Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Hall of Merit > Discussion
Hall of Merit
— A Look at Baseball's All-Time Best

Monday, May 15, 2006

Jim Bunning

Eligible in 1977.

John (You Can Call Me Grandma) Murphy Posted: May 15, 2006 at 01:19 PM | 218 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

Page 3 of 3 pages  < 1 2 3
   201. John (You Can Call Me Grandma) Murphy Posted: May 24, 2006 at 05:19 PM (#2035360)
Since folks have chewed the numbers up but good, I will merely volunteer the fact that when I watched Jim Bunning pitch I considered him a very good pitcher. But nothing extraordinary.

I think the majority of the electorate (including myself) would agree with your assessment when it comes to Bunning's "stuff" or peak. What makes him stand out was his durability and a fairly long career. Adding that to his fine pitching makes him a good candidate, though certainly not a "no-brainer."

As Chris points out, Bunning is not unique in our voting patterns, which says to me we have more voters who are inclined to the Shiny New Toy than are inclined to their favourite old Teddy Bear.

I actually downgraded Bunning because my "system" had him higher than I thought was realistic (especially with another close election). Banks could have been a little higher himself on my ballot, but I tried to be a little conservative with him, too.
   202. jimd Posted: May 24, 2006 at 07:09 PM (#2035502)
Still doesn't make him truly special.

Unfortunately, that's what the HOM is really about. Because of the number of electees required for the final total, we must sift through the HOVG to find the most meritorious.
   203. John (You Can Call Me Grandma) Murphy Posted: May 24, 2006 at 07:19 PM (#2035514)
we must sift through the HOVG to find the most meritorious.

Well, one person's HOVG is another's HoMer.

For our purposes, Bunning is considered a borderliner by the electorate.
   204. Paul Wendt Posted: May 24, 2006 at 07:35 PM (#2035543)
"shiny new toy" names the error in a degrading rather than a neutral way.
It suggests that the one who errs is infantile or subhuman because it is mainly babies and pets that get shiny new toys. Or it suggests an adult susceptible to automobile advertisements.

Rob Wood concluded of the shiny new toy:
Since the consequences of errors are asymmetrical, I take the "shiny new toy" phrase as a welcome caution to fully consider the strengths and weaknesses of all
players, most especially new players.


All well said, not only in conclusion.

I think the issue comes down to how a voter slots pitchers in with position
players.


It's true that someone may place Bunning and others all off ballot because it comes down to this. But the table by Chris Cobb might, probably should unsettle "everyone's" ranking of pitchers. I don't believe it all comes down to how much one values 11-year prime.


Chris Cobb's table, truncated:
Table 2 – Durability during Prime Years

Pitcher  Adj
IP  AvgIP Rfact IP+
Bunning  3006.0  2301.1   1.0   1.31
Newsom   3079.1  2400.5   1.0   1.28
Willis   3782.8  3322.4   0.9   1.27
Friend   2898.8  2370.4   1.0   1.22 


Chris,
Adj.IP and IP+ (columns 2 and 5) are calculated from
Avg.IP and Rfact and IP (columns 3 and 4 and not shown).
Right?

What about Rfact?
Suppose no rounding. What empirical data would make Rfact=0.900 precisely right?
In other words, how would you calculate if you were calculating rather than stipulating?
   205. jimd Posted: May 24, 2006 at 07:48 PM (#2035566)
we must sift through the HOVG to find the most meritorious.

Well, one person's HOVG is another's HoMer.

For our purposes, Bunning is considered a borderliner by the electorate.


I think we're really just talking terminology here.

From my point of view, it's obvious that the HOVG picks up where the HOM leaves off. And that when we're done, the bottom N members of the HOM will be pretty close to interchangeable with the same number of top HOVGees. IOW, the borderliners on both sides of the line.
   206. John (You Can Call Me Grandma) Murphy Posted: May 24, 2006 at 08:24 PM (#2035618)
I think we're really just talking terminology here.

We were, Jim. I understood your meaning, but I was just trying to make it clearer for any outsiders wading though this thread.

From my point of view, it's obvious that the HOVG picks up where the HOM leaves off. And that when we're done, the bottom N members of the HOM will be pretty close to interchangeable with the same number of top HOVGees. IOW, the borderliners on both sides of the line.

