Tyler Haltman: Drop Baseball Stats Like Nate Silver

Summary: New page: ...

[[Image:Baseballstats1.jpg|630px|thumb|left|Who says there's no such thing as a sports nerd? Photo by DeusXFlorida/[http://www.flickr.com/photos/8363028@N08/3381046750/ flickr]/CC]]

In the past, the ability to crunch and interpret baseball statistics was limited to spouting off the Triple Crown stats of a hitter?s batting average, home runs, and RBIs, a pitcher?s wins, ERA, and strikeouts. Now, however, we have sabermetrics- a new way to look at baseball statistics. Thanks to luminaries such as people like sabermetrics Godfather Bill James and stats guru Nate Silver, a true armchair statistician must be well versed in a series of increasingly-confusing definitions and formulas if he or she wants to be taken seriously.

Here, then, is a quick cheat sheet of some of the most important baseball statistics in use today, from the normal, to the more advanced, to the truly arcane sabermetric values.

==Hitting Statistics==

The object of a game of baseball is to, on offence, to score a lot of runs, and on defense, to keep the other team from scoring. Every single hitting statistic, therefore, is some measure of how much a hitter helps his team score runs.

===HR (Home Runs)===

The easiest (and most glamorous) way for a hitter to score a run is by hitting a home run. In one swing, he gets to round all of the bases and score. Anyone who was already on base also scores.

A player who hits 25 home runs in a season is considered a good power hitter (which is a player who hits a lot of Home Runs). Guys who hit 40 or even 50 home runs during the course of a 162-game season are among the best in the game.

===RBI (Runs Batted In)===

When a hitter hits the ball and causes one of his teammates to score a run, he is credited with an RBI. It doesn?t matter if he got a hit, a hitter can get an RBI with an out if their out allowed a teammate to score. A hitter who gets 100 RBIs in a season is generally considered to have had a good season.

===AVG (Batting Average)===

Batting Average is the percentage of time that a batter steps to the plate and gets a hit. It is expressed to three decimal points (.323 or .254). A hitter who got a hit every time he came up to bat would have a batting average of 1.000 and people would say he had an average of a thousand (meaning he got a hit every time he came to the plate) ? and that he was probably cheating.

The basic formula is:

'''# of hits / # at bats = Batting Average'''

A hitter who manages to have a batting average of .300 (?He?s batting three hundred.?) is having a successful season ? he?s getting a base hit 30% of the time. A league leader will generally have a batting average of between .340 and .360. Meanwhile, a hitter who is hitting less than .250 or so is having a bad season.

Taken together, these three stats are known as the ?Triple Crown? of hitting. For many, many years, these were the only three statistics that anyone cared about. A player who could hit .300 with 30 HR and 100 RBI was a stand-out player, and that was the end of the discussion.

But then people noticed that these stats did not tell the whole story.

===OBP (On Base Percentage)===

There are a number of ways a player can get on base or make an out that do not count for or against his batting average. They could get a walk, or be hit by the pitch, for example. This means that a player?s batting average is not an accurate indicator of how often a player gets on base. So they created OBP. Now every single at bat is taken into account, and every single result is included.

The formula for OBP is:

'''# times on base / # plate appearances = On Base Percentage'''

Where a .300 AVG was good, a .300 OBP is actually quite poor. Generally, you want a hitter to be able to get on base at about a .340 or .350 clip, and league leaders will have an OBP of .400 or more- that is, 40% or more times he comes to plate, he gets on base safely.

===SLG (Slugging Percentage)===

All hits are not created equal. A hitter who comes up and hits a home run is more valuable than one who comes up and hits a single. Slugging takes this into account, working exactly like a player?s batting average, but counting the total bases achieved per hit rather than just the hit itself. So, for example, a single still counts as one, but a double counts as two, a triple three, and a home run four.

The formula for SLG is:

'''(1 x # of singles) + (2 x # of doubles) + (3 x # of triples) + (4 x # of HR) / # of at bats = SLG'''

A player who hit a HR every at bat would have a SLG of 4.000 and is probably taking steroids. A decent hitter should have a SLG of at least .400 and league leaders will post a SLG of .600 or more.

These two stats, added to AVG, create what is known as a player?s ?slash line.? Someone who understands baseball statistics (and now that includes you) will see a player?s slash line listed as .295/.340/.437 and know that he has a Batting Average of .295, an OBP of .340, and an SLG of .437. This is a solid player who has a little power and is having a decent season.

