XBH% ?

[TooLiveBrew]

So what about a stat that measures the rate at which a player drives the ball for extra bases? I know SLG% adresses this, but it also accounts for singles (which are by no means bad).

 

I'll call it XBH%. It's simplistic, which may make it less useful, but IMHO, I don't think so. XBH / Total Hits.

 

Bill Hall's 2006 rate was .538

 

Albert Pujols, in an injury-shortened 2006, put up .467

His 2005 rate, when he played all but one game, was .415

 

I know this stat has a 'Russel Branyan/Jose Hernandez lean', in that a player with a terrible BA could conceivably lead the league in XBH%.

(The Hernandez mention refers to Jose's incredibly high BA when putting the ball in play -- when not K-ing)

 

But I think that, as with any stat, taken with a grain of salt - and used in corellation with other stats, namely BA - XBH% can be very effective.

 

IMHO, it's similar to FIP for pitchers, in that it takes the batter's teammates out of the equation (where RBI & R rely on others, and are sort of 'time-honored' stats --- which I do happed to find useful)

 

I'd love to hear some thoughts on this, or just someone telling me, "Hey, dummy, it alread exists, and it's called _____"

[endaround]

Sure you could use that, but most people already use ISP or Isolated Slugging Percentage. Its just SLG-BA to take away singles. The reason why this has some more value than just XBH% is that it keep the differences between 2Bs, 3Bs and HRs intact.

[TooLiveBrew]

Thanks, endaround!

 

How, precisely, does it keep the differentiation intact? I guess I'm a litte confused about that. I know I'm a newbie to these stats, but I'm trying http://forum.brewerfan.net/images/smilies/smile.gif

[BillScottCanRake]

Hopefully these examples illustrate why you can just subtract SLG and BA to get ISP.

 

If a player goes 4-for-12 with four singles. This gives that player:

 

- BA: .250

- SLG: .250

- ISP: .000

 

To get the players ISP we can use two methods. First, take the totals bases minus singles and divide that by the players ABs:

 

(tb - 1b) / ab = ISP

(4 - 4) / 12 = ISP

0/12= 0

 

The second method (and the most common) is to subtract SLG and BA:

 

slg - ba = ISP

.250 - .250 = 0

 

We can see the ISP is zero in both cases. The reason for this is BA treats all hits the same, while SLG takes into account the number of bases for each hit. When you subtract BA from SLG, you are in effect subtracting one base from each hit, leaving only the number of "extra bases" collected. You have, in effect, removed all of the singles and are only looking at extra base hits (or, more correctly, only the extra bases).

 

Just to make sure this works, you can look at a second example where a player also is 4-for-12, but they have 2 doubles and 2 singles for 6 total bases.

 

- BA: .250

- SLG: .500

- ISP: .250

 

Again, we can calculate ISP two ways. This time, lets use the easier method first and just subtract slugging and BA:

 

slg - ba = ISP

.500 - .250 = .250

 

The second, (total bases - singles) / ab:

 

(tb - 1b) / ab = ISP

(6 - 2) / 12 = ISP

4 / 12 = .250

 

Once again, we get the same result.

 

Hopefully that helps.

[deeswan]

How, precisely, does it keep the differentiation intact?

 

It doesn't give you exact numbers of XBH or anything, but the differentiation remains intact this way.

 

ISO does not correlate in any way to number of extra base hits. It gives something of a percentage of total bases derived from extra base hits. So, a hitter who has 10 XBH, all doubles, will score lower than someone who has 10 XBH, all HRs. By your method the 2 scores would be the same, by the ISO method the HR hitter will score better than the doubles hitter. Simplistic, but hopefully understandable.

 

Again, if you are looking to find the XBH percentage, keep using your method. But ISO is a great method for determining a level of power for an hitter.

[TooLiveBrew]

Wow! Thanks, endaround, BSCR, & deeswan http://forum.brewerfan.net/images/smilies/smile.gif http://forum.brewerfan.net/images/smilies/smile.gif http://forum.brewerfan.net/images/smilies/smile.gif

 

The new guy now has found a much better and objective way to assess offensive performance. I still think there's value in the XBH%, but that's in its breadth (ie. displaying the value of an Overbay or Hall, etc).

 

ISP is an efficient and simple stat. I knew there'd already be something out there! Thanks again guys....

 

[removes toe from 'Statistical Analysis' pool]

[1992casey]

It should be noted that if you look at Baseball Prospectus, they use a slightly different version of the Isolated Power stat. It operates on the assumption that speed, rather than power, is the cause of most triples.

 

To save a trip, I'll quote the entire stat glossary entry.

Isolated Power (ISO) is one of five primary production metrics used in identifying a hitter or pitcher's comparables. PECOTA uses a slightly modified version of Isolated Power that assigns the same value to triples as to doubles (extending a double into a triple is generally an indicator of speed, rather than additional power). Thus, the formula for isolated power as follows: ISO = (2B + 3B + HR*3) / AB

I don't think one method is better than the other. If you want to eliminate the speed factor, use the BP method. If you're simply looking at extra base hits, use the standard formula. Of course, if you want to calculate quickly and easily, use the standard formula. If you're into whipping up spreadsheets, what the heck, put both formulas in.

 

The most important thing to remember is if you're looking at Baseball Prospectus' site, you're going to see a different number than you'll see elsewhere.

[sbrylski]

I don't think I like BP's way. Because the guy is fast they shouldn't take away that base. Does it matter which extra bases came from speed and which from power? As long as he can get those extra bases, that is what the stat is trying to measure.

[1992casey]

They're really two different stats with the same name.

 

If you're a scout type looking for power, use the BP method.

 

If you don't care why extra base hits happen, use the standard method.

 

I see value in each method. But I do wish BP would have come up with a new name for its stat.

[sbrylski]

I see value in each method. But I do wish BP would have come up with a new name for its stat.

 

Yes, I agree.