I absolutely agree.
   207. DanG Posted: May 24, 2006 at 08:35 PM (#2035630)
Doing some quick and dirty calculating. There are now 148 HoMers. Through 2007 we will have 83 more for 231 total. About 50 no-brainers are among these. So I'm guessing about half of the remaining 33 are currently eligible. The top 16 also-rans from 1976 are:

George Sisler
José Méndez
Minnie Minoso
Cannonball Dick Redding
Ralph Kiner
Joe Sewell
Dobie Moore
Jake Beckley
Hugh Duffy
George Van Haltren
Ken Boyer
Cupid Childs
Billy Pierce
Pete Browning
Rube Waddell
Nellie Fox

Conveniently, there is a gap between Fox and the next guy. These 16 seem to be likely HoMers, from our current vantage point.

The next 14 are tightly grouped and comprise the rest of the viable candidates.

Quincy Trouppe
Bucky Walters
Charley Jones
Edd Roush
Mickey Welch
Tommy Leach
Roger Bresnahan
Bob Johnson
Gavy Cravath
Bob Elliott
Burleigh Grimes
Charlie Keller
Alejandro Oms
Larry Doyle

The 30 players listed could be considered as belonging to the HOVG. About half will be HoMers, and half left on the doorstep. The other 45 players receiving votes in 1976 are toast.
   208. Chris Cobb Posted: May 24, 2006 at 08:53 PM (#2035646)
Chris,
Adj.IP and IP+ (columns 2 and 5) are calculated from
Avg.IP and Rfact and IP (columns 3 and 4 and not shown).
Right


Adj. IP is IP * 162g/g in season.
Avg. IP is average IP for full-time starting pitchers in a season, also adj. to 162 g.
IP+ is Adj. IP / (Av.g IP * Rfact)
Rfact is an estimated adjustment arrived at by comparing the top IP+ scores (with no Rfact included) of different eras. Rfact more or less normalizes the top scores.

What about Rfact?
Suppose no rounding. What empirical data would make Rfact=0.900 precisely right?


I don't have a good answer for this question yet, but here are some thoughts.

I believe that an Rfact should only be necessary when the number of pitchers per team counted as full-time starters changes. I'm just entering into the study of the late 1970s, when the 5-starter rotation is beginning to develop. So far, I am sticking to the top 4 IP per team, but I will be watching the usage trends closely as I move forward into the 1980s. Observing what happens as the 5-man rotation becomes standard may provide more insight into the question of calculating Rfact.

To give a fuller explanation of my "method" for calculating Rfact: I have been estimating Rfact by calculating IP+ for the most durable pitchers of each era, to see what the practical maximum for that era is. This is represented by Christy Mathewson and Joe McGinnity (10 years only for the Iron Man) for the 1900s, Walter Johnson and Pete Alexander for the teens, Cy Young and Kid Nichols for the 1890s, and Galvin and Radbourn for the 1880s (though I'm going to be checking Keefe and Clarkson also). The Rfact for each era is the factor necessary to get the top IP+ value in line with those of the 4-man-rotation era, which has Roberts and Spahn as the top scorers so far at 1.44 and 1.38 respectively. The Rfact includes a significant judgment call concerning whether Walter Johnson should have an adjusted IP+ that is equal to Spahn, equal to Roberts, or higher than both (I set him at higher than both, btw). The durability patterns of Spahn and Young, in terms of appearances among league leaders in IP, is very similar, so I set Young's IP+ to Spahn's.

I suspect that rfact ought to be calculable on a less impressionistic basis by considering the effect of contracting the pool of starting pitchers in the 4-man rotation era. What happens, for example to IP+ if the bottom 25% of starting pitchers are removed from the pool that calculates the average IP? Studying those effects might help to provide a sounder empirical basis for this contextual adjustment. I have the data at my disposal to do such a study, but it would take a lot of number-crunching.

Does this seem like a promising approach, or is it unsound in some way?

I should note, btw, that this is essentially an attempt to fairly normalize pre-4-man rotation pitchers to 4-man rotation conditions. Different Rfacts could be developed to normalize to other rotation sizes. It is partly an attempt to provide an alternative perspective on this question to the changing replacement level for PRAR in WARP1, which I find unsatisfying as a solution to the problem.
   209. sunnyday2 Posted: May 24, 2006 at 11:43 PM (#2035875)
DanG, the guesstimate that half of the borderliners' borderliners will come from the current backlog and half from the future backlog is a reasonable guesstimate, though my saying so is only a wild wild guesstimate. Maybe you've given it more thought and have more of a rationale.