===OPS (On Base Plus Slugging)===

OPS is the obvious next step after OBP and SLG. It is created by adding OBP and SLG together. A decent hitter will have an OPS of at least .750, while elite hitters will post an OPS of .900 or even 1.000 or more.

In the above example, the player?s OPS is .777. That?s not bad. Not a superstar, but he?s not a drag on his team.

===RC27 (Runs Created Per 27 Outs)===

Now we come to some of the truly advanced metrics. The idea behind RC27 is, how many runs per game (a game is made up of 27 outs) would a team made up of nine copies of a single player score?

At first, this was done relatively simply with the following formula:

'''((Hits + Walks) x Total Bases / (At Bats + Walks) ) / 27 = RC27'''

But that didn?t take stolen bases into account. Or double plays. Or advancing runners. Or sacrifice hits. Or a long list of other individual factors. So now the formula reads:

'''(((2.4 x (At Bats + Walks + Hit By Pitch + Sacrifice Hits + Sacrifice Flies)) + (Hits + Walks ? Caught Stealing + Hit By Pitch ? Ground into Double Plays)) x (3 x (At Bats + Walks + Hit By Pitch + Sacrifice Hits + Sacrifice Flies)) + ((1.125 x Singles) + (1.69 x Doubles) + (3.02 x Triples) + (3.73 x HR) + .29 x (Walks ? Intentional Walks + Hit By Pitch) + .492 x (Sacrifice Hits + Sacrifice Flies + Stolen Bases) ? (.04 x Strikeouts))) / (9 x (At Bats + Walks + Hit By Pitch + Sacrifice Hits + Sacrifice Flies))) - .9 x (At Bats + Walks + Hit By Pitch + Sacrifice Hits + Sacrifice Flies) = RC27'''

There?s another version that includes situational hitting, but if I tried to explain it, your brain would explode. Just know that a decent player has a RC27 of around 5. League leaders are above 8.

===WAR (Wins Above Replacement Player)===

I?m not even going to try to explain how this is created. The basic idea is that a team can get a player in AAA (the highest minor league level) for cheap at any time. He would be a Replacement Player, and is not expected to be very good. They?ve done calculations that baffle NASA Engineers to find out that a team made up entirely of Replacement Players would only win between 20-30 games in a season. Since the record for fewest wins in a season since 1900 (held by the New York Mets in 1962) is 40, a team that wins only 20 or 30 games would be an historically bad team.

WAR measures how much better a given player is than an untalented Replacement Player. Take a look at the WAR scores of your team and you?ll begin to see which players aren?t holding there own, and are therefore worthy of your derision. A good player has a WAR of between three and five. A league leader will have a WAR over seven or eight.

==Pitching Statistics==

Just as every hitting statistic is about understanding how good a given player is at creating runs for his team, every pitching statistic measures some factor that has to do with run prevention.

===W/L (Wins and Losses)===

This is the easiest pitching metric to identify and, some feel, the most useless. At the end of any game, a pitcher on the winning side is given a Win and a pitcher on the losing side is saddled with a Loss. The Loss will go to the pitcher who allowed the base runner who would eventually score the run for the other team that gave that team a lead they never relinquished. He doesn?t have to actually give up the run, just be responsible for the runner who scores. A win goes to the ?Pitcher of Record? for the winning side at the time that run is scored.

Most wins and losses will go to a team?s starting pitcher. But there are rules. A starting pitcher must pitch five or more innings in order to get a win even if he is ahead. If he doesn?t go five innings, then the win will go to one of the pitchers who relieved him.

The magic benchmark for wins has always been 20. These days, very, very few pitchers get a chance to win 20 games in a single season, but a total of 17 or 18 wins is still a great season.

===S (Saves)===

Most relief pitchers won?t get as many wins or losses as a starting pitcher, so people started to look for statistics to quantify their value. Because of course there?s no way to know how good a player is without a statistic.

One of the most popular metrics created is the Save. This is given to a relief pitcher on a winning team who enters with his team ahead (though only ahead by a little) and finishes the game, never giving up that lead.

This is a case where a statistic actually created a role on the team. Before there were saves, there were no Closers. Now, there are pitchers paid an obscene amount of money to come in and pitch the 9th inning of a win in order to rack up a ton of saves.

Generally, every team will have a Closer who will get 20 or 30 saves. The top Closers will end up with between 45 and 50.

===ERA (Earned Run Average)===

This is the basic measure of how effective a pitcher is at keeping runs off the scoreboard, and one statistic in which a low value is better. A pitcher with an ERA of 3.50 is expected to give up an average of 3 1/2 earned runs over the course of an entire 9-inning game.

