Yeah, yeah, we all read Moneyball. We all get it. You only get so many outs, so don’t waste them. A walk is as good as a hit. As the King would say “et cetera, et cetera, et cetera”.
But really, do we actually get it?
Do we look at OBP the first thing, or are we still stuck in the idea of batting average first? I ask this because even though I like to think of myself as a convert, like to imagine myself as a modern analytical kind of guy, I have to admit to myself that my initial instincts (honed from years and years of reading baseball cards and whatnot) often lend themselves to scanning average before anything else. Tell me you don’t, right? I mean, maybe it’s a generational thing. But for me it’s a pure instinct. My gaze almost always goes to batting average first, and then my brain hauls its ass over to OBP and then OPS.
Anyway… this is a long way to go to state that somewhere along the line I asked myself a purely theoretical question that went like this: how does on-base percentage affect the number of times I can expect my team to bat around? In other words, if my team OBP is whatever it is, how often will my team send every player in the linep to the plate in an inning? It’s just math, right? And this is valuable to think about because I figure if my team bats around, they are guaranteed to score at least three runs even if every guy who gets on base gets on with nothing but a walk. If there are base hits in there, we’ll probably score more.
And yeah, crooked numbers are good.
THE PROCESS:
So I wrote a simple little script that let me set various OBP and a number of test innings, and then ran that number of innings, simulating outs and on base results. For those geeky enough to care, here’s the script:
Code: Select all
#/usr/local/bin/perl -w
$obp=.300;
$innings = 100000;
for($i=0; $i<$innings; $i++) {
$hitters[$i] = 0;
$outs = 0;
while($outs < 3) {
$hitters[$i]++;
if(rand() > $obp) { $outs++; }
}
}
for($i=0; $i<$innings; $i++) {
$pa[$i] = 0;
}
for($i=0; $i<$innings; $i++) {
$ab=$hitters[$i];
$pa[$ab]++;
}
print "Number of batters in an inning if OBP is:" . $obp . "\n" . "Total Innings: " . $innings . "\n\n";
for($i=0; $i<$innings; $i++) {
if($pa[$i] > 0) {
print $i . "\t" . $pa[$i] . "\n";
}
}
print "Close Screen? (Y/N): ";
$quit = <>;
For example:
Code: Select all
Number of batters in an inning if OBP is:0.3
Total Innings: 100000
3 34315
4 30646
5 18555
6 9272
7 4250
8 1795
9 750
10 264
11 92
12 34
13 16
14 3
15 3
16 3
17 1
18 1
Close Screen? (Y/N):
Cool, right?
But I decided I wanted a little more context, so I converted those number to game played. Or, in other words, I asked myself If every player on my team has a OBP of XXX, how many games should I expect it will take before my team bats around?
Let’s look at that example. If my team can be expected to bat around 1.16% of the time, how many innings should it take, on average, before my team sends at least 9 guys to the plate? This is essentially saying that every inning I’m flipping a coin that’s weighted to land on heads only 1.16% of the time…how many times do I have to flip it to expect 1 heads? The answer is 1/X, or in this case I have to play 85.9 innings before I’d expect by random nature to send at least 9 guys to the plate. Divide that by 9 innings in a game, and the answer comes back that, if everyone on my team OBPs at a .300 rate, all other things being equal, my offense should be expected to bat around once every 9.54 games.
Of course, all things are never equal. But you get the point.
THE SUMMARY
All right. If you’ve made it this far, I commend you (or I worry about you, but you’re in this league, so the chances of you being a normal sane person are already low, so maybe I shouldn’t, eh?). Anyway, I ran this script several times, increasing the team OBP from .300 on up to .400, and captured off the data. Then did that simple math to arrive at a nice little chart that shows the base power of OBP.
If my team OBP is .300 I can expect to put up a crooked number about every 9.5 games. At .350 that drops to about 4.4 games. A .400 team would bat around about every 2.1 or 2.2 games--about four and a half more times than that .300 team.
So, yeah, OBP for the win, right?
Of course, my brain is probably always going to have to make my gaze scan across to OBP and SLG and OPS. But, hey, we all have our crosses to bear I suppose.