A Math Project

by SaberR on Aug 12, 2008 2:35 PM PDT Comment 17 comments

So, I have been crunching some numbers on Adrian Beltre, trying to figure out given normalized luck, how his numbers might actually look.

Well I did some math

Adrian Beltre has had 446 AB's this season. In 327 of those AB's, Beltre put the ball in play. Beltre has a 21 % Line Drive percentage, meaning in 68 of his at bats this year he has hit a line drive. According to Hard Ball Times and history a line drive will land for a hit 75 % of the time. So of those 68 at bats Beltre should have 51 or so hits. Now if I did my math correctly, Beltre only has 23 hits this year from his line drives.

So if I go ahead and add 28 hits to Beltre's total (139 instead of 111) That would bump his average up to .311. So Beltre's line might look something like this...

.311 AVE and a .372 OBP

But Here's were I am I know my math has flaws.

1.) Beltre hasn't maintained a 21% LD% all year

2.) I don't know how many of his hits would have been for XB's

3.) I don't know how many of this line drives would still end up being caught due to bad luck

4.) Other factors I have forgotten to put in there.

In short. I need some help.

0 recs | Comment 17 comments

Story-email Email Printer Print

Comments

Basically you're looking at PrOPS

If you want to check the tRA posts and such you can find out the true hit values for each event (LD/GB/FB) and figure out what his projected luck-neutral BABIP should be. Don’t forget to factor out home runs though

The ♥ Jose Lopez ♥ Watch - 146 H, 16 BB, 44 G Left

by seattlebruin on Aug 12, 2008 2:37 PM PDT reply actions 0 recs

Thats the thing. I don't know how to do that.

by SaberR on Aug 12, 2008 2:47 PM PDT up reply actions 0 recs

Heh, I'll try to explain what I did then

(because I know I made a mistake somewhere so I’m hoping someone smart catches it)

Basically, Graham’s tRA post outlines the expected hit value for each type of batted ball
xH LD ~ .71
xH OFB ~ .25
xH GB ~ .28
xH IFB ~ .015

You just need to take the data that presented in FanGraphs on the player page and subtract home runs from total batted balls so you get a representation of what percentages of balls in play only were LDs, IFBs, GBs, OFBs.

From there you can figure out how many of each he’s hit this season and come to a projected BABIP, which is pretty easy to convert to average once you add K’s and HR’s back in.

The ♥ Jose Lopez ♥ Watch - 146 H, 16 BB, 44 G Left

by seattlebruin on Aug 12, 2008 3:05 PM PDT up reply actions 0 recs

That's not quite right

tRA uses xOut values and so includes double plays.

Why not just used PrOPS?

by Graham on Aug 12, 2008 3:27 PM PDT up reply actions 0 recs

I wanted to see if I could re-create it on my own

and double plays make perfect sense since every result I got seemed to be missing ~3% of batted balls. Thx.

The ♥ Jose Lopez ♥ Watch - 146 H, 16 BB, 44 G Left

by seattlebruin on Aug 12, 2008 3:41 PM PDT up reply actions 0 recs

You should've used a ♥ when signing the thx.

Finally, some recognition.

by NOLAmarinergirl on Aug 12, 2008 3:50 PM PDT up reply actions 0 recs

That shows up as a bold 'I'. In FF3 and Safari.

I’m not switching to Camino just to read some hearts. Not that your hearts are bad, I just don’t like Camino.

by dpseadv on Aug 12, 2008 6:20 PM PDT up reply actions 0 recs

It doesn't for me, and I have FF3

Finally, some recognition.

by NOLAmarinergirl on Aug 12, 2008 7:34 PM PDT up reply actions 0 recs

Huh, it's suddenly showing the hearts, in all threads.

My Mac has magical powers, if only I knew how to make them work on command.

by dpseadv on Aug 12, 2008 11:02 PM PDT up reply actions 0 recs

if you use PrOPS, subtract out .030 points -- the formula is a few years old and offense is down

League-average PrOPS is .030 points higher than league-average OPS this year.

my blog // calling BJ Upton lazy is lazy // past results do not guarantee future performance

by Sky Kalkman on Aug 16, 2008 3:38 PM PDT up reply actions 0 recs

Don't just look at his bad luck with line drives

Look at his luck with ground balls and flyballs too. Those should have corresponding percentages for base hits.

by johnbai on Aug 12, 2008 2:45 PM PDT reply actions 0 recs

Also, where did you get his hits on LDs from?

The ♥ Jose Lopez ♥ Watch - 146 H, 16 BB, 44 G Left

by seattlebruin on Aug 12, 2008 3:06 PM PDT reply actions 0 recs

LD% x Hits = number of line drive hits

.21×111 = 23

by SaberR on Aug 12, 2008 3:54 PM PDT up reply actions 0 recs

Doesn't work

Scroll all the way down to Hit Trajectory:

http://www.baseball-reference.com/pi/bsplit.cgi?n1=beltrad01&year=2008

by Jeff Sullivan on Aug 12, 2008 4:47 PM PDT up reply actions 0 recs

What?

Why would LD% be spread evenly amongst all batted ball results? It’s very unlikely, but he may have 68 hits on 68 line drives, and has just gotten insanely unlucky on GBs or FBs. Multiplying LD rate x hits forces the appearance of bad luck.
As an example, look at Ian Kinsler, currently a player who’s quite lucky according to PrOPS. But multiply his .244 LD% by his 157 hits, you get only 38 hits. In other words, he looks UNlucky, which is wrong.

You really need to start with his BABIP on each batted ball type, and I’m not sure you can get that without combing the PBP records. Or, I don’t know, ask Matthew.

by marc w on Aug 12, 2008 4:48 PM PDT up reply actions 0 recs