Tuesday, November 28, 2006

Sparking a Debate: Who's the Best Hitter, Part 2

As promised here is a more sabermetric approach to the question at hand. I'm not going to say that this set of data paints a clearer picture, but it does at least try to factor in a number of different variables the first set of data couldn't. I used 7 stats to come up with my rankings, BA diff, OBP diff, SLG diff, RC/27, OPS+, EqA, and Warp-3/season. Before you freak out on me, allow me to explain what each stat means and what each stat factors in.

BA, OBP, and SLG diff - Are the difference between a player's career BA/OBP/SLG compared to the league average BA/OBP/SLG. For example, to figure out Ruth's BA diff I took his career batting average, .342 and subtracted the league average for that time period, .285, and got a difference of .057.
It's hard to compare hitters from different eras; one person may have played during a time where high averages were expected with little power and another player may have been around during a time when the averages were down and pitching dominated. By comparing them to the league average, it's possible to see how far ahead of the pack they were. This is essentially my goal for this entire data set. In the first data chart I compared the players to eachother, but here I'm trying to quantify how much they outperformed players who were playing under similar circumstances and then comparing that to the other all-time greats.

RC/27 - RC on it's own is an estimator created by Bill James (of the Bill James Handbook). RC by itself doesn't help us much since it is a cumulative stat, so those players with longer careers would logically have high RC's, but if you take RC/27 which basically measures how many runs a lineup featuring 9 of the same player would produce per game; you have a good comparison stat.
Imagine 9 Babe Ruth's in one lineup; would you ever have him bunt?

OPS+ - An OPS+ is adjusted to compare the specific season(s) to which it is being applied to. OPS+ is just the OPS of a player measured against the league average for the same time period. Sort of what I did what BA/OBP/SLG diff, but it also accounts for park factors. An OPS+ of 100 is considered to be league average. An OPS+ of 125 means the player was 25% better than a a league average player.

EqA - I'm just going to straight quote Baseball Prospectus here, "A measure of total offensive value per out, with corrections for league offensive level, home park, and team pitching." EqA incorporates baserunning but not defense. EqA is derived from something called Raw EqA, which is calculated by the following formula: (H + TB + 1.5*(BB + HBP + SB) + SH + SF) divided by (AB + BB + HBP + SH + SF + CS + SB)"
It's an eye-sore, I will admit, but it's a damn good tool to compare hitters. EqA is set on a scale like that of batting average, so it's easy to tell what is good and bad. An average EqA is .260.

Warp-3/Season - Again this is straight from Baseball Prospectus "Wins Above Replacement Player, level 1. The number of wins this player contributed, above what a replacement level hitter, fielder, and pitcher would have done, with adjustments only for within the season."
Unfortunately this has a defensive component in it, but it is still a useful tool. I tweaked slightly for our purpose. Warp-3 is a cumulative stat over the span of a career, so I turned it into a players avg Warp-3 over a 162 game season. To get this, I took a players career WARP and multiplied by 162, and then divided by the total number of games he has played.

So enough talk, onto the data chart. Again, some of my thoughts follow the numbers, and if you want a good view just click on the chart and it will expand.













-Ruth is almost across the board, number 1. Considering the fact that he dwarfed most his contemporaries, I'm not surprised.

-Teddy Ballgame, Ruth, Gehrig, and Hornsby all stayed in the top-5. Bonds jumped up into the top-5 this time knocking out Pujols.

-I'm most surprised by Hornsby staying in the top-5. I think Hornsby and Gehrig are now tied for the prestigious, Player that Jeeves' Underrated the Most Award.

-I personally think Bonds' numbers were terribly inflated by his steroid use, but again there's no denying how impressive his stats are.

-A-Rod was the big dropper from data chart 1 to data chart 2, falling ten spots.

-It's disappointing to see Mays so far down the list again. He was always near the top of my list of best players, but I may have to reconsider where I place him now. More on that coming later, when I try and take all these numbers into consideration and change around my personal top-10.

-Thoughts anyone? Still think I'm crazy? Think I'm onto something? Let me hear it.

7 Comments:

Blogger jamesmnordbergjr said...

Poor Yaz, the last guy to win a triple crown can't get out of the bottom spot in either stat pack.

