you're reading...
Sidebars, The HoME 100

Timelining: The right tool for certain baseball jobs

In our HoME 100 series, Miller and I had a difference of opinion on how to compare players across eras. Rather than explain it exhaustively, here’s a couple quotes from that series that stake out the differences in our positions.

Miller:

On the other hand, some people like Eric are timeliners. That’s shorthand for saying Ruth didn’t have the competition of others so his numbers shouldn’t be taken on face value. While I disagree with this notion, I completely understand it and can accept it.

Me:

In any objective way of looking across time, we must ask ourselves the simple Jamesian question: Is there anything in the player’s record that may be creating illusions of context? Sometimes the answer is as simple as the ballpark. Sometimes it’s more subtly the run environment. Sometimes it’s something diabolical like pitching usage patterns. But in the case of the timeline, it’s not so specific or simple to see because we live in it.

So to summarize, I think it more reasonable to discount the stats of players the further back in time we go, while Miller believes that we should view time not as a slope but as a horizontal line.

Or to phrase it even another way, there are three schools of thought that I frequently read in online discussions.

  • If we put Babe Ruth in a time machine and brought him to 2016, he’d be lucky to hit .250 with 20 homers. He might be laughed out of the league!
  • If we put baby Babe Ruth in a time machine and brought him to 1986, and raised him the way Mike Trout was raised, he’d be amazing, though it’s unlikely he would dominate the league like he did in the 1920s.
  • Neither of those will ever happen, and all we have is the statistical record, so we should judge every player based on the context they played in.

I tend to fall into the second group, and Miller into the third. We both have good reasons why. So instead of dissecting those arguments, let’s take a different angle on timelining and ask this question: How could timelining be a useful tool for particular kinds of analyses or discussions?

One reason I want to think about this question is simple and personal. I don’t timeline for Halls of Fame/Merit/Miller and Eric. I did and would again for something like the HoME 100. I don’t see any inconsistency here either. The timeline is the right tool for certain kinds of jobs, and rank-order lists across history is one.

Why rank-order lists and not Halls? It’s about granularity. The HoME now has 217 players in it. Once a player gets his plaque, there’s no distinction among the quality of his career or anyone else’s who is in the HoME. In fact, the only distinction the HoME (or most Halls of Fame) make is simply in or out. Great, not great. Is a player one of the best 217 in history or not? But if we instead ask ourselves how the best 217 players in history rank, we must create defensible distinctions between number 1 and 2 or 51 and 52 or 101 and 102. That requires a level of specificity missing from the general question that a Hall usually answers.

The rules of the Hall of Miller and Eric specifically recommend that we observe some degree of balance among positions and eras. We already do a little generalized timelining by electing, say, Sal Bando instead of John McGraw or Tony Phillips instead of Cupid Childs. We know that the 1890s are already slightly overrepresented in our Hall, and we know it’s incredibly unlikely that this era stood out among all others for the quality of its play and the concentration of its talent. So we have to make mental adjustments for the timeline anyway.

But in a ranked list of players, unless the list maker asks for the best players since 1920, we have to consider all of baseball history. And we know that before integration the best players stood out more from the league than they did later when the talent pool widened and deepened. So it’s possible that the stats we have (absolute or derived) may not be telling us anything about how great a player was across time, only in his own time. Stats from 1927 may not be as comparable to those in 2007 as they appear on the surface. That’s why the timeline is a useful tool for making greatest-ever lists. Otherwise we will continually end up with lists with no one from the last 20, 30, or 40 years near the top.

So the timeline brings some much-needed equity to the question of Who’s number one? But it’s not terribly necessary for questions such as Who are the 217 best?

I don’t, yet, have a formal timelining procedure. That’s something I’ve got an idea for. Actually, it’s more like I’m going to steal someone else’s idea and adapt it. It is, however, a pretty good idea, or else I wouldn’t be poaching it. Somewhere in the next year, I’ll pull together an article in which I present that more formal approach, but we’ve got other fish to fry in HoMEland. For now, however, it’s good enough to make some kind of mental adjustment. For example, pretend that Manny Modern to Xavier Expansion to Ian Integration to Andrew Antique all have similar peaks and careers by whatever measure you like. Further pretend the group span 100 years from Manny Modern to Andrew Antique. We can ask ourselves questions such as:

  • Is it reasonable to believe that all four of them played under equally difficult conditions?
  • Do I believe that competition is 10% stiffer for Manny Modern than for Andrew Antique? 20%? 40%?
  • Is the slope of the timeline linear? Humped?
  • How much closer do I think the quality of play was between expansion and integration than between integration and antiquity?

This gets us surprisingly far. Maybe not as precise as we’d like but a long way along. But it not only helps modern players stand out a little more, it also reminds list makers not to unreasonably penalize older players. That is if we’re looking for well-reasoned rankings whether for our own enjoyment, to argue about with a friend, or to post on a website.

Finally, just because I’m timelining to answer a given question also doesn’t mean that I’m sworn to make my decisions based on it and only it. Of course, other factors may rise in importance. The timeline is a resource, not a system. But it’s a good tool for the right job.

Advertisements

Discussion

No comments yet.

Tell us what you think!

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

Institutional History

%d bloggers like this: