Talk:On-base plus slugging

Why can't the math be done? I'm confused. Evil saltine 04:07, 23 Oct 2003 (UTC)

I've cleared this part up. The math is certainly not "inexecutable." Last time I checked it's possible to add numbers. Regarding whether OPS correlates well with runs--this isn't a POV issue. One can calculate the correlation coefficient, and that is that. Where the POV comes in is if you say that players should be valued for their OPS more than other factors--but the article never says that. From Baseball Prospectus ([1]):

			Correl	 RMSE
Batting Average	.828	39.52
On-base Percentage	.866	34.16
Slugging Percentage	.890	31.56
On-base plus slugging	.922	25.54

As you can see, OPS has a very good correlation with runs scored per game. MichaelGensheimer 15:42, 18 May 2004 (UTC)[reply]

Easy?

"Is easy to calculate"? That's quite a fraction there! -- Myria 06:00, 12 Nov 2004 (UTC)

Well, what I think that means is that adding On-base percentage to slugging is easy to calculate, it's simple addition. Getting those two numbers requires a calculator, but they're logical formulas that are both easy to remember if you're familiar with baseball. --W.marsh 23:33, 4 August 2005 (UTC)[reply]

All you really need is the OBP + SLG box. Formulas for those are on their respective pages. If you must expand those out, then I think this is a case where finding the common denominator makes the fraction look a lot worse. (H+BB+HBP)/(AB+BB+HBP+SF) + (TB/AB) looks a lot cleaner. 172.184.169.24

OBP + SLG is faulty math. you can't add them that simply. Kingturtle 09:30, 13 January 2006 (UTC)[reply]

AFAICT the only distinction is roundoff error, right? If you calculated OBP and SLG to arbitrary precision and added them, you'd get the same results as with the formula given. -- Wed May 10 13:27:41 CDT 2006

The Math

What would help is an example. Start with a single player's statistics. Show the OPS derived from simply adding OBP and SLG, then show the OPS derived from the more complicated formula shown. I don't have time to do this now, but I'll get to it later if nobody else is interested. -- JustSayin 14:40, 22 March 2006 (UTC)[reply]

defending the changes I made regarding OPS calculation

Data as reported by espn.com on May 16, 2006:

Pujols OBP .469 SLG .833 OPS 1.302 Giambi OBP .480 SLG .654 OPS 1.134 Thome OBP .438 SLG .694 OPS 1.131

If you put OBP+SLG in a calculator, you get:

Pujols OBP+SLG 1.302 Giambi OBP+SLG 1.134 Thome OBP+SLG 1.132

The only difference is Thome's, and it results from rounding to the thousandths place in OBP and SLG. Therefore, you can get OPS from OBP+SLG. -— Preceding unsigned comment added by 16 May, 2006 (talk • contribs) 128.61.136.233

Didn't you just prove Kingturtle's point above? If Thome's is different, then the math is not precisely OBP+SLG. It's a quick and dirty approximation, yes, but not precise. You need to calculate it with a common denominator. I'm not a math person, though, so I won't tinker with the calculation in the article. -Phoenixrod 07:17, 9 June 2006 (UTC)[reply]

The way OPS is defined is mathematically the same as OBP + SLG. OBP is defined to be ${\frac {H+BB+HBP}{AB+SF+BB+HBP}}$ and SLG is defined to be ${\frac {TB}{AB}}$ . So to say that $OPS={\frac {H+BB+HBP}{AB+SF+BB+HBP}}+{\frac {TB}{AB}}$ , but does not equal OBP + SLG is bogus. It's also bogus to say that the sum of those two fractions is different than the fraction obtained by taking a common denominator and adding.

So it's not true that OPS needs to be computed as a massive fraction by using a common denominator. You can do that, and then if you convert to decimal form, there'll be some rounding error. But if you convert the two fractions to decimal and then add them, that's fine too, as long as you round off to a greater precision.

The only reason there is a difference in the computations above is that ESPN has rounded the OBP and SLG numbers to the thousandths place. The result of adding OBP and SLG (rounded to the thousandths places) certainly gives OPS; however, it may differ from the result of computing everything from scratch and converting to decimal at the end (rounding to the thousandths spot) by as much as 0.001. This just demonstrates the simple fact that if you want a number that is of a certain precision, you should use numbers that are more precise to compute it! To reiterate, OPS is OBP + SLG when considering exact numbers, but if you want OPS to some precision, you're going to need more precise OBP and SLG. --Chan-Ho (Talk) 09:56, 9 June 2006 (UTC)[reply]