Sports-Reference, with the help of researcher Greg Spira, is proud to present our definitive, comprehensive reference page regarding the performance-enhancing drug policies in place for various North American sports leagues/associations. In it, you'll find a list (in FAQ format) of the various PED regulations in pro sports leagues, in addition to lists of every athlete suspended under each league's policies.

We want this page to be your best source for information about steroid/PED policies in pro sports. However, pages like these are constantly growing and evolving with new information, so if you know of a policy or a suspended athlete not listed on the page, send us an e-mail and we'll be sure to include it on the site.

Recently, I was involved in some debate about Rich Harden. On one side, the case was being made that he might/should be a pitcher that teams should look to acquire for 2010. And, the point on my side was that he cannot be counted on to give you a regular turn for a full season - despite how well he may pitch when he does take toe to the rubber.

And, this whole thing got me wondering: Which starting pitchers were the ones that you could least rely on to give you a full season of taking a regular turn?

This is where Baseball-Reference.com's Play Index Pitching Season finder comes in handy. Using it, and setting the filters for "For single seasons, From 1901 to 2009, (requiring GS>=5, IP<=150, GS<=25 and At least 100% games started), sorted by greatest Seasons matching criteria" for each league, I came up with these lists:

American League:

National League:

Granted, this is far from being very scientific. It doesn't really factor in partial seasons that are the result of a pitcher breaking into the major leagues, for example. But, it's a nice, fun, quick way to look at this question.

It's interesting to see names here that you would expect - like Aaron Sele, Jason Bere, Juan Guzman, Sid Fernandez and Randy Wolf.

It's worth noting that Pedro Martinez makes the N.L. list because of his seasons from 2006 through 2009. Clearly, the Red Sox made the right call after the 2004 season by letting him go elsewhere.

In fact, many of the leaders on these lists had these types of seasons come towards the end of their career.

The one thing, above all others, that stands out to me here is that it's such a "modern" event to have pitchers post these types of seasons. You don't see any names on these leader boards of guys who pitched before 1984.

Here's a little gem brought to my attention by Raphy: The Best of Baseball-Reference.com's Sponsorship Comments by Baseball Junk Drawer. Apparently Raphy saw it linked in a piece written by Rob Neyer although I haven't found the original article yet.

Of the four types of hit in baseball, triples are by far the rarest. In 2009, there were 949 triples out of 43,524 total hits (2.18%). I went back and calculated historical values for some past years:

1999: 931/45,327 (2.05%)
1989: 868/36,293 (2.39%)
1979: 1,066/37,911 (2.81%)
1969: 849/32,581 (2.61%)
1959: 591/21,636 (2.73%)
1949: 761/22,168 (3.43%)
1939: 916/23,371 (3.92%)
1929: 1,166/24,642 (4.73%)
1919: 1,048/19,624 (5.34%)
1909: 1,002/19,655 (5.10%)

As you can see, triples used to be more than twice as common as they are today. Although many factors are involved, the single-biggest is the much smaller size of ballparks today as compared to back then. These days, outfielders cover a lot less territory and many balls that could have once gone for hits are either caught or go over the fence for homers. A typical triple requires a certain amount of open space on the field where the ball can get to a position in fair territory that's somewhat far away from all of the fielders. That's a lot rarer these days than it used to be.

Anyway, once upon a time, there were lots of guys who had more triples than doubles in a season.

Here are players from the 40-year period 1901-1940 who had more triples in a season than doubles, ranked by triples:

Notice too that all of these players had more triples than homers.

Now here is the same list compiled from the 40 most recent seasons (1970-2009):

As might be expect, these guys have much lower totals for extra-base hits and played in many fewer games than their counterparts from about 70 years earlier.

Here are the career leaders (1901-present) for most triples with a fewer number of doubles:

As it turns out, #3 Frank Shugart is a bogus entry because most of his career came before 1901 and he finished with 79 career triples and 110 career doubles.

Among active players, the only guys with more triples than doubles are as follows:

Slim pickings there, folks. And before anybody asks, B-R.com lists guys as active until they officially announce their retirement.

