Is the Craig Biggio the 35th Greatest Player Ever?
by Jeff Zimmerman on Jul 23, 2009 6:00 AM EDT
Email
Print
I have been doing some work on Negro League history and was looking at the all time rankings in the last Bill James's Historical Abstract. I saw Craig Biggio ranked as the 35th greatest player of all time using his formula of peak years, total production and other intangibles. Bill gives an explanation of his reasoning for having Craig so high, but I just can't buy it even though the publication is dated a bit.
The two hitters above and below him on Bill's list were Arky Vaughan, Pete Rose, Eddie Matthews and Carl Yastrzenski. Off the top of my head, I would consider the former 3 much better than Biggio and after doing a little reading Arky Vaughan was a great shortstop that ended up giving up the the second half of his career to the military for WWII. Here is the 5 batters with there Bill James ranking, WAR (from Rally's Database) and Win Share rankings and lifetime totals [Bill James ranking includes pitchers, but the WAR and WS values are for only batters].
| Bill James Top 100 | Name | WAR Rank | WAR | Win Shares Rank | Win Shares |
| 33 | Pete Rose | 43 | 75.3 | 12 | 540 |
| 34 | Eddie Mathews | 21 | 98.3 | 26 | 435 |
| 35 | Craig Biggio | 74 | 66.2 | 25 | 437 |
| 37 | Carl Yastrzemski | 28 | 88.7 | 18 | 490 |
| 39 | Arky Vaughan | 42 | 75.6 | 67 | 357 |
It can be seen that Bill's rankings are based on his win shares (which reward long term consistency, vice above average/replacement production). Looking at the WAR values though, Biggio is not within 9 WAR of any of the other players and there are 52 hitters between him and Mathews. Biggio was a good player, but he is not the same caliber player as these four. The compatibles in the WAR list (Ron Santo,Gary Carter, Rafeal Palmeiro and Jim Thome) seem to me to be a better fit.
Of course, what would an article comparing players at Beyond the Box Score be without a WAR comparison graph.
41 comments
|
0 recs |
Do you like this story?
Comments
WSAB
The biggest difference between Win Shares and WAR is essentially the inclusion of Loss Shares. WSAB takes care of that. As covered in this article…
http://www.hardballtimes.com/main/article/historical-win-shares-above-bench/
…Biggio is 37th all-time in WSAB. This is for players after 1900 — not sure which players are included in your rankings — and doesn’t include pitchers. Eddie Mathews is 19th, Pete Rose is 22nd, Yaz is 27th and Vaughn is 34th.
And (this isn’t meant for just you) may I just say that I find those WAR graphs to be indecipherable? I know you guys like them, but they’re just meaningless spaghetti charts to me. I’m partly color blind, and the colors you guys use to distinguish between players isn’t useful. Plus, line graphs with such similar players, even if the colors were distinguishable, wouldn’t be helpful. IMO.
Gotta admit...
Sometimes I feel the same way. I love the WAR graphs, I like to do them myself, but they are a strange beast to follow. The “contact corner OF” graph was REALLY difficult to read (outside of seeing that Ricky Henderson was ridiculous).
Here's an idea that would help everybody:
Everybody take the same data set and present the most useful implementation of a WAR-trajectory graph. Want the colors bolder? Want the baseline lines less prominent? Want more texture instead of color? Want something else? Present it, and we’ll discuss what’s best. Then maybe we can produce a formatted Excel graph that we can all use.
Thoughts?
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 23, 2009 10:46 AM EDT up reply actions
I just love that BtB has become the place for WAR career graphs.
@bs_uf15bosox9be:OverTheMonster-ALLERGEN WARNING:May contain PB.
skyking162 was supposed to be the place for it, but eh, some things don't work out ; )
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 23, 2009 11:54 PM EDT up reply actions
You've done a great job with BtB though.
Getting mentioned on MLB.com and in major newspapers = epic success of a blog
@bs_uf15bosox9be:OverTheMonster-ALLERGEN WARNING:May contain PB.
The biggest difference is loss shares, sure.
But there are many other significant upgrades from WS to Sean Smith’s implementation of WAR. Is WS passable? I guess, and more so for pre-retrosheet seasons. Should we use it for more than a check to WAR’s results? I don’t see why.
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 23, 2009 10:43 AM EDT up reply actions
they are different systems
Sky, your comment makes the assumption that the systems are comparable. Of course they are to an extent, but they aren’t in other ways. There’s a fundamental difference in philosophy between the two systems. WAR is a “theoretical” metric. It makes total sense within its framework. Win Shares attempts to do something different, however. It attempts to allocate each team’s specific wins (and losses, in James’ new version) to individual players. It can definitely be better but, even if it is perfect, it will yield different results than WAR.
Personally, I think both philosophies have merit, just as WPA has merit. Different ways of looking at the world of baseball. Why throw out any of them?
Um... 285 HBPs. That's why.
The ability to repeatedly get hit by a pitch is the overlooked 7th tool for professional baseball players, after the traditional 5 tools, and the 6th tool, which is ability to stay on the field and off the disabled list. Unless it’s the 8th tool now, having been overtaken by the ability to burn through pitch counts by taking 10 pitches per plate appearance.
Also, I’m a little curious about Biggio’s component WAR numbers from the BaseballProjection database – particularly his GIDP number from 1997, when Biggio famously became the only player ever to go 162 games without grounding into a double play. That effort gets him a “2” – just like most of his other seasons in the 90s. I’m sure there’s a logical explanation for that. But rather than pick apart WARs math, I’ll just point out that it may be a little unfair to certain players who don’t play a single position throughout there career, but instead move around based on their team’s overall need.
Anyway, the real reason Bill James ranks him so high is that Mr. James secretly knows the value of the HBP, but keeps quiet about it to keep from enraging all those people who think that HBPs are equivalent to cheating. You know them as “pitchers”.
HBPs are included both in Win Shares and WAR.
Do you think one or both systems underrates HBPs? Are you defining “greatest” beyond just value to team?
As to the GIDP thing, Biggio led off every game he started (he actually didn’t start in like five games according to B-Ref.) That means his first at-bat was never a GIDP situation and hitting after the pitcher/Ausmus/SS-of-the-day, there’s a good chance he wasn’t in a ton of other GIDP situations. So yes, he avoided GIDPs well, but the low absolute total was also a reflection of low GIDP opportunies (wish we actually had data on opportunities, though – might be worth asking Sean Smith.)
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 23, 2009 12:23 PM EDT up reply actions
James on GIDP opportunities
BJ points out in the Abstract that Biggio “beat the throw to first on a double play attempt” 36 times in 1997. I’m not sure how he calculated this, but Bill seems to think that this is a very high number.
I'm hopelessly biased
I think all systems underrate HBPs, but I’m hopelessly biased on the subject. I haven’t gotten as far as looking at the actual formula WAR uses and how it uses HBPs, so I’m not really qualified to comment on WAR vs WS. And particularly in Biggio’s case where he was easily 5 times better than the average player at that particular skill of being hit by pitches. In general, I’d like the wins above replacement concept better if instead of adding up the results of the individual seasons comparisons, we just took the whole block of seasons in which the player played and compared him against that – so if you’re going to compare Biggio to a “replacement level” player, let’s have a 20 season replacement instead of a series of 20 one-year replacements. But I can rant about that on my own blog.
On the GIDP thing, I might look into it at some point by grabbing all the retrosheet play-by-play files for those 20 years of Astros games, just because I can. I figured it was the number of chances, but it still seems illogical to me to have a perfect season score lower than an imperfect season, regardless of the number of chances to be imperfect, particularly when it’s a situation that the batter has no control of, like the number of people, and their location on base. I’m sure the GIDP piece of WAR works fine as a broad metric, but this is a case where it looks odd to me.
by plunkeveryone on Jul 23, 2009 4:00 PM EDT up reply actions
Question about HBPs:
Wouldn’t you rather have a player draw a walk? Less injury risk, more chances for the batter’s team to see what the pitcher’s got, etc.
@bs_uf15bosox9be:OverTheMonster-ALLERGEN WARNING:May contain PB.
HBPs tend to be a bit more valuale, even in just vanilla linear weights.
Why? Because pitchers are more likely to give up walks when they want to, which means they probably matter less (runner on second vs. nobody on, for example.) But HBPs are more random, so they happen more in bad situations, from the perspective of the pitchers. Minor difference.
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 23, 2009 11:56 PM EDT up reply actions
Per The Book
HBP win value: .029
Unintentional BB win value: .028
However, HBP has a value of 12 runs/win, whereas unintentional BB has a value of 11.6 runs/win. This suggests HBP is more likely to happen in an unimportant game state, no? However, IBBs have a value of 17.8 runs/win, so if you include them then walks are more likely to happen in unimportant situations.
by Tommy Bennett on Jul 24, 2009 12:08 AM EDT up reply actions
Maybe pitchers being less focused leads to more HBPs?
And the intentional ones aren’t usually carried out in high leverage situations, I would think.
IBBs, on the other hand, are generally performed when a much easier out is right behind the walked batter, so the value of it is minimal. Makes sense to me, anyway.
@bs_uf15bosox9be:OverTheMonster-ALLERGEN WARNING:May contain PB.
"Plus, line graphs with such similar players..."
“…even if the colors were distinguishable, wouldn’t be helpful. IMO.”
Isn’t that often the point? if you have three players- one a first ballot HoFer, one an almost certain first ballot HoFer and the third, a guy with virtually no shot with the HoF voters. With their yearly WAR plotted, the charts are so close you can barely tell what belongs to which player, doesn’t that tell the story? Essentially the three of them had the same value on the field, but one in particular is treated very differently by the media due to a skillset they don’t care about.
by erosen on Jul 23, 2009 12:01 PM EDT reply actions 2 recs
Not really
Showing a graph like this to show that you can’t distinguish players from each other is misleading, cause it’s the graph that is making the players indistinguishable, not necessarily the data. That’s one reason to be very careful about graphics. It’s even easier to deceive your readers with graphics than it is with data.
Depends
Here, the resolution is quite good, like about 1/3 or 1/4 WAR per season good. And we always have the total WAR given to us in tabular form as well. In this case, anyway, I don’t see how the graph gives us a false impression of how comparable the players are. Isn’t 1/4 WAR per season resolution enough when comparing entire careers? If the resolution was just 1 WAR, then I would tend to agree with you.
Resolution is only part of the story
If a graph is constructed in such a way that lines overlap continuously, with no distinguishing background, and the lines are hard to differentiate, with muted colors are the only way to differentiate the lines (no difference in width, line design, symbols), etc. etc…. well, you could have awesome resolution and it wouldn’t do an ounce of good.
Here are some simple ideas/rules for line graphs from Elements of Graphic Design (By Kosslyn). I don’t take this stuff literally all the time, but some might find it useful:
- Vary the salience of lines to indicate relative importance
- Ensure that crossing or nearby lines are discriminable
- Put labels at the end of lines
Just some suggestions, beyond resolution, for how to make line graphs discernable.
There might also be too many line and there should be 4 graphs instead of one
One of Biggio against each player with the default reference line(s)
Jeff Zimmerman - Protecting the world from RBI's and Wins from my mom's guest house.
by Jeff Zimmerman on Jul 23, 2009 7:09 PM EDT up reply actions
Questions answered
Charts cluttered
1. One problem is the SBN format where we lose 1/3 to 1/2 of the viewing area due to stuff of the side out my control. I wish they could be wider, but I have to deal with the site format. Let me look into making them vertical.
2. The chart probably shouldn’t have been and they is why the last sentence is not very serious and was a sad attempt at humor.
WAR vs WSAB
1. I just don’t like Win Shares at all and when I belonged to Bill James online, I got into “discussions” with Bill on them and they won’t change (I have heard he is inclined to including negative WS recently). I really don’t like them for several reasons (under valuing pitchers, no positional adjustments, over values playing time, etc) and WAR much more fair. For a while WS were the only way of comparing players from current and past ERAs. With Rally’s WAR dataset, it is now possible to use WAR vice WS (or WSAB).
