Sean Lahman database sample queries - mysql

Question

Is there any place where can I find sample queries (SELECT, UPDATE, DELETE) for Sean Lahman database? I wanna see what can be done with this database..

Answer 1

I use to ship the database with sample queries. Maybe I should revisit that idea. Here are a few to get you started.
A simple one to show all of the players named "Sean:"
SELECT nameLast, nameFirst, debut
FROM Master
WHERE (nameFirst="Sean")
ORDER BY nameLast;
Here's one to show a list of players with 50 HRs in a season:
SELECT Master.nameLast, Master.nameFirst, Batting.HR, Batting.yearID
FROM Batting INNER JOIN Master ON Batting.playerID = Master.playerID
WHERE (((Batting.HR)>=50))
ORDER BY Batting.HR DESC;
Here's one to show the all-time leaders in strikeouts:
SELECT Master.nameLast, Master.nameFirst, Sum(Pitching.SO) AS SumOfSO
FROM Pitching INNER JOIN Master ON Pitching.playerID = Master.playerID
GROUP BY Pitching.playerID, Master.nameLast, Master.nameFirst
ORDER BY Master.nameLast;
There are several websites with tutorials on using the database that include sample queries. See:
http://webdev.cas.msu.edu/cas992/weeks/week5.html
http://www.hardballtimes.com/main/article/databases-for-sabermetricians-part-one/
You can find more by googling 'sql lahman'

Answer 2

This database can do most anything except game-by-game analysis, for that you will need to go to Retrosheet {http://www.retrosheet.org/game.htm}.
But let's say that you want to replicate the totals you see on Baseball-Reference.com, you could easily do that.
If you want some advanced metrics (Sabermetric-like stats), I recommend Tom Tango's website. There you can find help to do your own queries for wOBA. You can also formulate (try to duplicate) FanGraph's or Baseball-Reference's WAR.
Basically, anything you want (provided you can do the calculations/master the SQL syntax) from this except game-by-game or pitch-by-pitch types of data.
Here's a query to determine Salary by games played (offensively and defensively) to figure out how much a player cost/makes per game. (T-SQL w/ SQL Server 2012 Express)
select
m.namefirst,
m.namelast,
s.yearID,
s.teamID,
s.salary,
Cast ('162' as Int) as FullSeason,
round(sum(s.salary)*1.00/162,0) as Game_Rate,
sum (case when s.playerID=b.playerID then f.g else 0 end) as Gm_App_Field,
b.g as Batting,
--sum(case when s.playerID=b.playerID and s.yearID=b.yearID then b.g else 0 end) as Gm_App_Hit,
sum (case when s.playerID=b.playerID then f.innouts else 0 end) as InnOuts,
sum(F.InnOuts)/27 as FullGames,
round((sum (case when s.playerID=b.playerID then f.g else 0 end)/162.0)*s.salary,0) as PayByGmFielding,
round(sum(b.g*s.salary)/162,0) as PayByGmHitting,
round((sum(F.InnOuts)/27)*(s.salary/162),0) as PlayingSalary
from Fielding f
inner join batting b
on f.playerID=b.playerID and f.yearID=b.yearID
inner join salaries s
on f.playerID=s.playerID and f.yearID=s.yearID
inner join [master] m
on b.playerID=m.playerID and f.playerID=m.playerID and s.playerID=m.playerID
where
f.yearID = '2013' and f.POS <> 'P' --b.playerID = 'zimmejo02'
group by
m.namefirst,m.namelast, s.yearID , s.teamID, s.salary, b.g
Which outputs this:
namefirst namelast yearID teamID salary FullSeason Game_Rate Gm_App_Field Batting InnOuts FullGames PayByGmFielding PayByGmHitting PlayingSalary
A.J. Pollock 2013 ARI 491000 162 9093 119 137 2897 107 360672 1245685 324302
You can also create your own searches, here's one with players who have more BB/SO with full player cards, including the WAR I came up with (may be off a little from FanGraphs or Baseball Reference) - (T-SQL w/ SQL Server 2012 Express)
--1. Retrives Full Player Records of guys with more BB than SO
select
m.namefirst,
m.namelast,
b.yearID,
b.yearID-m.birthyear as Age,
b.G,b.AB,b.R,b.H,b.[2B],b.[3B],b.HR,b.RBI,b.SB,b.BB,b.SO, left(round((b.bb*1.000/b.SO),3),4) [BB/SO Rate], left(round((b.h*1.000/b.ab),3),5) as Average
,b.IBB,b.HBP,b.SH,b.SF,b.SF,b.GIDP,case when br.yearID=b.yearID and br.playerID=b.playerID then br.War else 'error' end as WAR
from [master] m
inner join batting b on b.playerID=m.playerID
inner join BR_WAR_2013 br on br.playerID=m.playerID
where b.SO <> 0 and b.AB > 300 and b.bb>b.SO
group by
m.namefirst,
m.namelast,
b.yearID,
b.yearID-m.birthyear,
b.G,b.AB,b.R,b.H,b.[2B],b.[3B],b.HR,b.RBI,b.SB,b.BB,b.SO, left(round((b.bb*1.000/b.SO),3),4), left(round((b.h*1.000/b.ab),3),5)
,b.IBB,b.HBP,b.SH,b.SF,b.SF,b.GIDP,case when br.yearID=b.yearID and br.playerID=b.playerID then br.War else 'error' end
having case when br.yearID=b.yearID and br.playerID=b.playerID then br.War else 'error' end <> 'error'
order by b.yearID desc, left(round((b.bb*1.000/b.SO),3),4) desc