I think how this pans out--whether we get about 16 deep on the current backlog or not--will be a test of the "SNT" hypothesis. If it is true that this electorate is susceptible to the allure of the SNT, then maybe we go less than 16 deep on the current backlog.

I think the issue is not really really just SNT's, however. There have been SNT's since 1899. Today, however, there are newly eligible players who are 1) SNT's and 2) players many of us saw play, in many cases as impressionable youths (us, not the players, well, maybe them too).

If we get 16 deep on the current backlog that will be fair evidence that as a whole we have remained objective about those whom we never saw play as compared to those we have.
   210. Joey Numbaz (Scruff) Posted: May 26, 2006 at 05:12 AM (#2037854)
Chris, I'm going to repost this portion of my comment from the 1977 ballot discussion, where I'm working on a new pitcher pennants added system, regarding normalizing IP:

Thanks for the note Chris - I will look at the Bunning discussion for the translated innings discussion.

I would say that my initial position, before reading, is that I don't mind setting the league leaders for innings pitched as the standard for that year. One pitcher in a year can be an outlier (Wilbur Wood in 1971 for example), but the 5 or 10? That's the era norm for top pitchers, which is what we are evaluating.

If we had stats on IP per GS for every league ever, I agree that could work (actually batters faced would be even better).

I would say ideally you could use say drop the #1 guy in the league from the equation - so you don't have a Wilbur Wood throwing the numbers off and changing the average. I think once you get past the top guy or two, it starts to level off.

Looking at 1901-1977 NL and AL, here are the averages of the top 5 in each league for each season, the number in parenthesis is the distance to the next guy up the chain:

NL1: 322.4
NL2: 303.4 (19.0)
NL3: 291.8 (11.6)
NL4: 282.8 (9.0)
NL5: 277.1 (5.7)

AL1: 322.9
AL2: 301.4 (21.5)
AL3: 290.2 (11.2)
AL4: 281.2 (9.0)
AL5: 274.4 (6.8)

After the first guy throws the number off, it follows a pretty steady progression with the difference getting smaller and smaller. I'd say drop the #1 and #2 and set your league norm based on the others.

So maybe drop the first 2 in an 8 team league, first 4 in a 16 team league. # of pitchers to base it on based on league size:

8 team: 3,4,5,6
10 team: (3*.5),4,5,6,7,(8*.5)
12 team: 4,5,6,7,8,9
14 team: (4*.5),5,6,7,8,9,10,(11*.5)
16 team: 5,6,7,8,9,10,11,12

Basically drop the top pitcher for every 4 teams, and take .75 pitcher for every team in the total.

Then normalize your pitchers to that standard, and pick the normalized number (Prospectus' 275) based on the all time average of the answers.

Now I'll read the Bunning thread and completely change my mind.


I'm not sure I agree that 5-man rotations didn't exist before the late 1970s.

Back in the 1920s and 1930s, you would have one or two (or if you were lucky 3) horses, but a total 6 or so guys rotating in and out because of all of the double-headers. Just pulling a random team, the 1930 Yankees, you have Pipgras getting 30 starts, Ruffing 25, Sherid 21, Wells 21, Pennock 19 and Johnson 15. But Pipgras also threw 14 games in relief, Ruffing 9, Sherid 16, Wells 6, Pennock 6 and Johnson 29.

I still think you are better off removing the top 2-4 (depending on the number of teams in the league) and normalizing based on the rest of the league leaders. That's the era norm for what the top pitchers in the league were able to do. I'm open to being convinced otherwise.

Also, it's MUCH easier to calculate something like that, then having to look at the top 4 or 5 starters on every team for every year.
   211. Joey Numbaz (Scruff) Posted: May 26, 2006 at 05:25 AM (#2037857)
By the way, what I described above is normalizing innings to what the #1 pitcher on a middle 1/2 (25-75 percentile) team throws. I think that's a very reasonable standard.

Also, by doing it for leagues as opposed to seasons, you inherently adjust for things like higher scoring leagues, DH (allows starters to throw more innings), etc.. Unfortuntately, it's still easier to throw more innings in a pitcher's park than a hitter's park, but I'm not sure how we'd adjust for that, or if it's worth the trouble.
   212. Dr. Chaleeko Posted: May 26, 2006 at 02:34 PM (#2038042)
Joe and Chris,

I think there's important information to be gained by including the league's top innings guys. Chris is asking himself: What were the practical limits for innings pitched in the pitcher's era/season. Including the top innings is important in establishing what exactly the usage patterns of the era looked like.

I think using the first and fifth place pitcher (in an eight-team league) combined as a barometer of usage is the way to go: doing so shows you how much a better-than-average #1 starter in the league would pitch, while also offering a sense of the ceiling that usage patterns of the day put on a pitcher.
   213. Joey Numbaz (Scruff) Posted: May 26, 2006 at 04:02 PM (#2038169)
But Dr. Chaleeko . . . if you use the #1 guy to set the baseline, you can't reward the guy who breaks all the barriers (like a Wilbur Wood). And you penalize the pitchers who were unfortunate enough to have been active during the time of the true historical outliers (Wood, Chesbro 1904, etc.).

By removing the true outliers from the equation, you can reward extreme performance. I think that's really important.

Over on the 1977 ballot thread I compared the results of doing it my way and WARP's way, if you want to take a look.
   214. Dr. Chaleeko Posted: May 26, 2006 at 04:38 PM (#2038225)
I'll stand down, you've studied it A LOT more closely than I have.
   215. Paul Wendt Posted: May 27, 2006 at 01:14 PM (#2039724)
To give a fuller explanation of my "method" for calculating Rfact: I have been estimating Rfact by calculating IP+ for the most durable pitchers of each era, to see what the practical maximum for that era is. This is represented by Christy Mathewson and Joe McGinnity (10 years only for the Iron Man) for the 1900s, Walter Johnson and Pete Alexander for the teens, Cy Young and Kid Nichols for the 1890s, and Galvin and Radbourn for the 1880s (though I'm going to be checking Keefe and Clarkson also). The Rfact for each era is the factor necessary to get the top IP+ value in line with those of the 4-man-rotation era, which has Roberts and Spahn as the top scorers so far at 1.44 and 1.38 respectively. The Rfact includes a significant judgment call concerning whether Walter Johnson should have an adjusted IP+ that is equal to Spahn, equal to Roberts, or higher than both (I set him at higher than both, btw). The durability patterns of Spahn and Young, in terms of appearances among league leaders in IP, is very similar, so I set Young's IP+ to Spahn's.

I suspect that rfact ought to be calculable on a less impressionistic basis by considering the effect of contracting the pool of starting pitchers in the 4-man rotation era. What happens, for example to IP+ if the bottom 25% of starting pitchers are removed from the pool that calculates the average IP? Studying those effects might help to provide a sounder empirical basis for this contextual adjustment. I have the data at my disposal to do such a study, but it would take a lot of number-crunching.

Does this seem like a promising approach, or is it unsound in some way?


It seems promising but I can't go beyond the superficial seeming. It's challenging to comprehend and bring sound thinking to bear. Now that I see the bottom line, "an alternative perspective on this question to the changing replacement level for PRAR in WARP1, which I find unsatisfying as a solution to the problem."

I expect to be able to do this kind of think :-) with oldies radio on low. This weekend is The History of Rock and Roll; just now, "I get high with a little help from my friends."

I would be inclined to peg number two to number two; to Spahn if he is the 1.38. But I would also be inclined to make it nearly continuous all derived from 5-year moving averages or something like that, without roundoff. Is it now a quantum measure with 0.975 one quantum below 1.000?
   216. Paul Wendt Posted: May 27, 2006 at 01:20 PM (#2039727)
workload analysis - how many forays have we seen here in three years?

Chris Cobb, Are you fussing with multi-team or multi-league pitchers?
JoeD, how about you?
   217. Joey Numbaz (Scruff) Posted: May 27, 2006 at 05:07 PM (#2039791)
Paul, I'm probably just going to skip multi-league guys in terms of calculating the baseline for a season - I don't think it's worth the trouble. It's not going to move the average more than a couple of innings, since I'm using anywhere from 4 to 8 pitchers to calculate the baseline for each league/season.

Multi-team, same-league pitchers are included.
   218. Chris Cobb Posted: May 27, 2006 at 05:36 PM (#2039814)
I haven't been fussing with multi-league or multi-team pitchers. I recognize this as a small source of error, but it isn't worth the trouble at this point to correct. I hope, in a third or fourth iteration of the system, to catch those pitchers and account for them.
Page 3 of 3 pages  < 1 2 3

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Traderdave
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.2043 seconds
49 querie(s) executed