It?s important to remember, however, that these are EARNED runs, not total runs. A run is deemed ?unearned? if it is scored as a result of a fielding error. In this case, it will not count against a pitcher?s ERA.

The formula to create a pitcher?s ERA is:

'''(Earned Runs Allowed x 9) divided by Innings Pitched = ERA'''

A pitcher with an ERA under 2.00 is having a great season, between 3.50 and 4.50 is average, and anything over 5.00 is considered a bad year.

===QS (Quality Start)===

Sometimes a pitcher will pitch well and still lose, say 1-0. He is saddled with a Loss, even thought he pitched quite well. The other guy just pitched better. So they came up with the Quality Start as a way to measure how many times a starting pitcher pitched well enough for his team to win, whether or not they actually did.

A pitcher who pitches at least six full innings and gives up no more than three earned runs is thought to have done his job and receives a Quality Start. A look at how many Quality Stars a pitcher has will give you an idea of how often he gives his team a legitimate chance to win.

===BABIP (Batting Average on Balls In Play)===

This stat is a measure of how good the pitcher is at making the hitter hit the ball poorly. Taking away walks and strikeouts, this measures what the opposing hitters are doing when they actually make contact with the ball.

The odd thing is, Sabermetricians combed through thousands upon thousands of baseball games and discovered there was almost a universal BABIP constant that was remarkably consistent. That number was .300. In general, the average pitcher will allow hitters to hit .300 against him when they actually make contact with the ball.

When a pitcher?s BABIP is lower than .300, they are generally having a lucky season, and more often than not, their BABIP will revert to the .300 mean the next season and they will not pitch as well. Likewise, if they are having an inexplicably bad season and their BABIP is over .300, there is a very good chance that they are being unlucky, and that they will pitch better next year.

There have been a few pitchers throughout the years who have defied this logic, such as knuckleball pitcher Tim Wakefield, but for the vast majority of pitchers, this rule holds true.

===K/BB (Strikeouts (K) Per Walk (BB))===

A pitcher who strikes a lot of hitters is a good thing, because those are outs that they control, regardless of who their teammates are. Conversely, a pitcher who walks a lot of hitters is giving the opposing team free base runners, and there?s nothing the greatest fielding team can do to stop it.

Naturally, sabermetricians combined the two stats to come up with yet another way of measuring pitching success. Simply put, K/BB gives us an idea of how many outs he gets via the strikeout (meaning his defense doesn?t have to do any work for the out) verses how many extra base runners he?s putting on the bases.

The formula is simply:

'''# of strikeouts / # of walks.'''

A pitcher with a K/BB of 1.00 means that for every batter he strikes out, he also walks one. Generally, you want a pitcher?s K/BB over 2.00, the best can be as high as 5.00, 6.00, or more, meaning for every extra baserunner the pitcher puts on base for free, he?s getting six outs without having to trouble his defense. A pitcher with a K/BB under 1.00 is walking more batters than he is striking out, and he is not long for the league.

===WHIP (Walks and Hits Per Innings Pitched)===

This is a good measure of how well the pitcher is at keeping runners off the bases. Simply add Hits Allowed and Walks Allowed and divide by Innings Pitched.

A decent WHIP would be between 1.25 and 1.40. A pitcher who can get their WHIP under 1.00 is having a great season, and a bad season would be a pitcher whose WHIP approaches 2.00.

===AGS (Average Game Score)===

One of his more celebrated creations of Bill James is the Game Score, which measures how dominating any pitcher is on a particular day.

To figure out a pitcher?s game score:

# Start with 50 points.
# Add 1 point for every out by the pitcher recorded
# Add 2 points for each inning completed by the pitcher AFTER the 4th inning.
# Add 1 point for each strikeout.
# Subtract 2 points for each hit allowed.
# Subtract 1 point for each walk allowed.
# Subtract 4 points for each Earned Run allowed.
# Subtract 2 points for every Unearned Run allowed.

A pitcher has done a decent job if their final Game Score is 50 or higher. A pitcher who manages a Game Score over 80 or even 90 has had a fantastic game.

Average Game Score, then, is simply the average game score that pitcher has had during the season. To figure it out, simply add up all the Game Scores and divide by the games pitched.

''Original article by David Neilsen, Wired.com.''

[[Category:Sports]]
[[Category:Math]]
[[Category:Statistics]]
[[Category:Baseball]]