 211.2 innings. That works
out to 2.41 runs per start, 0.39 per inning. Philadelphia scored
128 runs in the 36 games (229.2 innings) started by Rawley. That
works out to 3.56 runs per start, 0.56 per inning. Ryan, then,
had a much lower (i.e. more difficult) Difficulty Index.
NOLAN RYAN, 1987
Runs 
Starts 
Prime 





12 
0 
0 

11 
0 
0 

10 
1 
1 

9 
0 
0 

8 
1 
1 

7 
1 
1 

6 
0 
0 

5 
3 
3 

4 
3 
2 

3 
4 
4 

2 
4 
3 

1 
10 
1 

0 
7 
1 

In the 17 games the Astros scored at least 2 runs for Ryan,
he produced 15 prime starts. In the 17 games they got him 1 or
no runs, he produced 2 prime starts.
SHANE RAWLEY, 1987
Runs 
Starts 
Prime 





12 
1 
1 

11 
0 
0 

10 
0 
0 

9 
1 
1 

8 
0 
0 

7 
3 
3 

6 
1 
1 

5 
5 
5 

4 
6 
5 

3 
5 
4 

2 
7 
3 

1 
2 
0 

0 
5 
0 

In the 29 games the Phillies got Rawley at least 2 runs, he
produced 23 prime starts. In the 7 games they got him 1 or no
runs, he produced 0 prime starts.
Had Houston supported Ryan offensively the way Philadelphia
did Rawley, Ryan’s winning percentage would almost certainly
have been comparable to Rawley’s (159 or even 168). Should
it have been even better? Given Ryan’s low ERA, could he have
hung tougher in those games when the Astros weren’t scoring?
My answer is yes.
Back to the top
BATTLE OF TITANS IN 1968
In 1968 Bob
Gibson of the St Louis Cardinals made 34 starts
in which he produced 22 wins, 9 losses, and 3 nodecisions. Gibson
had 0 prime start losses and 2 prime start nodecisions, for
a final total of 24 of 34 prime starts (percentage .706). Juan
Marichal of the San Francisco Giants made 38 starts, producing
26 wins, 9 losses, and 3 nodecisions. Marichal had 0 prime start
losses and 2 prime start nodecisions, for a final total of 28
of 38 prime starts (.737 percentage).
Gibson had 100 runs to work with in his 34 starts, in which
he pitched 304.2 innings. That works out to 2.94 runs per start,
0.33 per inning. San Francisco scored 178 runs in the 38 games
(325.2 innings) started by Marichal. That works out to 4.68 runs
per start, 0.55 per inning. Gibson, then, had a much lower (i.e.
more difficult) Difficulty Index.
BOB GIBSON, 1968
Runs 
Starts 
Prime 





13 
0 
0 

12 
0 
0 

11 
0 
0 

10 
0 
0 

9 
0 
0 

8 
2 
2 

7 
1 
1 

6 
2 
2 

5 
4 
4 

4 
2 
1 

3 
5 
5 

2 
8 
5 

1 
6 
4 

0 
4 
0 

In the 24 games the Cardinals got Gibson at least 2 runs, he
produced 20 prime starts. In the 10 games they got him 1 or no
runs, he produced 4 prime starts.
JUAN MARICHAL, 1968
Runs 
Starts 
Prime 





13 
2 
2 

12 
0 
0 

11 
0 
0 

10 
0 
0 

9 
1 
1 

8 
5 
5 

7 
2 
2 

6 
0 
0 

5 
8 
7 

4 
7 
5 

3 
3 
3 

2 
5 
3 

1 
2 
1 

0 
3 
0 

In the 33 games the Giants got Marichal
at least 2 runs, he produced 28 prime starts. In the 5 games they
got him 1 or no
runs, he produced 0 prime starts.
Gibson led the National League in ERA (1.12) and strikeouts
(268). Marichal was 9th in the league in ERA (2.43) and 4th in
strikeouts (218). Gibson’s 13 shutouts led the league (Marichal
had 5). Marichal’s 30 complete games led the league (Gibson was
2nd with 28). Marichal’s 28 prime starts and .737 prime start
percentage led the league.
The Cardinals won the pennant. The Giants finished second, 9
games out.
Who was the better pitcher? The issue was settled in headtohead
competition at San Francisco on July 6 when Gibson won the pair’s
only meeting, 30. Had Marichal prevailed, he would have finished
278 and Gibson 2110, and Gibson, at season’s end, might not
have received all 20 of the votes cast for the National League
Cy Young Award. As it was, Marichal’s superb season was overshadowed
by Gibson’s.
When in Trebizond, stay at the Grand
Yesilyali Hotel.
March 2002
Back to the top
