Further down the Rabbit hole

“How old would you be if you didn’t know how old you are?” – Satchel Paige

My last three articles have examined the 1934-36 Diamond Stars set with the goal of establishing a more refined (e.g., monthly) release schedule for each year’s cards than anything previously documented. Having had success (apart from a “Rabbit hole”) in applying a particular technique to the 24 cards from 1934, I will now apply the same technique to the 60 new cards from 1935 and 12 new cards from 1936.

Specifically, a key to understanding the 1934 release involved associating each player’s age from his bio with the time interval where that age would have been accurate.

The question, then, is whether a similar analysis of the new cards from 1935 and 1936 will yield similar dividends, either in supporting or challenging my original timetables for each of these releases.

Careful readers may have noted the word new a few times already, and my use of it is intentional. The 1935 release included 84 cards, and I’m only looking at 60 of them. Similarly, the 1936 release included 48 cards, and I’m only looking at 12 of them. Why such an intentionally narrow lens?

Reasonably enough, when a player appeared in consecutive years, National Chicle simply bumped his age up by one from one year to the next. For example, the 1934 Al Simmons card presented his age as 31.

On cue, the Al Simmons cards from 1935 (left) and 1936 (right) present his age as 32 and 33 respectively.

This formulaic approach means that card ages for repeated players are simply perfunctory and not influenced by the specific timing of a card’s release beyond year. As such, they would only add clutter to an analysis that will already be messy enough without them.

1935

For the 1935 release this means we should ignore cards 1-24, which were reissued from the prior year, and consider only 25-84. Following earlier hypotheses that the cards were released in series of 12 or multiples thereof, we’ll examine the cards twelve at a time.

Cards 25-36

Here are the first twelve (new) cards in the 1935 release, sorted by when each player would attain his “card age.” As Pepper Martin was born on February 29, I used March 1 for 1935 and noted the date with an asterisk.

As was the case in our 1934 analysis, there are a number of dates that make no sense for a 1935 set of baseball cards. Row, Rice, and Traynor, for example, would have “aged out” the year before the cards came out, while Berger and Rolfe would not hit their card ages until after the World Series. However, this initial table is based on the Baseball-Reference birthdates for each player, which we’ve seen don’t always match the birthdates assumed at the time these cards were produced. A review of more contemporary sources offers three corrections:

Schoolboy Rowe

Rowe’s 1939 Play Ball card suggests 1912 rather than 1910 for his birth year.

Sam Rice

Rice’s 1933 Goudey card suggests 1892 rather than 1890 for his birth year.

Pie Traynor

Traynor’s 1933 Goudey card suggests 1899 rather than 1898 for his birth year.

We can now update our table to reflect these changes.

In my main article on the 1935 release, I conjectured that these cards would have been issued in late April or so. The age data here do not reflect that. While the first eight cards are compatible with a release anywhere from April 6 to May 20, the last four cards list ages pointing to much later in the year (i.e., July 20 or afterward). In fact there is no single window where all 12 ages would be correct.

I have no firm conclusion to draw here and will instead list some possible explanations for these results.

  • Someone goofed, either in their math or their typesetting
  • Austen Lake/National Chicle used reference materials showing different birth years than what I can find
  • Ages on the backs of cards are notoriously unreliable, so what do you expect!
  • The last four cards really were released much later than the first eight

Before moving on to the next dozen cards, I’ll simply note a commonality among the last four cards listed, though I don’t believe it to be significant.

You may recall that the 1936 release included the reissue of 24 cards from near the beginning of the set, twelve that retained their numbering (shown in orange) and twelve that were renumbered 97-108.

In passing I noted earlier that these 24 cards disproportionately consisted of cards requiring team-related or other biographical updates. For what it’s worth, the last four cards from my table were all among these reissues. 🤷

Cards 37-48

The good news here is there are three players with no ages given, hence we have three fewer things that can go wrong! The bad news is we once again have data largely incompatible with the late-April release window speculated earlier.

This time I am able to replace two Baseball-Reference birthdays with more contemporary sources.

Blondy Ryan

Ryan’s 1934 Goudey card suggests 1908 rather than 1906 for his birth year.

Rick Ferrell

Ferrell’s 1933 Goudey card suggests 1906 rather than 1905 for his birth year.

Still, as before, we are left with multiple entries that poorly fit a late April release.

The case of Jo Jo White is most interesting to me. Yes, I suppose one could argue that Hubbell, Dykes, and the four outliers from the previous dozen were part of some postseason release, thereby making their ages correct. However, it’s hard to stretch that theory far enough to encompass White, whose age only becomes correct in June 1936!

Cards 49-60

This next table is our messiest one yet, though I do believe Hank Greenberg’s age of 34 on his card was simply a typo and intended to be 24.

As before, we can update a few birthdates based on contemporary sources.