Among players with more career triples than homers, the leaders 1901-present include many of the greatest players of the early 20th century:

Even the list of leaders among active players is somewhat impressive:

I'm inclined to guess that Carl Crawford will eventually fall off this list. Generally speaking, he's hit fewer triples in recent seasons while his homers have remained fairly constant. (That's a very rough analysis.) If he loses a little speed in the coming years, which is likely, I'd expect his triples to drop even more. However, I think Juan Pierre is safe

As a final tidbit, there are six players in baseball history to finish their career with more than 1 triple, and also more triples than singles. Limb McKenry leads the way with 7 career hits that included 4 triples. It was done most recently by Gary Moore of the 1970 Dodgers. Moore had 3 career hits: a single and two triples.

Wired magazine recently published a short opinion piece about the suggestion that American professional sports consider changing their playoff system to be more similar to the one used in European soccer.  Over the pond, each soccer league is split into levels. Teams that finish with the worst record (or records) in a given level are demoted to the next lower level, while teams winning lower levels can climb up to a next higher one. I don't know a lot of the specifics of how these leagues work, but that is the basic idea as I understand it.

The Wired article suggests that the guys who have come up with the suggestion feel that this sort of system used in MLB would encourage creation of new teams, take away some leverage from ownership (such as threats of moving the team) and force all teams to be as competitive as possible to avoid the lower levels.

My friend Eric P brought the article to my attention and we discussed how such a baseball playoff system might look in baseball. I'm not suggesting this is a good idea--here's just a summary of what we discussed.

First, we limited our discussion to the current 30 teams and how such a system might perhaps increase competitive balance while also making the full season more meaningful for a larger fraction of the teams.

Suppose we break the 30 teams into 6 levels, each containing 5 teams. Let's go by 2009 records to determine the levels.

LEVEL 1

Yankees, Angels, Red Sox, Dodgers, Phillies

LEVEL 2

Rockies, Cardinals, Giants, Rangers, Marlins

LEVEL 3

Twins, Braves, Tigers, Mariners, Rays

LEVEL 4

Cubs, Brewers, White Sox, Reds, Blue Jays

LEVEL 5

Athletics, Padres, Astros, Diamondbacks, Mets

LEVEL 6

Indians, Royals, Orioles, Pirates, Nationals

So here are some thoughts on how such a league would work in 2010 and forward:

• Teams play a fairly unbalanced schedule against mostly teams within their own level. For example, Level 1 teams play 18 games each against the 4 other teams in their level, 6 games each against the teams in Level 2, and 3 games each against the teams in Levels 3, 4, 5, and 6. That's 72 games against Level 1, 30 games against Level 2, and 60 games against Levels 3 through 6, for a total of 162 games.
• The top four teams from each level make the playoffs. The seeding and home-field advantage are determined by best records. There are six sets of playoffs. The ultimate winner from Level is is the overall champion. The winner from the playoffs in Levels 2 through 6 moves up to the next higher level. The team in each level that did not make the playoffs drops down to the next lower level, with the exception of the worst team in Level 6, which can't drop to a lower level.

Again, I'm not suggesting that this idea is necessarily wonderful, but it does offer some interesting benefits:

• By adding the extra playoff series the lower levels, there are more games and therefore more revenue. MLB will never change anything about the game unless they see more revenue.
• By virtue of the ability to move up or down, many more teams will remain invested in game outcomes for most of the year. Fans will remain interested too, especially those in Level 2 or 3 who hope to move up a level. It might be OK to have the moving team determined strictly by best or worst record (and not add the lower level playoffs) but again that would eliminate the extra game revenue.
• By having the in-level unbalanced schedule, there would be new rivalries that would remain somewhat constant from year-to-year. For example, in 2010 there would be lots of games between the Braves and Twins and the Cubs and White Sox. Those rivalries would continue until one team moved up or down from its current level.

