Posted by Jtpsops on 2/28/2012 2:31:00 PM (view original):
My point is, BABIP is a useless stat - as has been stated numerous times over the last 500 pages.

Lowering your BABIP lowers your WHIP.

If you have next to no control over your BABIP, WHIP becomes relatively useless.

WHIP takes into account K's, HR and BBs. BABIP doesn't.

Ok, then let's measure K's, HR's, and BB's.  We don't need WHIP to do that.

Point is, WHIP takes into account everything. Outs are outs, hits are hits, walks are walks - it doesn't isolate outs and hits "in play". It's an all-encompassing stat. BABIP is not.

But half the numerator in WHIP (the H) is heavily dependent on a pitcher's BABIP.  If BABIP is out of the pitcher's control, then almost half the numerator in WHIP is essentially useless, making WHIP essentially useless.

In other words, if BABIP is useless, we should take it out of WHIP.  If you take BABIP out of WHIP, you are left with HR, BB, and Ks...or FIP.

I don't recall saying it's out of a pitcher's control. My point was, it's meaningless without context. A scorching liner or a dribbling grounder that's fielded for an out shows up as an out, period. A hit, as far as BABIP goes, just shows up as a hit. But what's better - 3 singles or 3 doubles? It's like a hitter with a .340 BA - someone might say "ya, but they're all singles - a guy hitting .280 with all doubles is better". Which would be true. It's inversely true with pitchers. A pitcher who has a .320BABIP but only gives up singles is better than a pitcher with a .270BABIP who gives up all extra base hits.

Those are obviously extreme examples, but BABIP eliminates a lot of context. WHIP doesn't necessarily take that context into account either, but at least it includes everything and gives you an accurate number to determine how many baserunners a pitcher allows. BABIP is arbitrary because it decides "we're not gonna count K's, HR or BBs". That means a pitcher who puts a lot of balls in play with a great defense behind him is gonna look better than a high K guy, and a pitcher who gets ripped for a lot of long balls could actually look better than a pitcher who gives up his share of hits but keeps the run damage to a minimum.

What if I wanted to measure a pitcher's greatness by his OAV second or third time through the batting order? I just randomly decided to cut out the first time through. I doubt you'd accept it.

Some of these numbers are good to have if you're a player and need info on the opposing pitcher, but on their own, they're useless to try and measure how good a player is.

So is BABIP a stat the pitcher can control?

According to BABIP, a HR is not in play.  To a baseball team, it is.  Foul balls into the stands are out of play.  Swings resulting in no contact are not in play.  Any ball hit into fair territory SHOULD be considered in play.  If it happens to go over the fence, not only was it "in play" but devastatingly so.

I am not arguing with you that BABIP doesn't count homeruns as balls in play.  I know it doesnt.  I just dont see how the person who developed the statistic did not realize that any ball that went into fair territory as a result of contact with a bat was "in play". 

BABIP is not a stat that a pitcher can COMPLETELY control.  No.  In fact, it may be the least controllable stat for a pitcher.  They only have control of how much velocity is on a ball put in play that doesn't leave the park.  After that....   if a duck snort finds a hole, or a rocket goes right at a fielder who was out of position at the time of the catch, it is all up to the stars.  Thereby making it one of the least effective stats at telling us anything.

So if the stat isn't controllable and tells us very little, why would we want to use a stat (WHIP) that relies heavily on BABIP?

LETS PLAY DISTRACT THE STAT NERDS!

Posted by jrd_x on 2/28/2012 4:46:00 PM (view original):
So if the stat isn't controllable and tells us very little, why would we want to use a stat (WHIP) that relies heavily on BABIP?

You shouldn't rely solely on any one statistic.  Nor even a huge pile of them.  Liars.  Damned Liars.  Those who use statistics.

However, using a complete set of data and statistics in conjunction with having watched the players and having a decent understanding of the game allows for a much better conclusion than any one aspect on it's own.

That being said....  WHIP is a decent START to a pitcher's effectiveness.  It's obvious to me (at least I think it is) that Hunter had a higher ERA and a lower ERA+ even though he allowed fewer baserunners than did Carlton because he stranded fewer runners with whiffs than did Carlton, and he allowed homers at a much higher rate than did Carlton.

Posted by antoncresten on 2/28/2012 4:47:00 PM (view original):
LETS PLAY DISTRACT THE STAT NERDS!

And thanks for this.  Great distraction.

Posted by jrd_x on 2/28/2012 9:24:00 AM (view original):
You didn't answer my question. Does "based on" mean you changed them or the years aren't consecutive?

Here's another hypothitical for you. Based on real numbers by two different pitchers.

Player A: 3621 IP, 3.42 ERA, 100 ERA+, 4.48 FIP, 1.191 WHIP, .243 OAV, .299 OBP, .390 SLG, .689 OPS
Player B: 3659 IP, 3.23 ERA, 104 ERA+, 3.91 FIP, 1.287 WHIP, .249 OAV, .317 OBP, .410 SLG, .728 OPS

The last four items (OAV, OBP, SLG, OPS) were for hitters against that pitcher.

Who's the better pitcher?

It's a hypothetical based on real stats by two real pitchers.

For the sake of this example, full careers. Including ramp-up at the beginning, decline at the end.

What does "based on" mean?  Did you change them?  Are they consecutive seasons?