I also wonder how much league size is taken into account, especially considering that lgAVG is being used. (More guys hitting at the Mednoza line would mean a lower league average.) There weren't nearly as many players in the early days, and that might skew the stats a little too. Sorry, Jeeves, I'm not trying to rain on the parade by saying that, I'm just trying to add some more perspective. But you did add the disclaimer about comparing guys from different eras, so I guess that should be taken for what it is.

Thoughts on who is best...

Frank Thomas...very very underrated in historical terms if not here in Chicago.

Albert Pujols...kinda like Terrell Davis in football, not enough time (for me) to be talked about yet. Maybe in time.

Barry Bonds...I can't believe his career BA is that low, but you can't deny his OBP. In a perfect lineup, he'd bat second.

Those late 60's, 70's era hitters just don't stack up. The ptiching, maybe?

11/29/2006 5:09 AM  
Blogger jamesmnordbergjr said...

Oh, yeah...it's nice to see that statistics class paying early dividends for you. :-)

11/29/2006 6:37 PM  
Blogger Jeeves said...

No need to apologize. That's exactly the type of feedback I'm looking for. I want to get the most possible out of all this stuff.
I'm going to try and look into the numbers and see if I can get a feel for lgAVG; we'll see what the Internet has to offer, maybe there's a BA+ stat like OPS which is normalized.

Just glancing at the stats that I have on the chart it almost looks like a cyclical rise and fall. I don't know how much merit that has, I'm just looking at things as they are.
It started in the mid .270's with Anson (1871 (when he started playing)) and then dropped a bit during Wagner's time but went on the upswing during Cobb's and Shoeless' time, and through Ruth's time before peaking during Foxx's prime which was the mid 20's to mid 40's. Average really plummetted for a while, and now may or may not be coming back up. A-Rod's league average is pretty high up there, but Pujols', who is the youngest, is lagging a bit.

We'll see what the wonders of the Internet yeild.

11/30/2006 1:34 AM  
Blogger Jeeves said...

I'll take a look into Mays, at least for the first chart. I don't know a site that can give me breakdowns of his career totals part way through his career, and I can't figure out these sabermetrics for the life of me.

In think that the fact that there were so many good hitters in the 50-60's and not as many teams as there are today does have some effect on league averages and what not, since the talent pool isn't as watered down and there seem to be a large number of studs in that time. I can't quantify the effect, but there has to be some.

Ruth does seem like a freak of nature, just imagine how good he would have been if he conditioned himself like todays athletes!! Good thing IBB's weren't in vogue then, because Ruth would have been walked an insane amount before Gehrig was around to give him protection. I've got more on this coming later, I'm thinking Thursday night, just to wrap things up and address such things as the league not being integrated and all that, but just some other ridiculously dominant athletes that pop into my head...Tiger Woods, Roger Federer, Gretzky, Edwin Moses (won the 400-meter hurdles 120+ consecutive time), Jesse Owens.

11/30/2006 1:52 AM  
Blogger UCLASoxFan said...

I'd throw Lance Armstrong in there too when it comes to absolute dominance. Roy Jones Jr, before he got old.

Yay for the Big Hurt getting some love from the stats!!!:)

Luckily Pujols' career doesn't look like it's about to be cut short like TD's was. (knock on wood)

It must have been the pitching that kept down the 60's and 70's era hitters. When McGwire hit his 65 HR's, he had an OPS+ of 178. In '68 when Yaz had 23 HR's (and a .922 OPS) he had a 171 OPS+. I don't know what was going on there.
Maybe you should take a look at the all time greatest pitchers, Jeeves. See if a bulk of them coincide with the 60's and 70's.

What was that name of the player in the Negro Leagues that hit an insane number of HR's? I wonder what his career stats would look like compared to all these guys.

Maybe BA is cyclical. I can't wwait to see whatcha find out about that.

11/30/2006 3:50 AM  
Blogger UCLASoxFan said...

As far as my top 5 goes I'm gonna have to put in Gehrig now. I always associated him with the streak, didn't realize he was such a damn good player.
Hornsby is looking good to. I might just have to make a 4 way tie for 5th place in my top 5.

11/30/2006 4:00 AM  
Blogger jamesmnordbergjr said...

usf: I think the negro league player you're thinking of is Josh Gibson...I also agree that gehrig would make my top five just from looking at jeeves stats

11/30/2006 6:12 AM  

Post a Comment

<< Home

Free Hit Counter