However there are many negatives:

• At the beginning of each year, the overall champion can come from only the 5 Level 1 teams, regardless of how well those teams actually play. It gives those teams an unfair advantage. It also means that no team can ever have a significant reversal, such as Tampa Bay making it all the way to the World Series in 2008 after never having had a winning season.
• The scheduling would be a nightmare and in all likelihood, impossible. The 2010 schedule couldn't have been created until the end of the 2009 season, which isn't enough time. Also, check out what Level 1 would be like in 2010: three east coast teams and two west coast teams. That means that Yankees, Red Sox, and Phillies fans would have to deal with 27 games on the west coast, and these games are against primary rivals. Imagine, too, if there were 4 west coast teams in Level 1 with a single east coast team. That team would be flying west for 36 in-level games (as well as other out-of-level games against west coast opponents.) It would be such a disadvantage for that team. (Of course, the Mariners face such a disadvantage now by having no nearby divisional rival. Even the Althletics are a multi-hour plane flight away. Compare this to the proximity of Boston, New York, Philadelphia, Baltimore, Washington, and Pittsburgh.)
• The teams in the lower levels know that they can't compete for the championship for a minimum of a certain number of years For example it takes a minimum of 5 years to move from Level 6 to Level 1, and only then if the team wins its level every single season. This goes against the spirit of professional sports as we know it in the U.S., where hope springs eternal and any team can have championship aspirations each season. The idea of keeping fans interested would probably backfire for a team like the Royals, which by such a system would have been stuck in the doldrums for many consecutive years.

The true spirit of the Wired article is different from what I suggest above. To follow its suggestion would mean something more like this:

• Create a 12-team "developmental league." Any owner can put together players he or she acquires as free agents and pay a yearly fee to place his team in the league. The league fee would need to be quite large to cover costs associated with travel, locating stadiums to play in, etc.
• Each year, the winner of the developmental league moves up to the major leagues, while the major-league team with the worst record moves down to the developmental league. If creating your own baseball team is exceptionally difficult, then the major league team that falls into the D-league should easily win the D-league championship, and the D-league team playing in MLB should finish with by far the worst record. Then they just switch places the following year. Eventually, though, as things develop, the D-league might get good enough to be able to promote some of its teams to the majors and keep them there.
• Falling out of the MLB and into the D-league would represent a major loss of revenue for that team and therefore should strongly motivate the owners of all teams to try to avoid losing a lot of games.

It seems remarkably unlikely that the playoff system in MLB will ever shift to anything akin to what I've written here, but I still find it fun to think about.

Let's take a look at post-1954 HR leaders in their team's wins and losses. For starters, here's a little lesson about new capabilities of the PI.

For a long time, we could search for career games in which players hit at least one home run. Here are the top 20:

But this doesn't tell the entire story since it doesn't take into account multi-homer games. With the new Batting Game Finder, the total stats in such games are also provided and can then be sorted. Re-running the list above and then sorting for total homers gives a new top 20:

Reggie Jackson drops from 4th down to 7th. If I had to guess why, I'd say it's because he played longer than most and added on more games with single homers.

Now here are the most total homers in team losses:

Hmm, wow, this is interesting. Firstly, Bonds and Sosa pass Aaron for more total homers in losses. I think that can be attributed to 2 general factors: 1) I'm guessing that Aaron's teams had a better overall W-L% than Bonds' or Sosa's teams. Obviously, all other things being equal, playing on better teams makes it harder to hit home runs in losses. 2) In Aaron's day, one guy hitting a home run gave his team a much better chance of winning. With so much more offense these days, a single home run does less to win games for your team and means that guys like Bonds and Sosa tend to have their homers spread more evenly across wins and losses.

A bunch of new names pop up on this top 20 list for homers in losses: Banks, Thomas, Kingman, and Winfield. I suspect that these guys turn up here but not on the top 20 list for team wins because of the same 2 factors I mentioned above.

Every now and then there is game that seems to be headed in no particular direction. One team has the lead , then other, and round and round they go. It's almost as if the team batting last is sure to win. Games with multiple lead changes are usually sloppy, but exciting nonetheless. Lets take a look at a couple of these bizarre games.

PI does not yet have the ability to easily provide us with a list of games with the most lead changes. However, we can still find some crazy see-saw contests by searching for the games with the most blown saves:

Read the rest of this entry »

The recent addition of the arithmetic option to the PI created the possibility of searching for rate stats on a game by game basis. For example, if I wanted to find all of the teams since 1954 that had a  batting average of .500 during the course of a game, I could search for H=.5*AB.  Here are the results of a few of these searches.

