I have been curious about the normalized statistic WIS displays for HR/100AB. WIS labels this HR/100AB#, but I’m going to call it HR#. It seems odd that WIS displays only a single digit for HR#. With most other statistics WIS calculates and displays to greater precision, more digits. I thought it was unusual that WIS appeared to round this off to a whole number.
When I looked more closely I discovered HR# is not rounded at all. For seasons where the actual league average for HR/100 AB is very close to the historic average, normalization will result in only very small changes to a player’s actual HR/100AB. For these years players with actual HR/100AB like 2.1 had HR# of 2. That is what you would expect. But players with actual HR/100AB of 2.8 or 2.9 also had HR# of 2, not 3. This was the case for players across the spectrum of HR# numbers. Clearly HR# is truncated, not rounded. So a player with a HR# of 2 has a normalized HR# between 2.00 and 2.99. The same is true for larger and smaller HR# numbers. So the difference between the calculations done by the WIS game engine, which I assume does not throw away the digits beyond the decimal place, can be significantly different from the ones displayed in search results in the WIS Team Center.
I thought it would be possible to calculate a more precise version of the HR# than displayed by WIS using a variation of the Bill James log5 formula that another WIS owner shared with me. It worked to a degree, but the actual formula must be slightly different. I made some adjustments based on how much above or below the historic average for HR/100AB each season was. Using this adjustment I could create estimates for every player that would fit the whole number truncated values for HR# reported by WIS in the Team Center. I think these estimates are accurate to .1 plus or minus .1.
Here are some of my results for players with the highest WIS HR#. My estimates are listed as EST HR#. AB, HR, and HR/100AB are actual (RL) statistics. In a few cases my estimates round up to one greater than the WIS HR#. Babe Ruth in 1927 was above 13.95 and rounded to 14. WIS lists his HR# as 13. The digits they dropped make him much closer to 14 than 13.
I’d like to know if anyone else has looked closely at this.
|
Player
|
Year
|
B
|
AB
|
HR
|
HR/100AB
|
WIS HR#
|
EST HR#
|
|
Ruth, Babe
|
1920
|
L
|
458
|
54
|
11.8
|
15
|
15.6
|
|
Ruth, Babe
|
1927
|
L
|
540
|
60
|
11.1
|
13
|
14.0
|
|
Williams, Ted
|
1953
|
L
|
91
|
13
|
14.3
|
13
|
13.8
|
|
Ruth, Babe
|
1921
|
L
|
540
|
59
|
10.9
|
13
|
13.5
|
|
Ruth, Babe
|
1928
|
L
|
536
|
54
|
10.1
|
12
|
12.3
|
|
Ruth, Babe
|
1926
|
L
|
495
|
47
|
9.5
|
12
|
12.1
|
|
Bonds, Barry
|
2001
|
L
|
476
|
73
|
15.3
|
11
|
11.6
|
|
Spencer, Shane
|
1998
|
R
|
67
|
10
|
14.9
|
11
|
11.6
|
|
McGwire, Mark
|
1997
|
R
|
174
|
24
|
13.8
|
11
|
11.4
|
|
Ruth, Babe
|
1924
|
L
|
529
|
46
|
8.7
|
11
|
11.3
|
|
McGwire, Mark
|
1998
|
R
|
509
|
70
|
13.8
|
11
|
11.2
|
|
Foxx, Jimmie
|
1932
|
R
|
585
|
58
|
9.9
|
10
|
10.6
|
|
Ruth, Babe
|
1929
|
L
|
499
|
46
|
9.2
|
10
|
10.5
|
|
Medina, Luis
|
1988
|
R
|
51
|
6
|
11.8
|
10
|
10.4
|
|
Ruth, Babe
|
1930
|
L
|
518
|
49
|
9.5
|
10
|
10.3
|
|
Ruth, Babe
|
1922
|
L
|
406
|
35
|
8.6
|
10
|
10.3
|
|
Gehrig, Lou
|
1927
|
L
|
584
|
47
|
8.0
|
10
|
10.2
|
|
Greenberg, Hank
|
1938
|
R
|
556
|
58
|
10.4
|
10
|
10.1
|
|
McGwire, Mark
|
2000
|
R
|
236
|
32
|
13.6
|
10
|
10.1
|
|
Ruth, Babe
|
1931
|
L
|
534
|
46
|
8.6
|
10
|
10.1
|
|
Hill, Glenallen
|
1993
|
R
|
87
|
10
|
11.5
|
10
|
10.0
|
|
Ruth, Babe
|
1923
|
L
|
522
|
41
|
7.9
|
9
|
10.0
|
|
Bond, Walt
|
1962
|
L
|
50
|
6
|
12.0
|
9
|
9.9
|
|
Ruth, Babe
|
1919
|
L
|
432
|
29
|
6.7
|
9
|
9.8
|
|
Ruth, Babe
|
1932
|
L
|
457
|
41
|
9.0
|
9
|
9.6
|
|
McGwire, Mark
|
1995
|
R
|
317
|
39
|
12.3
|
9
|
9.6
|
|
Foxx, Jimmie
|
1933
|
R
|
573
|
48
|
8.4
|
9
|
9.5
|
|
McGwire, Mark
|
1999
|
R
|
521
|
65
|
12.5
|
9
|
9.5
|
|
York, Rudy
|
1937
|
R
|
375
|
35
|
9.3
|
9
|
9.4
|
|
Connors, Merv
|
1938
|
R
|
62
|
6
|
9.7
|
9
|
9.4
|
|
Schmidt, Mike
|
1981
|
R
|
354
|
31
|
8.8
|
9
|
9.4
|
|
Wilson, Hack
|
1930
|
R
|
585
|
56
|
9.6
|
9
|
9.3
|
|
Keller, Charlie
|
1947
|
L
|
151
|
13
|
8.6
|
9
|
9.3
|
|
McGwire, Mark
|
1993
|
R
|
84
|
9
|
10.7
|
9
|
9.3
|
|
Greenberg, Hank
|
1946
|
R
|
523
|
44
|
8.4
|
9
|
9.3
|
|
Aaron, Hank
|
1973
|
R
|
392
|
40
|
10.2
|
9
|
9.2
|
|
Bonds, Barry
|
2004
|
L
|
373
|
45
|
12.1
|
9
|
9.2
|
|
Kiner, Ralph
|
1949
|
R
|
549
|
54
|
9.8
|
9
|
9.2
|
|
Ruth, Babe
|
1935
|
L
|
72
|
6
|
8.3
|
9
|
9.2
|
|
Williams, Cy
|
1923
|
L
|
535
|
41
|
7.7
|
9
|
9.2
|
|
Bonds, Barry
|
2002
|
L
|
403
|
46
|
11.4
|
9
|
9.2
|
|
Gehrig, Lou
|
1934
|
L
|
579
|
49
|
8.5
|
9
|
9.1
|
|
Aaron, Hank
|
1971
|
R
|
495
|
47
|
9.5
|
9
|
9.1
|
|
Bonds, Barry
|
2003
|
L
|
390
|
45
|
11.5
|
9
|
9.1
|
|
Monday, Rick
|
1981
|
L
|
130
|
11
|
8.5
|
9
|
9.0
|
|
Hisle, Larry
|
1980
|
R
|
60
|
6
|
10.0
|
9
|
9.0
|
|
Stargell, Willie
|
1971
|
L
|
511
|
48
|
9.4
|
9
|
9.0
|
|
McGwire, Mark
|
1996
|
R
|
423
|
52
|
12.3
|
8
|
9.0
|
1/2/2011 3:00 AM (edited)