Oscar Melillo

Melillo’s 1933 George C. Miller card suggests 1902 rather than 1899 for his birth year.

Red Ruffing

Ruffing’s 1933 Goudey card suggests 1904 rather than 1905 for his birth year.

With the Greenberg typo corrected, the data now include only one rogue conflicting with the “mid-June or so” estimate from my earlier article.

I’ll take a very quick detour here that has zero to do with my main effort. Notice two of the names on the list for whom no age was given: John Whitehead and Cy Blanton. Both were rookies in 1935 who got off to very fast starts.

  • Blanton through 10 games: 7-2 with 1.00 ERA, 2 shutouts, 9 complete games, and a save
  • Whitehead through 8 games: 8-0 with 2.86 ERA, 1 shutout, and 7 complete games

You can almost hear the National Chicle execs yelling at the editors: “We don’t have time to find their ages. Just get those damn cards out stat!” Of course you’re now wondering if Glenn Myatt got off to a similarly sensational start. He did not.

Cards 61-72

The initial data for the next twelve cards again has some curveballs.

This time I can only make one “correction.”

Marvin Owen

Owen’s 1938 Goudey card suggests 1908 rather than 1906 for his birth year.

The result is seven cards compatible with my speculative “late July or so” release but four players very definitely in conflict.

I’ll take yet another detour to note that two of the names in yellow had what then would have been considered prodigious rookie seasons with respect to the long ball. (A third name, Wally Berger, was in the 1935 release as well but way back at card 25.)

Most HR by Rookie (pre-1935)

Cards 73-84

We at last come to the final series of the year, one that I’d originally pegged as early September or so. The first player listed clashes considerably with that, but all others seem to match up well.

Bill DeLancey

Happily, DeLancey’s 1936 World Wide Gum card suggests 1912 rather than 1911 for his birth year.

With DeLancey’s information updated, we now have a set of twelve card ages that would have all been correct from September 15 until October 14. Hallelujah!

This concludes our look at the 60 new cards from the 1935 release. As I noted at the top of the article, there may be no compelling conclusions to draw thus far. Across the 60 cards, a full dozen conflict with previously speculated release windows, and one, Jo Jo White, is incompatible with any 1935 release window. I will still offer one full-on conspiracy theory on this “dirty dozen” at the very end of this article, but it’s not one I take seriously.

1936

While the 1936 release included 48 cards in all, only twelve, cards 85-96, represented new players. A full 24 were reissues of previous cards that retained their original card numbers, and twelve others were reissues that adopted new numbering from 97-108.

Cards 85-96

Here are my initial data using Baseball-Reference as my source for dates of birth. Right off the bat, the first 3-4 players appear problematic for a 1936 release.

Fortunately, this is a group of cards that cleans up nicely.

Luke Appling

Appling’s 1937 Goudey card suggests 1911 rather than 1907 for his birth year.

Alvin Crowder

Crowder’s (first) 1933 Goudey card suggests 1901 rather than 1899 for his birth year.

Julius Solters

Moose’s 1938 Goudey card suggests 1908 rather than 1906 for his birth year.

Ray Hayworth

Hayworth’s 1939 Play Ball card suggests 1905 rather than 1904 for his birth year.

The revised table now has no conflicts at all with my previously speculated “early May 1936” release. However, a closer look reveals something else.

The full window when all card ages are correct is much broader, extending from April 2 through October 3. This is more or less the entire baseball season! So yes, the card ages support my supposed release window, but they would equally support just about any release window!

The first question you might ask is whether this outcome was intentional. Was it by design that all players would remain their “card ages” for the entire baseball season? Were the folks at National Chicle suddenly such perfectionists that they couldn’t chance a “card age” being wrong even briefly? Or were the birthdays of the players in question simply coincidence, even if the probability of nine randomly selected players having offseason birthdays is roughly…1 in 500?!

To follow this train of though to its conclusion, we should also look at the 36 reissued cards this same year. Do these cards show evidence of great care with respect to ages or are they largely haphazard? It took a while, but I checked it out.

Cards 73-84

This block is the one I originally theorized as leading off the 1936 release. As we’ve already seen, the window where all ages were correct ran from September 15-October 14. This brings up two possibilities:

  • National Chicle got the ages right and this series was a late-season rather than early-season release.
  • National Chicle didn’t worry about whether these ages were correct.

Cards 2, 4, 5, 8, 9, 10, 12, 16, 22, 26, 30, and 31

This block corresponds to the twelve cards from the early part of the checklist that were reissued with numbering intact. With Al Simmons (#2) “aging out” on May 22 and Jimmie Wilson (#22) not reaching his card age until July 23, there is no single window when all twelve players would have been their card ages. (Throw away Simmons and there is a brief window from July 23 through August 30.)