Batting Average

Since 1954 there have been 84 teams that have had a batting average of more than .500 in a game.  Another 97 teams have hit exactly .500.  The teams did it in different ways, some with reduced ABs with multiple sacrifices and some with an explosion of hits. 180 of the teams won, one did not.

The highest single game batting average since 1954 belongs to the 2003 Boston Red Sox who went 28-46 against the Florida Marlins on June 27th. Their .609 batting average was the only time since 1954 in which a team hit over .600.

Slugging Percentage:

Since 1954 there have been 67 teams that posted a single game slugging percentage of 1.0 or higher. The only team to lose is the one with the fewest at bats:

The highest single game team slugging percentage since 1954 was posted by the '87 Blue Jays whose 53 total bases with only 44 AB were god for a slugging percentage of 1.205

TOB/PA

While we can't sort by OBP for an individual game, we can use a similar stat of  Times on Base/Plate Appearances. This ratio is similar to OBP, but will reward a team for reaching on an error and penalize it for a sacrifice bunt. Still it can be revealing, nonetheless.
The highest TOB/PA produced by a team in a game since 1954 was by the 2008 Cardinals who combined 9 walks with 26 hits in a single game against Atlanta.

This is another study that was done a while ago using the old version of the PI. It's much easier to do with the new version.

Using the Team Batting Game Finder and my favorite new feature, the arithmetic criterion, I have done several searches setting H = XBH (extra base hits) to achieve the following results.

Anyway, since 1954, a team has had all of its hits go for extra bases a total of 901 times.

These are the 31 times from just 2009. Right away you'll notice that most of these games were losses. That makes sense--if all your hits are for extra-bases, there's a good chance that you didn't get a lot of hits. It's pretty hard to have 7 or more hits all go for extra bases and it's tough to win games with 6 or fewer hits.

In 2009, teams achieving this "feat" went 5-26 (.161). It's easy to find the overall record in such games since 1954 by limiting the searches to just wins or losses.  Since 1954, teams have won 189 such games while losing 712 games (.265 winning percentage.)

The highest number of hits in game where all hits were XBH was 9:

As has been noted on this blog a few times by bloggers and readers alike, game #1 on the above list is highly notable as not only the most hits in a game where all hits were extra-base hits, but also because all nine of the hits were doubles. Notice that in these 11 extraordinary games, the teams went 8-3 (.727) meaning that in games with 5 hits or fewer they went 181-709 (.203.)

Again as has been mentioned elsewhere, since 1954 there have been 4 games where all the hits in the game (by both teams) were extra-base hits. These are really easy to find by doing the same kind of Team Batting Game Finder search but clicking on the bubble near the top to sort by greatest number of teams matching the criteria.

The 3 older games were all 1-0 scores and the game from 2 years ago was a 2-1 game. Not surprisng.

I ran a bunch of other searches looking just at games where all of a team's hits were singles. Since 1954 there have been 19,139 such games. Of those, 3,860 were wins and 15,253 were losses, yielding a win percentage of .201. It's interesting that getting either all singles or all extra-base hits is bad. It's getting a combination of both that is needed. Makes them sound like food groups.

Here are the most hits in games where all of a team's hits were singles. Man, games #2 through #5 were some bad luck! Getting 18 or 19 hits and still losing is rough.

Finally, of those 19,139 times that team has had all its hits go for singles since 1954, more than 2,000 of the occurrences came in the same game when both teams achieved it. In fact, there have been 1,026 games in which that's happened.

Here are the 13 times it happened in 2009:

The teams achieving the feat went 13-13 in these games. (That's a little humor for you...)

According to the PI Pitching Event Finder (using multiple searches) there have been only 2 games since 1954 which featured at least 6 bases loaded walks. One of them involved 6 bases loaded walks in the first 2 innings. The other included an inning that is hard to imagine.

April 22, 1959

Heading into 7th inning in their game against the A's, the White Sox only held a small 8-6 lead. However, they had no reason to worry. In the 7th the White Sox  would score  11 runs on 1 hit. That inning would feature a single, two players reaching on errors (plus an additional error), a base-loaded hit batsman, and 10 walks;  8 of them with the bases loaded. Although their lone single made it into RF, Chicago's three outs never even made it past the pitcher. Here is the play by play: