04.28.06

‘Sunny’ Jim, and Estimated Unassisted Putouts by a First Baseman

Posted in Query at 7:51 pm by Administrator

I recently added the capability of calculating estimated Unassisted Putouts (UPO) by First Baseman to the Team Fielding query.

I was very intrigued by Bill James’ article in ‘Win Shares’ of the same title. I think that his logic is sound with this new stat, and in combination with Assists, I think that UPO+A is probably the most effective measure of a first baseman’s range that anyone has ever come up with. Yes, it is a team stat, but for teams with a regular first baseman, you can easily transfer the responsibility for most of these plays to one man. As James states, the real proof of concept is to see who rises to the top when the statistic is calculated. Go ahead and run it yourself, and see who appears. UPO+A leaders are typically teams with players like Keith Hernandez, Vic Power, Don Mattingly, unquestionably talented glove men. For more on UPO, see the Fielding Glossary, or, more to the point, ‘Win Shares’.


Anyway, in testing out my addition of Unassisted Putouts by 1B to the sortable stats for Team Fielding, and the associated UPO+A, I noticed that there were some teams with VERY low assist totals at 1B. So I ran the query for teams Assists per Game at 1B ASCENDING from 1901 to present. The St. Louis Cardinals of 1924, 1926, and 1928 all show up in the top 15 all-time, at #15, #7, and #10 respectively. All of these teams are around 0.37 assists per game at 1B. Well, there’s got to be a story there.

Off to the Player Fielding query I go, and there he is, the 1B for all of these teams, Jim Bottomley. Hmmm? But wait, there’s more. Appearing at #8 on the list are the 1938 St. Louis Browns. Guess who the 1B on that team was? Yep, none other than ‘Sunny’ Jim Bottomley. Now we really may have something. He switched teams and leagues, with a totally different set of pitchers to play behind, and still put up a truly historically low number of assists per game. What is up with this guy? Well, there is one commonality, all of these teams played in the same home park, Sportsman’s Park. But frankly, I can’t imagine how the dimensions of a ballpark, foul ground for example, could possible affect assists by a first baseman. Any ground ball in foul ground isn’t eligible for an assist, because it’s a foul ball. I could see Putouts being affected, but not Assists.

Overall, Jim Bottomley’s teams occupy #7, 8, 10, 15, 33, 46 on the top 50 list of lowest Team Fielding Assist Per Game at 1B all time. That’s got to be way more than a coincidence. So was this guy just an abysmal fielder with no range whatsoever? Or was he just a ball hog, never giving up the ball to pitcher for the play at first, and instead making all of those putouts himself? Well, we can attempt to answer these questions by looking at the estimated Un-assisted Putout totals for these teams and comparing them to the league average at that time.

Running the Team Fielding query for 1B, UPO Descending, 1922-1937, I’ve got Bottomley teams at #5, #8 and #14 on the top 50 list. As a side note, I need to say that Lou Gehrig’s Yankees teams occupy 5 of the top 9 spots in this query! That is a very strong showing by the Iron Horse, and it makes me want to diverge into another research project, but back to ‘Sunny’ Jim. So his teams do appear on the UPO query, but not overwhelmingly so, at least, not enough to balance out his paltry assist totals, in my opinion.


Well, now I’m off to the Player Career Fielding query page. I want to see how Bottomley’s assist totals truly rank among 1B all-time. The query is Assists per Game ascending, 1B, 1901-2005, 1000 Games Played minimum.  And, oh yeah, he is truly abysmal! His 0.43 Assists per Game is the worst career rate by a 1B ever. Since I am calculated Range+ at the 1B Player level with Assists only (an admittedly flawed approach), he also owns the lowest Range+ of all time at 76. No one else is really even close, except for a man who just hit 1000 games played, Sean Casey. Casey comes in at a stellar 0.46 Assists per game, and a Range+ of 79, both good for 2nd worst all time.


From my results, I would say that Bottomley appears to have been a first baseman with decent range, perhaps even good range, but one who rarely threw the ball to anyone else once he got his hands on it. Maybe he was a Garvey before there was Steve Garvey, although undoubtedly with less range. On the other hand, this is far from definitive, and it would be hard for me to argue with someone who claimed he was a statue over at first. That is also a distinct possibility. If anyone out there has any insights regarding the fielding talents of ‘Sunny’ Jim, I’d love to hear them. It seems that I didn’t answer the question here, sorry about that.


While cursing me for wasting your time reading an article with no real conclusion, check out ‘UPO’ and ‘UPO+A’ in the Team Fielding query, I am convinced that they are the definitive way to measure range at 1B.

 

 

04.14.06

Cannon for an Arm

Posted in Query at 9:56 pm by Administrator

Career Fielding: OF, ARM+ descending, all time, min 250 games

Is it possible that Jesse Barfield didn’t have the best throwing arm of all time? Sure, it’s possible. Is it possible to prove this? Not a chance. Jesse’s ARM+ rating of 202 is simply amazing. This means he gunned down twice as many runners as the average OF in his time! Wow. I believe that Barfield had the best outfield throwing arm ever. Yes, this is a highly subjective conclusion, as it is based on only one stat, assists per game.

There are a lot of factors that come into play when looking at assists per game. The most glaring problem is that baserunners tend to stay glued to the bases more often when a guy with a reputation for gunning out runners is in the outfield. So the strong armed OF has less chances to throw out baserunners than an OF with a poor arm does. Case in point, using the Single Season Batting query, check which OF had the most assists last year. Yep…. Mr. Defense himself, Manny Ramirez, with 17 assists. Why? Because everyone and their brother runs on Manny. For those 17 assists, I’ll bet there were 100+ runners that took extra bases off of ManRam over the course of the season. So you can see the inherent flaw. 

The thing is, everyone knew Barfield had a cannon. It didn’t matter. He still led the league in assists from the OF from 1985 to 1987. If we could somehow measure opportunities for outfield assists, then create an Assists per Chance stat, I believe that Barfield would lap the field. 

Yes, Clemente had an incredible arm as well, and was an amazing player, far superior to Barfield overall. But Barfield had the gun.

  
Note that King Kelly appears above Barfield on the list with an ARM+ of 203. I discount this, for a few reasons. #1 - It was the 19th century. #2 - Look at Kelly’s career Range+, 70! That is truly terrible, people. How could he have been this bad?The answer to this, again, comes from my bible, Bill James’ Abstract. Allegedly, Kelly may have played a VERY shallow OF at times, almost as a fifth infielder. If true, this would create a lot of Assist chances, but in turn take away a lot of outfield Putout chances. I can’t see how this would have helped the team. Think about all of those fly balls landing in the outfield behind him! Oh yeah, one more thing. Kelly owns the worst OF fielding average of all time (>500 G at OF), at 82.0%! This is even worse than his Range+ problem. Even if he did play shallow, what is up with this error rate? There must be some very interesting stories about the way King Kelly played the game. I may have to hunt down a book. :)