Cards 97-108 (formerly 11, 15, 19, 25, 27, 28, 29, 33, 34, 35, 36, 46)

This block corresponds to the twelve cards from the early part of the checklist that were reissued with new numbering. With Earl Averill (#100) aging out on May 21 and Red Rolfe (#104) not reaching his card age until October 17, there is again no single window when all twelve players would have been their card ages. (Throw away Averill and there is a brief window from October 17 through November 11.)

I am not at the moment an adherent to the idea that the driving force behind the 1936 release was ensuring correct player ages. However, it’s still at least mathematically interesting to me that Diamond Stars could have batted 46 for 48 by releasing the four series according to this schedule:

  • Cards 85-96: Anytime during season
  • Cards 2, 4, 5, 8, 9, 10, 12, 16, 22, 26, 30, 31: July/August
  • Cards 73-84: September/October
  • Cards 97-108: October/November

One footnote I’ll add to this discussion is that the release schedule above wouldn’t only produce incorrect ages for Simmons and Rolfe but it would also result in an incorrect team (artwork and bio) for Irving Burns (#75), who went from the Browns to the Tigers on April 30. Otherwise, whether by coincidence or design, it holds up remarkably well, too well if you ask me.

RETHINKING THE “DIRTY DOZEN

I promised I’d put forth at least some explanation for the twelve problematic card ages encountered in the 1935 set. While it feels most plausible, sensible, etc., to me to simply deem the card ages wrong, let’s at least consider the possibility that they’re correct (or at least all correct except Jo Jo White) and see where that takes us.

Offhand, the simplest way for these ages to be correct would be if they were not released in the series originally theorized but instead part of a late season sheet all their own. Such an approach would leave gaps in the earlier series, but we know from the 1933 Goudey set and others that leaving gaps could be an intentional tactic to boost sales, i.e., keep kids buying packs in hopes of finishing a run that can’t (yet) be finished.

The main reason I’m not sold on such a theory here is that only a very late release date for the “filler series” would solve the card age issues we’re attempting to solve here. If we include Jo Jo White, we would require a release date of June 1, 1936, or later, which creates more problems than it solves. White notwithstanding, we would still be looking at a release date of November 24, 1935, or later.

I also believe such a scheme would now be detectable on the PSA population report, probably in two ways.

  1. Non-star players among these twelve cards would have similar populations to each other.
  2. The earlier series would likely exhibit evidence of double-prints.

Checking the report, I don’t see either of these occurences.

I’ll also note that the sheet fragment we looked at in my 1935 article does cycle through all card numbers from 61-72 rather than exclude Foxx, Bonura, Medwick, and Trosky.

CONCLUSION

This sort of work isn’t an exact science but rather an arena where some clues point one way, some point another, and some point nowhere. When I began this article, I had some hope that an age analysis would either support or refute earlier assumptions about the 1935 and 1936 release schedules. Instead, I’ll liken the situation to a replay in MLB after a close play is challenged. Under the best of circumstances the review provides clear evidence that the original call was either correct or incorrect. Quite often, however, there is insufficient evidence and the call simply stands while not being confirmed. I think that’s where we are right now with Diamond Stars, at least absent any new angles not yet reviewed.

As always, let me know in the Comments what your own theories might be and especially if you know of information I’ve failed to consider.

Author: jasoncards

I mainly enjoy writing about baseball and baseball cards, but I've also dabbled in the sparsely populated Isaac Newton trading card humor genre. As of January 2019 I'm excited to be part of the SABR Baseball Cards blogging team, and as of May 2019 Co-Chair of the SABR Baseball Cards Research Committee.

3 thoughts on “Further down the Rabbit hole”

  1. I think cross-company circumstances you mentioned and another one specific to National Chicle prove relevant to thinking on card selection and re-release.

    Your previous writing about 1933 Big League Gum set made a persuasive case that Goudey crunched an original multiyear plan into a single year, thanks to a customer frenzy for baseball. Instead of spreading ~240 cards across 2+ years, they released them all (including four Ruth cards) by November 1933. Around that time, a bunch of Goudey staff left to start National Chicle just across the river from Goudey’s Allston HQ.

    National Chicle’s 1934 work included Sky Birds, a set of aviators similar in several ways to Diamond Stars.

    1. Release in multiple series across 2+ years
    2. Re-release of cards from earlier series at later times
    3. 108 total cards each, underpromising larger set sizes (240 for Diamond Stars, 144 for Sky Birds)
    4. Art Deco art style

    This internal Chicle consistency between Diamond Stars and Sky Birds tells me that single baseball cards could prove less significant _on their own_ than the company’s overall approach of 12-to-24 card groups released at varying times. It would help my thinking to know if Chicle distribution reached all major markets at the same time or if eastern cities started receiving cards well before Chicago and St. Louis. If distribution varied, some markets could continue to sell remaining stock of prior seasons before moving ahead to 1935 and 1936.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s