Jeff Zimmerman - Protecting the world from RBI's and Wins from my mom's guest house.
by Jeff Zimmerman on Jul 23, 2009 12:20 PM EDT reply actions
We could do something like a smaller version before the jump, then a large version after the jump...
Beyond the Boxscore Not a member? Sign up.
by Sky Kalkman on Jul 23, 2009 12:25 PM EDT up reply actions
I still don't think the post jump is still big enough, but not much to do but try something else.
Jeff Zimmerman - Protecting the world from RBI's and Wins from my mom's guest house.
by Jeff Zimmerman on Jul 23, 2009 12:33 PM EDT up reply actions
Link to a full size version?
Also, on the Visual View creator/editor, which used to be WYSIWYG, can’t you click and drag to make the graph larger?
@bs_uf15bosox9be:OverTheMonster-ALLERGEN WARNING:May contain PB.
Its the width of the page. from the FanPost information on one side.
And on the front page is a bunch of other stuff
Jeff Zimmerman - Protecting the world from RBI's and Wins from my mom's guest house.
by Jeff Zimmerman on Jul 23, 2009 5:38 PM EDT up reply actions
But the editor's window is about the same width as even the narrow version of the front page, no?
And the link would still be the best idea.
@bs_uf15bosox9be:OverTheMonster-ALLERGEN WARNING:May contain PB.
You can click on the picture to get a link most of the time.
Derosa.
by vivaelpujols on Jul 24, 2009 3:34 AM EDT up reply actions
WSAB corrects some of that
WSAB includes negative Win Shares, gives more value to starting pitchers and essentially adds losses to compensate for people who get credit just for playing time. I’m not saying it’s perfect, but it’s still pretty good.
Plus, I guess I didn’t state my main point loudly enough, which is that WSAB seems to corroborate your point.
I agree about Win Shares
But two additional things strike me
1. In the methodology chapter of the Historical Abstract, James says he would adjust it if it didn’t “feel right.” I believe he did that with Biggio (can’t check at the moment because my copy is currently in a box right now), which brings me to…
2. Craig Biggio is one of Bill James’ favorite players. We’ve all got “our guys.”
He obviously did adjustvalues with Arky Vaughan
Using WAR, Arky was averaging 3-4 a year and when he came back from WWII, it was near 0.
Bill James did this with several other wartime vets (and other that left us early like Clemente) which I don’t have a problem with. There is just no reason I can really find to have Craig higher; never lead his team to a World Series, didn’t break any new ground (color barrier), never was a coach later.
He just looks totally out place there between Matthews and Yastrzemski.
Jeff Zimmerman - Protecting the world from RBI's and Wins from my mom's guest house.
by Jeff Zimmerman on Jul 23, 2009 2:25 PM EDT up reply actions
I must have missed something...
There is just no reason I can really find to have Craig higher; never lead his team to a World Series, didn’t break any new ground (color barrier), never was a coach later.
Never led his team to a World Series? Wasn’t Biggio with the Astros in the 2005 World Series?
Didn’t break any new ground? 2nd all time hit-by-pitch record, 5th all-time in doubles (most for a RHB), and is the only player to ever have 3000 hits, 600 doubles, 400 stolen bases, and 250 home runs. He has the most homeruns of any leadoff hitter. Not to mention the first player to ever be an All-Star as both a catcher and two-bagger. Ok, so maybe not “new” ground, but certainly rare ground.
Never a coach later? He’s a coach now. Granted, it’s as a high school coach, but he’s only been out of baseball for 2 years, give him some time. The Astros love him, I bet they’ll offer him something before too long.
he deserves somthing like that
great player, 300 hits, played a ton of positions, he is worth the 35th spot
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Beyond the Box Score
FanPosts
Community blog posts and discussion.
Recent FanPosts
Follow us on Twitter!
BTB: @BtBScore
Justin: @justinbopp
Satchel: @SatchelPrice
J-Doug: @RationalPastime
Daniel: @CamdenCrazies
Peter: @CapitolAvenue
Josh: @garik16
Julian: @splashinpumpkin
Matt: @devil_fingers
Bill: @BillPetti
JD: @JDSussman
Lucas: @DBITLefty
Jeff: @jeffwzimmerman
Lewis: @LewsOnFirst
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
